Healthcare analytics, AI solutions for biological big data, providing an AI platform for the biotech, life sciences, medical and pharmaceutical industries, as well as for related technological approaches, i.e., curation and text analysis with machine learning and other activities related to AI applications to these industries.
Vincent Motto-Ros of Lyon 1 University, in Lyon, France, is combining the ability of atomic spectroscopy techniques to detect and quantify metals with the mapping approaches most often used with molecular techniques. He has combined laser-induced breakdown spectroscopy (LIBS) with electron microscopy to map the metals and metallic nanoparticles in biological tissue, as a way of studying the update and clearance of these materials by biological systems. In this interview, he discusses his work applying LIBS to biological analysis, including the methods, advantages, and future directions.
In recent years, many researchers have been advancing the application of molecular spectroscopy techniques, such as Raman and infrared (IR) spectroscopy, to image biological tissues, such as for the analysis of tumors. Atomic spectroscopy techniques such as inductively coupled plasma-mass spectrometry (ICP-MS), on the other hand, are being used extensively to study metals and metallic nanoparticles in biological matrices. Vincent Motto-Ros of the Light and Matter Institute at Lyon 1 University, in Lyon, France, is combining the ability of atomic techniques to detect and quantify metals with the scanning approaches most often used with molecular techniques. He has used laser-induced breakdown spectroscopy (LIBS) to image metals and metallic nanoparticles in biological tissue, as a way of studying the uptake and clearance of these materials by biological systems. He recently spoke to us about this work.
You are developing LIBS techniques for the imaging analysis of metallic nanoparticles and other metals in biological tissue. How does LIBS imaging work?
In LIBS imaging, laser-induced plasma are generated continuously while scanning the sample surface over the region of interest. Elemental images are obtained, in a pixel-by-pixel manner, after extracting the intensity of the interesting species (atoms, ions, or molecules) from each recorded spectrum (Figure 1).
Figure 1: (a) Principle of LIBS imaging instrument with the major components, including the microscope objective used to focus the UV laser pulse, the motorized platform supporting the sample, and the optical detection system fiber-connected to the spectrometer. (b) Example of single-shot emission spectra in the 315–345 nm spectral range recorded in three different regions of the mouse kidney sample indicated in (a) with emission lines of calcium (Ca), sodium (Na), and gadolinium (Gd). (c) Relative-abundance images of Gd (green), Ca (yellow), and Na (red) represented in a false color scale. (d) Multielemental biodistributions in a coronal murine kidney section, 24 h after Gd-based nanoparticle administration (spatial resolution of 10 µm).
In a recent study in mice, you used LIBS to investigate the renal clearance of gadolinium-based nanoparticles that are being studied for their potential as both diagnostic tools and therapeutics (1). What exactly were you looking for?
Nanomaterials represent a huge potential for future medical applications, but the characterization of these attractive candidates within biological tissue remains very challenging, especially because of their small size. Any modification applied to such small nanoparticles, such as fluorescence labeling, may indeed modify their shape, size, or charge, and therefore affect their biodistribution. In this work, we wanted to study the renal biodistribution of gadolinium-based nanoparticles with the aim to show the potential of LIBS for label-free imaging of nanoparticles, and metals more generally, in their biological environment.
What benefits can LIBS offer for such studies compared to existing elemental methods?
The LIBS approach is the only optical technique able to image elements with parts-per-million sensitivity and a resolution in the range of a few micrometers. The simplicity of the setup endows LIBS with a series of advantages including an all-optical design compatible with standard optical microscopy, reduced cost, and operation in ambient atmosphere. In addition, although LIBS is not as sensitive as laser-ablation ICP-MS (LA-ICP-MS), which has a sensitivity of
< 500 ppb, or as high-resolving as micro X-ray fluorescence spectrometry (< 1 µm), it has a fast operating speed, which can be up to 100 times faster than other techniques. These advantages enable LIBS imaging to provide large-scale images over entire organs (> 1 cm²) with a cellular scale resolution (~10 µm) and in reasonable time periods.
How did you optimize the balance between spatial resolution and detection sensitivity?
Achieving an adequate balance between spatial resolution and detection sensitivity is indeed one of the critical points for setting up a LIBS imaging instrument. In LIBS, sensitivity and the resolution are tightly linked. Resolution is governed by the size of the ablation craters, whereas sensitivity relies on the amount of vaporized material and on the excitation capability of the laser pulse. For example, improving the resolution (using for example an objective with a higher magnification) may dramatically lower the sensitivity. Considering our given experimental configuration (UV laser pulse, x15 magnification objective, collection system, and so on) the best performance, both in terms of signal-to-noise ratio and crater size, was obtained for a pulse energy of 800 µJ.
In this study, the kidney samples were embedded in epoxy before analysis. Would it be possible to do this kind of analysis on fresh tissue?
Yes, our first studies were carried out on fresh tissue slices (2,3). However, mastering the laser ablation of such soft materials is difficult and the accessible resolution is quite poor (~100 µm). The use of epoxy-embedded samples offers better ablation control compared with fresh tissue, which greatly improves the sample’s mechanical stability, the ablation shot-to-shot repeatability, and thus the accessible resolution.
How were the elements of interest quantified using the imaging technique?
A standard quantification methodology was used. Standards were made in epoxy resin with graduated concentration of the elements of interest (Gd, Si, Na, Fe). Calibration curves were then built for each element and the relative-abundance images were subsequently transformed into quantitative-abundance images.
What limits of detection were you able to attain for the various elements you measured?
We typically obtained detection limits of 10 ppm for Gd, 2 ppm for Si, and 5 ppm for Na, values estimated from single-shot measurements.
What were the initial results in this study, particularly as you studied the renal clearance of the nanoparticles over time?
These LIBS experiments made it possible to image the components of the nanoparticles (Gd and Si) throughout the entire kidney and provided evidence of both their rapid renal uptake and their elimination. We also imaged elements naturally present in the tissue such as Na, Ca, Fe, Cu, P, and Mg. The kinetics of nanoparticle elimination was evaluated from kidney samples collected at various times after nanoparticle administration (Figure 2). These results highlight the appropriate elimination of nanoparticles from the organism, which is a fundamental necessity for the clinical development of such agents.
Figure 2: Quantitative imaging of Gd and Na in kidney coronal sections showing the fast renal uptake and elimination of nanoparticles. Mouse kidney samples were collected at various times after nanoparticle administration (ranging from 5 min to 1 week).
Overall, what did your study show about the suitability of LIBS to this type of work?
The LIBS imaging capability is very complementary to conventional methods for biological research, such as transmission electron microscopy or immunohistochemistry. Importantly, this work has drawn the attention of biologists working in the field of nanotechnology. They are seeing in LIBS a promising approach for preclinical studies with the idea to provide the biodistribution of nanomaterials at the scale of an entire organ. Since doing this study, we have worked on several applications where LIBS imaging has been used among other techniques to evaluate the biological activity and toxicity of various types of nanoparticles (4–6).
How large an area did you sample, and how long did it take? How feasible is this method for mapping larger samples?
Imaging large areas by LIBS is not technically difficult as long as the sample is perfectly flat (or alternatively the instrument has to be equipped with an autofocus system). The scanning speed is critical, however, because recording high-definition images with a large number of pixels can take a considerable amount of time. All the images shown above were obtained using an instrumentation operating at 10 Hz. With this acquisition rate, 36,000 pixels were recorded within 1 h. This year, we updated our instruments to increase the scanning rate to 100 Hz. This obviously provides more comfortable conditions for analyzing larger samples. An example is shown in Figure 3. This image of 500,000 pixels represents a rat kidney section (~2 cm²) and was recorded in about 80 min at 100 Hz.
Figure 3: High-definition image showing the biodistribution of gold-based nanoparticles in a rat kidney. This image of around 500,000 pixels was recorded with a resolution of 20 µm and covers a surface of about 2 cm².
What are your next steps in this work?
As already mentioned, we are working on the improvement of the acquisition rate. We should be able to make our first kilohertz tests in a couple of weeks. In addition, we are also investigating the three-dimensional capability of the technique with the aim both to provide a global organ distribution of metals and to focus on specific regions of the tissue with cellular-scale resolution. At the present time, we are also working on samples with more complex architectures such as the brain, lungs, or tumors, and with other types of metal-based nanoparticles (gold, silver, platinum, and so on).
References
L. Sancey, V. Motto-Ros, B. Busser, S. Kotb, J.M. Benoit, A. Piednoir, F. Lux, O. Tillement, G. Panczer, and J. Yu, Scientific Reports4, 6065 (2014).
V. Motto-Ros et al., Spectrochim. Acta, Part B83, 168–174 (2013).
V. Motto-Ros et al., Appl. Phys. Lett.101, 223702 (2012).
L. Sancey et al., ACS Nano9, 2477 (2015).
A. Moussaron et al., Small11, 4900 (2015).
S. Kunjachan et al., Nano Letters15, 7488 (2015).
Vincent Motto-Ros is an associate professor at the Light and Matter Institute at Lyon 1 University, in Lyon, France.
Chemotherapy Benefit in Early Breast Cancer Patients
Larry H Bernstein, MD, FCAP, Curator
LPBI
Agendia’s MammaPrint® First and Only Genomic Assay to Receive Level 1A Clinical Utility Evidence for Chemotherapy Benefit in Early Breast Cancer Patients
Clinical high-risk patients with a low-risk MammaPrint® result, including 48 percent node-positive, had five-year distant metastasis-free survival rate in excess of 94 percent, whether randomized to receive adjuvant chemotherapy or not
MammaPrint could change clinical practice by substantially de-escalating the use of adjuvant chemotherapy and sparing many patients an aggressive treatment they will not benefit from
Forty-six percent overall reduction in chemotherapy prescription among clinically high-risk patients
April 19, 2016 / B3C newswire / —Agendia, Inc., together with the European Organisation for Research and Treatment of Cancer (EORTC) and Breast International Group (BIG), announced results from the initial analysis of the primary objective of the Microarray In Node-negative (and 1 to 3 positive lymph node) Disease may Avoid ChemoTherapy (MINDACT) study at the American Association for Cancer Research Annual Meeting 2016 in New Orleans, LA.
Using the company’s MammaPrint® assay, patients with early-stage breast cancer who were considered at high risk for disease recurrence based on clinical and biological criteria had a distant metastasis-free survival at five years in excess of 94 percent.The MammaPrint test—the first and only genomic assay with FDA 510(k) clearance for use in risk assessment for women of all ages with early stage breast cancer—identified a large group of patients for whom five-year distant metastasis–free survival was equally good whether or not they received adjuvant chemotherapy (chemotherapy given post-surgery).
“The MINDACT trial design is the optimal way to prove clinical utility of a genomic assay,” said Prof. Laura van ’t Veer, CRO at Agendia, Leader, Breast Oncology Program, and Director, Applied Genomics at UCSF Helen Diller Family Comprehensive Cancer Center. “It gives the level 1A clinical evidence (prospective, randomized and controlled) that empowers physicians to clearly and confidently know when chemotherapy is part of optimal early-stage breast cancer therapy. In this trial, MammaPrint (70-gene assay) was compared to the standard of care physicians use today, to decide what is the best treatment option for an early-stage breast cancer patient.”
The MINDACT trial is the first prospective randomized controlled clinical trial of a breast cancer recurrence genomic assay with level 1A clinical evidence and the first prospective translational research study of this magnitude in breast cancer to report the results of its primary objective.
Among the 3,356 patients enrolled in the MINDACT trial, who were categorized as having a high risk of breast cancer recurrence based on common clinical and pathological criteria (C-high), the MammaPrint assay reduced the chemotherapy treatment prescription by 46 percent.Using the 70-gene assay, MammaPrint, 48 percent of lymph-node positive breast cancer patients considered clinically high-risk (Clinical-high) and genomic low-risk (MammaPrint-low) had an excellent distant metastasis-free survival at five years in excess of 94 percent.
“Traditionally, physicians have relied on clinical-pathological factors such as age, tumor size, tumor grade, lymph node involvement, and hormone receptor status to make breast cancer treatment decisions,” said Massimo Cristofanilli, MD, Associate Director of Translational Research and Precision Medicine at the Robert H. Lurie Comprehensive Cancer Center, Northwestern University in Chicago. “These findings provide level 1A clinical utility evidence by demonstrating that the detection of low-risk of distant recurrence reported by the MammaPrint test can be safely used in the management of thousands of women by identifying those who can be spared from a toxic and unnecessary treatment.”
MINDACT is a randomized phase III trial that investigates the clinical utility of MammaPrint, when compared (or – “used in conjunction with”) to the standard clinical pathological criteria, for the selection of patients unlikely to benefit from adjuvant chemotherapy. From 2007 to 2011, 6,693 women who had undergone surgery for early-stage breast cancer enrolled in the trial (111 centers in nine countries). Participants were categorized as low or high risk for tumor recurrence in two ways: first, through analysis of tumor tissue using MammaPrint at a central location in Amsterdam; and second, using Adjuvant! Online, a tool that calculates risk of breast cancer recurrence based on common clinical and biological criteria.
Patients characterized in both clinical and genomic assessments as “low- risk” are spared chemotherapy, while patients characterized as “high- risk” are advised chemotherapy. Those with conflicting results are randomized to use either clinical or genomic risk (MammaPrint) evaluation to decide on chemotherapy treatment.
The MINDACT trial is managed and sponsored by the EORTC as part of an extensive and complex partnership in collaboration with Agendia and BIG, and many other academic and commercial partners, as well as patient advocates.
“These MINDACT trial results are a testament that the science of the MammaPrint test is the most robust in the genomic breast recurrence assay market. Agendia will continue to collaborate with pharmaceutical companies, leading cancer centers and academic groups on additional clinical research and in the pursuit of bringing more effective, individualized treatments within reach of cancer patients,” said Mark Straley, Chief Executive Officer at Agendia. “We value the partnership with the EORTC and BIG and it’s a great honor to share this critical milestone.”
Breast cancer is the most frequently diagnosed cancer in women worldwide(1). In 2012, there were nearly 1.7 million new breast cancer cases among women worldwide, accounting for 25 percent of all new cancer cases in women(2).
Crystal Resolution in Raman Spetctoscopy for Pharmaceutical Analysis, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 1: Next Generation Sequencing (NGS)
Crystal Resolution in Raman Spetctoscopy for Pharmaceutical Analysis
Curator: Larry H. Bernstein, MD, FCAP
Investigating Crystallinity Using Low Frequency Raman Spectroscopy: Applications in Pharmaceutical Analysis
Crystallinity is an important factor when producing pharmaceuticals because it directly affects the bioavailability of the drug. Low-frequency Raman spectroscopy offers some advantages to the detection and analysis of crystallinity in pharmaceutical samples. Here the experimental requirements for low-frequency Raman measurements are described. The application of the technique to the study of crystallinity with a number of examples is discussed and the advantages and limitations are highlighted and compared with other techniques.
Raman spectroscopy is typically a nondestructive technique that uses lasers to probe for information about intra- and intermolecular bond vibrations. It involves irradiating the sample with monochromatic light to excite molecules within the sample to a virtually excited state. The molecules then relax to a higher (Stokes scattering) or lower (anti-Stokes scattering) vibrational level—resulting in the scattered light being lower or higher in frequency than the irradiating photons, respectively. Raman scattering is produced when this inelastic scattering occurs, and can be used to deduce information about the nature of the sample. Relative to the incident light, Raman scattering is a rare phenomenon, occurring in only 10-6 of the irradiated molecules (1). What occurs more commonly is elastic scattering, Rayleigh scattering, where the light emitted from the sample has the same energy as the incident light. The intensity of Raman scattering is determined by whether the vibrational mode of the molecule has a change in polarizability along its normal coordinate—similar to how infrared (IR) spectroscopy requires a changing dipole moment (1). For example, H2O has vibrational modes that have large dipole moment changes along the normal coordinate and as such it has strong IR bands. However, as H2O is an σ-bonded compound, the electrons have low polarizability and the change in polarizability with vibration is also low, hence the Raman scattering is weak. Conversely, the active pharmaceutical ingredients (APIs) in many drugs are generally π-bonded and easily polarizable, thus producing strong features in Raman spectra (2). This is one of the reasons why Raman spectroscopy is actively used in pharmaceutical studies.
Crystallinity and Why It Is Important for Pharmaceuticals
The term crystalline is used to describe solids in which the atoms or molecules are arranged in an ordered manner. For many pharmaceuticals, the crystalline form is more kinetically stable than the amorphous form, which typically results in crystalline solids being less soluble, and therefore less bioavailable than their amorphous counterparts. The influence that crystallinity has over the solubility of a solid is what makes this an important factor when manufacturing pharmaceuticals, as those which are supposed to be fast-acting should be readily soluble in the body, while slow-acting drugs should dissolve relatively slowly. This has been shown to be the case in previous studies of bioavailability of APIs in the literature (3–5). Because of this, it is important to be able to control the crystallinity of a drug and monitor it. However, the crystallinity of a sample cannot be assumed to be simply either 100% ordered or 100% amorphous. Instead, the literature has indicated that disorder occurs along a continuous scale where a sample can gradually become more or less crystalline before becoming fully ordered or disordered (6). Crystallinity can be controlled in a number of different ways. One such method involves creating a fully crystalline pharmaceutical product and introducing disorder into the structure mechanically by milling the sample (4,5). Previous studies have also shown that amorphous pharmaceuticals can be produced through different drying methods (7,8). Even accidental adjustment of properties such as moisture level can lead to a change in crystallinity (9,10). Methods to monitor the crystallinity of a sample are therefore useful.
Established methods for monitoring crystallinity include calorimetry and X-ray diffraction (XRD) (6). Terahertz spectroscopy is arguably a more recently implemented method of analysis of crystallinity and is the only one of these three techniques described in any detail here. The first terahertz technique described is terahertz absorption spectroscopy or far-IR spectroscopy. The terahertz radiation has a wavelength that is between that of IR and microwaves (0.1–1 mm or 10–100 cm-1) and has the ability to allow observation of various low energy vibrations within the sample (11–13). Vibrations can essentially be divided into two categories when studying crystalline samples: external and internal vibrational modes. Internal vibrations can be considered the local vibrations that occur within each molecule (that is, intramolecular bond vibrations). External vibrations can be considered the vibrations involving the overall lattice of the crystal, such as phonon modes or torsional vibrations (14,15). Terahertz spectroscopy tends to focus on the external vibrations. The technique has been known for more than 100 years, and two experimental challenges have inhibited its widespread use. The first of these is the presence of strong water absorption above 60 cm-1, which has complicated analysis of wet samples (16) and the second is the presence of ambient terahertz radiation that compromises spectral detection (17). The second of these issues has been solved in large part with the advent of terahertz time-domain spectroscopy, also known as terahertz pulsed spectroscopy (TPS), which generates terahertz radiation using femtosecond laser pulses, using a laser wavelength of around 780 nm. For more in-depth detail about terahertz spectroscopy, refer to references 11–13. Despite its relative underutilization, this technique has been found to be an effective method for observing and even quantifying crystallinity in various papers in the literature. Model compounds such as cellulose and gelatin–amino acid mixtures have been used previously to determine the efficacy of the technique when compared to XRD (18) or to determine whether it would be an efficient method for in-line and off-line analysis of pharmaceutical crystallinity (19). In particular, the use of terahertz spectroscopy with a multivariate analysis technique—partial least squares (PLS)—permitted the quantification of the crystallinity index (CI) of cellulose (18). However, not only has the crystallinity of model systems been analyzed using the technique, drugs such as indomethacin, ketoprofen, carbamazepine, enalapril maleate, fenoprofin, irbesartan, and diclofenac acid have also been studied (11,19–23). The use of PLS also proved effective when working with pharmaceutical products, allowing not only real formulations to be distinguished from one another, but also the different crystalline forms to be distinguished. These forms were indicated by Strachan and colleagues to include polymorphic, liquid crystalline, and amorphous states (11,24). Therefore, terahertz spectroscopy has seen growing interest based on its success with the study of pharmaceuticals. Other reviews have also come to compare terahertz spectroscopy with other forms of vibrational spectroscopy for pharmaceutical analysis (25,26).
Low-Frequency Raman
Raman spectroscopy has been used for decades to analyze and characterize polymorphs (27,28); however, like terahertz spectroscopy, low-frequency Raman spectra have been used more recently (29–31). Mid-frequency and high-frequency regions of Raman spectra are typically used to analyze the intramolecular bonding of molecules within a sample (internal vibrations). However, low-frequency Raman spectroscopy involves the analysis of the low-frequency regions of Raman spectra, which tend to contain features attributable to external vibrations of the crystalline lattice (14). Frequencies in the same region as those observed with terahertz spectroscopy can be observed using standard continuous wave lasers. Because Raman scattering is such a rare process compared to Rayleigh scattering, the Rayleigh signal must be removed before recording spectra. Many ways of doing this have been developed. Until the 1990s it was most commonly performed by the use of a triple-grating spectrograph in conjunction with a photomultiplier tube (32,33). This method is reliable but because of the many mirrors and diffraction gratings necessary, the ultimate efficiency or throughput of the system is inevitably low. To compensate for this, high laser powers and long acquisition times are needed (30). A less commonly used method involves the stabilization of iodine gas at specific temperatures to allow for the absorption of the unwanted laser signal (34). However, these methods are cumbersome and it later became more common to use holographic notch or edge filters that reject light with a specific wavelength. These filters can be designed to reject nearly any wavelength of light to match that of the laser used while allowing the rest of the spectrum to pass through. This method allows for the use of more efficient single grating spectrographs. In conjunction with array detectors or charge coupled devices (CCDs) an entire Raman spectrum can be acquired simultaneously. However, the filters also block a portion of the Raman scattering signal in addition to the Rayleigh line. In most cases, the spectra can only be collected above about 100 cm-1 (35).
Over the past decade new filtering technologies have been developed that allow for the acquisition of Raman spectra reaching very close to the laser frequency (35) while using standard dispersive Raman systems with all of their advantages. These filters are typically based on holographic reflective volume Bragg gratings (VBGs) and can achieve frequencies as low as 5 cm-1(34).
In most cases low-frequency Raman is measured using near-IR (NIR) diode laser sources for excitation. This is because the selectivity of VBGs decreases as wavelength decreases. Therefore, it is easier to acquire low-frequency Raman data with longer-wavelength lasers (34).
An issue that arises when performing low-frequency Raman measurements are artifacts caused by the laser itself (34). These may not be apparent when standard holographic filters are used because they appear so close to the laser line and would normally be blocked. These artifacts arise from laser instability, temperature fluctuations, amplified spontaneous emission, and plasma lines. VBG filters are therefore also used to clean up the laser line before it is focused onto the sample. In some optical designs a single VBG is used for both purposes simultaneously (34,35).
The resulting spectrum then allows detection of crystalline or amorphous forms, as the low-frequency Raman bands are associated with the low energy bond vibrations, hydrogen bonds, and phonon modes. Therefore, the overall environment and arrangement of the molecules will have a considerable effect on these vibrations. If the sample is crystalline, then the molecules will be highly ordered, and the low energy bond vibrations will show up in the spectrum as sharp peaks because the bonds will have a very similar environment. Conversely, if the sample is amorphous, there will be very little order and there will be a wide variety of molecular environments which, in turn, will typically produce only one broad band, known as the boson peak, and no other low-frequency features (36). Illustrations of crystalline and amorphous spectra are provided later in this article. This distinct difference between crystalline and amorphous materials is what suggests that low-frequency Raman spectroscopy could prove to be extremely useful in the study of pharmaceuticals.
Analysis
Overview of the Setup of a Low-Frequency Raman System
The exemplar data presented in this article were collected using two different low-frequency Raman systems, the first of which (Figure 1) is a home-built system based on a wavelength-stabilized 80-mW, 785-nm laser module (Ondax, Inc). The laser line is initially filtered by use of two BragGrate reflective VBGs (OptiGrate Corp.) to remove amplified spontaneous emission. The sample is arranged in a 135° backscattering geometry relative to a collection lens (f/2.3). The collected light is passed through a pair of VBGs (OptiGrate Corp. BragGrate 785 nm, OD3). The collimated, filtered light is then focused by a second f/2.3 lens onto a fiber-optic cable. The cable is coupled to an LS 785 spectrograph (Princeton Instruments). A third VBG is used to further filter light imaged onto the entrance slit of the spectrograph. Detection is achieved using a CCD (Princeton Instruments thermoelectrically cooled PIXIS 100 BR CCD). A typical spectral range for this experiment is about 2400 cm-1. Both the Stokes and anti-Stokes region of the spectrum can be seen in addition to a small remaining laser line signal. Raman standards such as sulfur and 1,4-bis(2-methylstyryl)benzene (BMB) are used to calibrate the spectrometer (37).
Figure 1: Illustration of an exemplar low-frequency Raman setup with a 785-nm laser.
The second system is based on a pre-built SureBlock XLF-CLM THz-Raman system from Ondax Inc. The laser (830 nm, 200 mW), cleanup filters, and laser line filters are all self-contained inside of the instrument but operate on the same principles as the 785-nm system. The sample is arranged in a 180° backscattering geometry relative to a 10× microscope lens. This system is then coupled via a fiber-optic cable to a Princeton Instruments SP2150i spectrograph and PIXIS 100 CCD camera. The 0.15-m spectrograph is used in conjunction with either a 1200- or 1800-groove/mm blazed diffraction grating to adjust the resolution and spectral range.
Crystalline Versus Amorphous Samples
The Raman spectrum of crystalline and amorphous solids differ greatly in the low-frequency region (see Figure 2) because of the highly ordered and highly disordered molecular environments of the respective solids. However, the mid-frequency region can also be noticeably altered by the changing environment (Figure 3).
A potential issue is optical artifacts, and these may be identified by the analysis of both Stokes and anti-Stokes spectra. One advantage of the experimental setups described is that signal from the sample may be measured within minutes and it is nondestructive, thus allowing Raman spectra to be collected from a single sample using both techniques at virtually the same time. This approach permits the examination of low-frequency Raman data with 785-nm and 830-nm excitation and allows comparison with Fourier transform (FT)-Raman spectra, in which it is possible to collect meaningful data down to a Raman shift of 50 cm-1. The benefits are demonstrated in Figure 4. In this data, each technique produces consistent bands with similar Raman shifts and relative intensities. While Raman data were not collected below 50 cm-1 using the 1064-nm system, the bands at 69 and 96 cm-1 are consistent with the 785- and 830-nm data. Furthermore, the latter two methods show consistency with bands appearing around 32 and 46 cm-1 for both techniques.
Figure 4: Comparison of the low-frequency region of three Raman spectroscopic techniques.
Case Studies
So far there have been few studies to utilize low-frequency Raman spectroscopy in the analysis of pharmaceutical crystallinity. Despite this, the literature does contain articles that demonstrate the promising applicability of the technique.
Mah and colleagues (38) studied the level of crystallinity of griseofulvin using low-frequency Raman spectroscopy with PLS analysis. In this study a batch of amorphous griseofulvin (which was checked using X-ray powder diffractometry) was prepared by melting the griseofulvin and rapidly cooling it again using liquid nitrogen. Condensed water was removed by placing the sample over phosphorus pentoxide and the glassy sample was then ground using mortar and pestle. Calibrated samples of 2%, 4%, 6%, 8%, and 10% crystallinity were then created though geometric mixing of the amorphous and crystalline samples; following this mixing, the samples were then pressed into tablets. Many tablets were then stored in differing temperatures (30 °C, 35 °C, and 40 °C) at 0% humidity. Low-frequency 785-nm, mid-frequency 785-nm, and FT-Raman spectroscopies were performed simultaneously on each sample. After PLS analysis, limits of detection (LOD) and limits of quantification (LOQ) were calculated. The results of this research showed that each of these three techniques were capable of quantifying crystallinity. It also showed that FT-Raman and low-frequency Raman techniques were able to both detect and quantify crystallinity earlier than the mid-frequency 785 nm Raman technique. The respective LOD and LOQ values for FT-Raman, low-frequency Raman, and mid-frequency Raman are as follows: LOD values: 0.6%, 1.1%, and 1.5%; LOQ values: 1.8%, 3.4%, and 4.6%. The root mean squared errors of prediction (RMSEP) were also calculated and, like the LOD and LOQ values, indicated that the FT-Raman data had the lowest error, followed by the low-frequency Raman, and mid-frequency Raman had the largest errors of the three techniques. The recrystallization tests that were performed indicated that higher temperatures showed a distinct increase in the rate of recrystallization and that each technique provided similar results (within experimental error). It is also important to note that each technique gave similar spectra (where applicable), which provides supporting evidence that the data is meaningful. Overall, the conclusions of this research were that low-frequency predictions of crystallinity are at least as accurate as the predictions made using mid-frequency Raman techniques. It is arguable that low-frequency Raman is better because of the presence of stronger spectral features and because they are intrinsically linked with crystallinity.
Hédoux and colleagues (36) investigated the crystallinity of indomethacin using low-frequency Raman spectroscopy and compared the results with high frequency data. The ranges of interest were indicated to be 5–250 cm-1and 1500–1750 cm-1 regions. Samples of indomethacin were milled using a cryogenic mill to avoid mechanical heating of the sample, with full amorphous samples being obtained after 25 min of milling. Methods used in this study include Raman spectroscopy, isothermal differential scanning calorimetry (DSC), and X-ray diffractometry as well as the milling technique. The primary objective of this research was to use all of these techniques to monitor the crystallization of amorphous indomethacin to the more stable γ-state while the sample was at room temperature–well below the glass transition temperature,Tg = 43 °C. The results of this research did in fact show that low-frequency Raman spectroscopy is a very sensitive technique for identifying very small amounts of crystallinity within mostly amorphous samples. The data was supported by the well-established methods for monitoring crystallinity: XRD and DSC. This paper particularly noted the benefit of low acquisition times associated with low-frequency Raman spectroscopy compared with the other techniques used.
Low-frequency Raman spectroscopy was also used to monitor two polymorphic forms of caffeine after grinding and pressurization of the samples (39). Pressurization was performed hydrostatically using a gasketed membrane diamond anvil cell (MDAC), while ball milling was used as the method of grinding the sample. Analysis methods used were low-frequency Raman and X-ray diffraction. Low-frequency Raman spectra revealed that, upon slight pressurization, caffeine form I transforms into a metastable state slightly different from that of form II and that a disordered (amorphous) state is achieved in both forms when pressurized above 2 GPa. In contrast, it is concluded that grinding results in the transformation of each form into the other with precise grinding times, thus also generating an intermediate form, which was found to only be observable using low-frequency Raman spectroscopy. The caffeine data, as well as the low-frequency data obtained for indomethacin were further discussed by Hédoux and colleagues (40).
Larkin and colleagues (41) used low-frequency Raman in conjunction with other techniques to characterize several different APIs and their various forms. The other techniques include FT-Raman spectroscopy, X-ray powder diffraction (XRPD), and single-crystal X-ray diffractometry. The APIs studied include carbamazepine, apixaban diacid co-crystals, theophylline, and caffeine and were prepared in various ways that are not detailed here. During this research, low-frequency Raman spectroscopy played an important role in understanding the structures while in their various forms. However, more importantly, low-frequency Raman spectroscopy produced information-rich regions below 200 cm-1 for each of the crystalline samples and noticeably broad features when the APIs were in solution.
Wang and colleagues (42) investigated the applicability of low-frequency Raman spectroscopy in the analysis of respirable dosage forms of various pharmaceuticals. The analyzed pharmaceuticals were involved in the treatment of asthma or chronic obstructive pulmonary disease (COPD) and include salmeterol xinafoate, formoterol fumarate, glycopyrronium bromide, fluticasone propionate, mometasone furoate, and salbutamol sulfate. Various formulations of amino acid excipients were also analyzed in this study. Results indicated that the use of low-frequency Raman analysis was beneficial because of the large features found in the region and allowed for reliable identification of each of the dosage forms. Not only this, it also allowed unambiguous identification of two similar bronchodilators, albuterol (Ventolin) and salbutamol (Airomir).
Heyler and colleagues (43) collected both the low-frequency and fingerprint region of Raman spectra from several polymorphs of carbamazepine, an anticonvulsant and mood stabilizer. This study found that the different polymorphs of this API could be distinguished effectively using these two regions. Similarly, Al-Dulaimi and colleagues (44) demonstrated that polymorphic forms of paracetamol, flufenamic acid, and imipramine hydrochloride could be screened using low-frequency Raman and only milligram quantities of each drug. In this study, paracetamol and flufenamic acid were used as the model compounds for comparison with a previously unstudied system (imipramine hydrochloride). Features within the low-frequency Raman regions of spectra were shown to be significantly different between forms of each drug. Therefore this study also indicated that the polymorphs were highly distinguishable using the technique. Hence, like all other previously mentioned case studies, these investigations further demonstrate the utility of low-frequency Raman spectroscopy as a fast and effective method for screening pharmaceuticals for crystallinity.
Conclusions
Low-frequency Raman spectroscopy is a new technique in the field of pharmaceuticals, as well as in general studies of crystallinity. This is despite indications in previous studies showing an innate ability of the technique for identifying crystalline materials and in some cases, quantifying crystallinity. Arguably one of the most beneficial aspects of this technique is the relatively small amount of time necessary to prepare and analyze samples when compared with XRD or DSC. This should ensure the growing use of low-frequency Raman spectroscopy in, not only pharmaceutical crystallinity studies, but also crystallinity studies of other substances as well.
References
J.R. Ferraro and K. Nakamoto, Introductory Raman Spectroscopy, 1st Edition (Academic Press, San Diego, 1994).
K.C. Gordon and C.M. McGoverin, Int. J. Pharm. 417, 151–162 (2011).
D. Law et al., J. Pharm. Sci.90, 1015–1025 (2001).
G.H. Ward and R.K. Schultz, Pharm. Res.12, 773–779 (1995).
M.D. Ticehurst et al., Int. J. Pharm.193, 247–259 (2000).
M. Rani, R. Govindarajan, R. Surana, and R. Suryanarayanan, Pharm. Res.23, 2356–2367 (2006).
M.J. Pikal, in Polymorphs of Pharmaceutical Solids, H.G. Brittain, Ed. (Marcel Dekker, New York, 1999), pp. 395–419.
M. Ohta and G. Buckton, Int. J. Pharm.289, 31–38 (2005).
J. Han and R. Suryanarayanan, Pharm. Dev. Technol.3, 587–596 (1998).
S. Debnath and R. Suryanarayanan, AAPS PharmSciTech. 5, 1–11 (2004).
C.J. Strachan, T. Rades, D.A. Newnham, K.C. Gordon, M. Pepper, and P.F. Taday, Chem. Phys. Lett.390, 20–24 (2004).
Y.C. Shen, Int. J. Pharm.417, 48–60 (2011).
G.W. Chantry, in Submillimeter Spectroscopy: A Guide to the Theoretical and Experimental Physics of the Far Infrared, 1st Edition (Academic Press Inc. Ltd., Waltam, 1971).
D. Tuschel, Spectroscopy30(9), 18–31 (2015).
P.M.A. Sherwood, Vibrational Spectroscopy of Solids (Cambridge University Press, Cambridge, 1972).
L. Ho et al., J. Control. Release. 119, 253–261 (2007).
V.P. Wallace et al., Faraday Discuss. 126, 255–263 (2004).
F.S. Vieira and C. Pasquini, Anal. Chem. 84, 3780–3786 (2014).
J. Darkwah, G. Smith, I. Ermolina, and M. Mueller-Holtz, Int. J. Pharm.455, 357–364 (2013).
S. Kojima, T. Shibata, H. Igawa, and T. Mori, IOP Conf. Ser. Mater. Sci. Eng.54, 1–6 (2014).
T. Shibata, T. Mori, and S. Kojima, Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 150, 207–211 (2015).
S.P. Delaney, D. Pan, M. Galella, S.X. Yin, and T.M. Korter, Cryst. Growth Des.12, 5017–5024 (2012).
The concept of group wavenumbers is defined, and the importance of recognizing patterns in infrared (IR) spectra is discussed. Continuing our theme of investigating the IR spectra of hydrocarbons, we look at the nature of aromatic bonding and why aromatic rings have unique structures, bonding, and IR spectra. The IR spectrum of benzene is analyzed in detail as a prototype example of an aromatic hydrocarbon.
A look at the spectrum of any pure molecule will disclose that many functional groups have multiple peaks. It is this pattern of peaks that defines the presence of a specific functional group in a sample, not one specific peak. Thus, interpreting spectra is not about memorizing peak positions, but is instead an exercise in pattern recognition. The human brain has evolved to be great at pattern recognition. Computers are good at many things except pattern recognition, hence the need to use our eyeballs and brains to interpret infrared (IR) spectra. A computer’s inability to interpret spectra is part of why there is a need for a column series like this one. Since humans are good pattern recognizers, I believe most people can learn to interpret IR spectra. My approach going forward will be to emphasize the pattern of peaks that defines the presence of a functional group in a sample rather than throwing hundreds of peak positions at you.
The peaks that define a functional group are what I will call its group wavenumbers. Some people call these “group frequencies,” but this is a misnomer. The x-axes of IR spectra are plotted in wavenumber, not frequency. A good group wavenumber peak has three useful properties:
It will be intense so it is easy to see.
It will appear in a unique wavenumber region where no other functional groups absorb.
It will fall in a narrow wavenumber range regardless of what molecule the functional group appears in.
We have talked about group wavenumbers and peak patterns in previous articles without realizing it. For example, the methyl and methylene C-H stretches that were introduced previously (1). Now, we have given these peak patterns a proper name.
To be clear, not all of the peaks from a given functional group will be useful group wavenumbers. For example, a peak may be too weak to be seen reliably, may appear in a region where lots of other peaks appear, or move around a lot from molecule to molecule. Many of the upcoming columns will be devoted to the useful group wavenumbers of economically important functional groups.
One final thought: I compare interpreting IR spectra to playing a piano. You can’t just walk up to a Steinway and play a Beethoven sonata, you have to practice first. Similarly, you can’t just walk up to a spectrum and pull information out of it, you have to practice first. The purpose of these columns is to give you the knowledge you need to interpret spectra and the problem spectra give you the opportunity to practice what you have learned.
Introduction to Aromatic Molecules
To be able to interpret IR spectra, one must have a nodding familiarity with the nomenclature and structures of organic chemistry. In today’s world many people interpreting spectra do not have this background, so I have and will continue to discuss the structure and nomenclature of functional groups whose spectra we will discuss.
Aromatic molecules were originally named because some of them smell nice. However, if you have ever smelled pyridine you know that not all aromatic molecules do smell nice. It has been found that the bonding in aromatic molecules is unique.
The prototypical aromatic molecule is benzene, C6H6, whose structure is represented in Figure 1.
Figure 1: The chemical structure of the benzene molecule, C6H6.
The drawing in Figure 1 is that of a six-membered ring or hexagon. A carbon atom is located at each vertex of the hexagon and a hydrogen atom is attached to each carbon, although it is not written in. The circle inside the ring represents that the electrons are delocalized which is illustrated in Figure 2.
Figure 2: Top: The P orbitals on each of the six carbon atoms in benzene that contribute an electron to the ring. Bottom: the collection of delocalized P orbital electrons forming a cloud of electron density above and below the benzene ring.
Each of the carbon atoms in a benzene ring contains two P orbitals containing a lone electron, and one of these orbitals is perpendicular to the benzene ring as seen in the top of Figure 2. There is enough orbital overlap that these electrons, rather than being confined between two carbon atoms as might be expected, instead delocalize and form clouds of electron density above and below the plane of the ring. This type of bonding is called aromatic bonding(2), and a ring that has aromatic bonding is called an aromatic ring. It is aromatic bonding that gives aromatic rings their unique structures, chemistry, and IR spectra. Benzene is simply a commonly found aromatic ring. Other types of aromatic molecules include polycyclic aromatic hydrocarbons (PAHs), such as naphthalene, that contain two or more benzene rings that are fused (which means adjacent rings share two carbon atoms), and heterocyclic aromatic rings which are aromatic rings that contain a noncarbon atom such as nitrogen. Pyridine is an example of one of these. The interpretation of the IR spectra of these latter aromatic molecules will be discussed in future articles.
Figure 3: The IR spectrum of benzene, measured as a capillary thin film between two KBr windows.
……
The area labeled B in Figure 3 refers to a region in aromatic ring spectra called the summation bands. More information on these peaks will come in a later column. The label C in Figure 3 at 1478 cm-1 is an example of a ring mode peak. A ring mode is a vibration that involves the stretching and contracting of the carbon-carbon bonds in an aromatic ring. These are typically sharp, but vary in number and intensity depending upon the molecule. They fall between 1620 and 1400 cm-1 as stated in Table I. The region labeled D in Figure 3 is where aromatic ring C-H in-plane bending peaks fall. These peaks are generally medium to weak in intensity, show up in a very busy spectral region, and hence are not useful group wavenumbers.
The most intense peak in Figure 3, labeled E, is an out-of-plane C-H bend. Since aromatic rings are planar, all the hydrogens are in the plane of the molecule. When these hydrogens bend above and below the plane of the molecule they are undergoing a C-H out-of-plane bend, which is sometimes called a wag because of the vibration’s resemblance to the wagging of a dog’s tail. This vibration gives rise to a large peak that typically falls between 1000 cm-1 and 700 cm-1. In the spectrum of benzene, this peak falls at 674 cm-1 because the molecule is unsubstituted.
Conclusion
To review then, the useful group wavenumbers for benzene rings are one or more C-H stretches between 3100 and 3000 cm-1, one or more sharp ring modes between 1620 and 1400 cm-1, and an intense ring bend from 1000 to 700 cm-1. Most of the time when a benzene ring is encountered it contains one or more substituents. The IR spectra of mono- and disubstituted benzene rings will be the topic of the next installment of this column.
Super-Resolution Fluorescence Microscopy: Where To Go Now? Bernd Rieger, Quantitative Imaging Group Leader, Delft University of Technology
09:30
Keynote Presentation
From Molecules To Whole Organs Francesco Pavone, Principal Investigator, LENS, University of Florence
Some examples of correlative microscopies, combining linear and non linear techniques will be described. Particular attention will be devoted Alzheimer disease or to neural plasticity after damage as neurobiological application.
10:15
Super-Resolution Imaging by dSTORM Markus Sauer, Professor, Julius-Maximilians-Universität Würzburg
10:45
Coffee and Networking in Exhibition Hall
11:15
Correlated Fluorescence And X-Ray Tomography: Finding Molecules In Cellular CT Scans Carolyn Larabell, Professor, University of California San Francisco
11:45
Integrating Advanced Fluorescence Microscopy Techniques Reveals Nanoscale Architecture And Mesoscale Dynamics Of Cytoskeletal Structures Promoting Cell Migration And Invasion Alessandra Cambi, Assistant Professor, University of Nijmegen
This lecture will describe our efforts to exploit and integrate a variety of advanced microscopy techniques to unravel the nanoscale structural and dynamic complexity of individual podosomes as well as formation, architecture and function of mesoscale podosome clusters.
12:15
Multi-Photon-Like Fluorescence Microscopy Using Two-Step Imaging Probes George Patterson, Investigator, National Institutes of Health
12:45
Lunch & Networking in Exhibition Hall
14:15
Technology Spotlight
14:30
3D Single Particle Tracking: Following Mitochondria in Zebrafish Embryos Don Lamb, Professor, Ludwig-Maximilians-University
15:00
Visualizing Mechano-Biology: Quantitative Bioimaging Tools To Study The Impact Of Mechanical Stress On Cell Adhesion And Signalling Bernhard Wehrle-Haller, Group Leader, University of Geneva
15:30
Superresolution Imaging Of Clathrin-Mediated Endocytosis In Yeast Jonas Ries, Group Leader, EMBL Heidelberg
We use single-molecule localization microscopy to investigate the dynamic structural organization of the east endocytic machinery. We discovered a striking ring-shaped pre-patterning of the actin nucleation zone, which is key for an efficient force generation and membrane invagination.
16:00
Coffee and Networking in Exhibition Hall
16:30
Optical Imaging of Molecular Mechanisms of Disease Clemens Kaminski, Professor, University of Cambridge
17:00
3-D Optical Tomography For Ex Vivo And In Vivo Imaging James McGinty, Professor, Imperial College London
17:30
End Of Day One
Wednesday, 15 June 2016
09:00
Imaging Gene Regulation in Living Cells at the Single Molecule Level James Zhe Liu, Group Leader, Janelia Research Campus, Howard Hughes Medical Institute
09:30
Keynote Presentation
Super-Resolution Microscopy With DNA Molecules Ralf Jungmann, Group Leader, Max Planck Institute of Biochemistry
10:15
A Revolutionary Miniaturised Instrument For Single-Molecule Localization Microscopy And FRET Achillefs Kapanidis, Professor, University of Oxford
10:45
Coffee and Networking in Exhibition Hall
11:15
Democratising Live-Cell High-Speed Super-Resolution Microscopy Ricardo Henriques, Group Leader, University College London
Information In Localisation Microscopy Susan Cox, Professor, Kings College London
12:45
Lunch & Networking in Exhibition Hall
14:15
Technology Spotlight
14:30
High-Content Imaging Approaches For Drug Discovery For Neglected Tropical Diseases Manu De Rycker, Team Leader, University of Dundee
The development of new drugs for intracellular parasitic diseases is hampered by difficulties in developing relevant high-throughput cell-based assays. Here we present how we have used image-based high-content screening approaches to address some of these issues.
15:00
High Resolution In Vivo Histology: Clinical in vivo Subcellular Imaging using Femtoseceond Laser Multiphoton/CARS Tomography Karsten König, Professor, Saarland University
We report on a certified, medical, transportable multipurpose nonlinear microscopic imagingsystem based on a femtosecond excitation source and a photonic crystal fiber with multiple miniaturized time-correlated single-photon counting detectors.
15:30
Coffee and Networking in Exhibition Hall
16:00
Lateral Organization Of Plasma Membrane Constituents At The Nanoscale Gerhard Schutz, Professor, Vienna University of Technology
It is of interest how proteins are spatially distributed over the membrane, and whether they conjoin and move as part of multi-molecular complexes. In my lecture, I will discuss methods for approaching the two questions, and provide biological examples.
16:30
Correlative Light And Electron Microscopy In Structural Cell Biology Wanda Kukulski, Group Leader, University of Cambridge
Imaging of Cancer Cells, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 1: Next Generation Sequencing (NGS)
Imaging of Cancer Cells
Larry H. Bernstein, MD, FCAP, Curator
LPBI
Microscope uses nanosecond-speed laser and deep learning to detect cancer cells more efficiently
April 13, 2016
Scientists at the California NanoSystems Institute at UCLA have developed a new technique for identifying cancer cells in blood samples faster and more accurately than the current standard methods.
In one common approach to testing for cancer, doctors add biochemicals to blood samples. Those biochemicals attach biological “labels” to the cancer cells, and those labels enable instruments to detect and identify them. However, the biochemicals can damage the cells and render the samples unusable for future analyses. There are other current techniques that don’t use labeling but can be inaccurate because they identify cancer cells based only on one physical characteristic.
Time-stretch quantitative phase imaging (TS-QPI) and analytics system
The new technique images cells without destroying them and can identify 16 physical characteristics — including size, granularity and biomass — instead of just one.
The new technique combines two components that were invented at UCLA:
A “photonic time stretch” microscope, which is capable of quickly imaging cells in blood samples. Invented by Barham Jalali, professor and Northrop-Grumman Optoelectronics Chair in electrical engineering, it works by taking pictures of flowing blood cells using laser bursts (similar to how a camera uses a flash). Each flash only lasts nanoseconds (billionths of a second) to avoid damage to cells, but that normally means the images are both too weak to be detected and too fast to be digitized by normal instrumentation. The new microscope overcomes those challenges by using specially designed optics that amplify and boost the clarity of the images, and simultaneously slow them down enough to be detected and digitized at a rate of 36 million images per second.
A deep learning computer program, which identifies cancer cells with more than 95 percent accuracy. Deep learning is a form of artificial intelligence that uses complex algorithms to extract patterns and knowledge from rich multidimenstional datasets, with the goal of achieving accurate decision making.
The study was published in the open-access journal Nature Scientific Reports. The researchers write in the paper that the system could lead to data-driven diagnoses by cells’ physical characteristics, which could allow quicker and earlier diagnoses of cancer, for example, and better understanding of the tumor-specific gene expression in cells, which could facilitate new treatments for disease.
The research was supported by NantWorks, LLC.
Abstract of Deep Learning in Label-free Cell Classification
Label-free cell analysis is essential to personalized genomics, cancer diagnostics, and drug development as it avoids adverse effects of staining reagents on cellular viability and cell signaling. However, currently available label-free cell assays mostly rely only on a single feature and lack sufficient differentiation. Also, the sample size analyzed by these assays is limited due to their low throughput. Here, we integrate feature extraction and deep learning with high-throughput quantitative imaging enabled by photonic time stretch, achieving record high accuracy in label-free cell classification. Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individual cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. This system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells.
references:
Claire Lifan Chen, Ata Mahjoubfar, Li-Chia Tai, Ian K. Blaby, Allen Huang, Kayvan Reza Niazi & Bahram Jalali. Deep Learning in Label-free Cell Classification. Scientific Reports 6, Article number: 21471 (2016); doi:10.1038/srep21471 (open access)
Supplementary Information
Deep Learning in Label-free Cell Classification
Claire Lifan Chen, Ata Mahjoubfar, Li-Chia Tai, Ian K. Blaby, Allen Huang,Kayvan Reza Niazi & Bahram Jalali
Deep learning extracts patterns and knowledge from rich multidimenstional datasets. While it is extensively used for image recognition and speech processing, its application to label-free classification of cells has not been exploited. Flow cytometry is a powerful tool for large-scale cell analysis due to its ability to measure anisotropic elastic light scattering of millions of individual cells as well as emission of fluorescent labels conjugated to cells1,2. However, each cell is represented with single values per detection channels (forward scatter, side scatter, and emission bands) and often requires labeling with specific biomarkers for acceptable classification accuracy1,3. Imaging flow cytometry4,5 on the other hand captures images of cells, revealing significantly more information about the cells. For example, it can distinguish clusters and debris that would otherwise result in false positive identification in a conventional flow cytometer based on light scattering6.
In addition to classification accuracy, the throughput is another critical specification of a flow cytometer. Indeed high throughput, typically 100,000 cells per second, is needed to screen a large enough cell population to find rare abnormal cells that are indicative of early stage diseases. However there is a fundamental trade-off between throughput and accuracy in any measurement system7,8. For example, imaging flow cytometers face a throughput limit imposed by the speed of the CCD or the CMOS cameras, a number that is approximately 2000 cells/s for present systems9. Higher flow rates lead to blurred cell images due to the finite camera shutter speed. Many applications of flow analyzers such as cancer diagnostics, drug discovery, biofuel development, and emulsion characterization require classification of large sample sizes with a high-degree of statistical accuracy10. This has fueled research into alternative optical diagnostic techniques for characterization of cells and particles in flow.
Recently, our group has developed a label-free imaging flow-cytometry technique based on coherent optical implementation of the photonic time stretch concept11. This instrument overcomes the trade-off between sensitivity and speed by using Amplified Time-stretch Dispersive Fourier Transform12,13,14,15. In time stretched imaging16, the object’s spatial information is encoded in the spectrum of laser pulses within a pulse duration of sub-nanoseconds (Fig. 1). Each pulse representing one frame of the camera is then stretched in time so that it can be digitized in real-time by an electronic analog-to-digital converter (ADC). The ultra-fast pulse illumination freezes the motion of high-speed cells or particles in flow to achieve blur-free imaging. Detection sensitivity is challenged by the low number of photons collected during the ultra-short shutter time (optical pulse width) and the drop in the peak optical power resulting from the time stretch. These issues are solved in time stretch imaging by implementing a low noise-figure Raman amplifier within the dispersive device that performs time stretching8,11,16. Moreover, warped stretch transform17,18can be used in time stretch imaging to achieve optical image compression and nonuniform spatial resolution over the field-of-view19. In the coherent version of the instrument, the time stretch imaging is combined with spectral interferometry to measure quantitative phase and intensity images in real-time and at high throughput20. Integrated with a microfluidic channel, coherent time stretch imaging system in this work measures both quantitative optical phase shift and loss of individual cells as a high-speed imaging flow cytometer, capturing 36 million images per second in flow rates as high as 10 meters per second, reaching up to 100,000 cells per second throughput.
Box 1: The pulse train is spatially dispersed into a train of rainbow flashes illuminating the target as line scans. The spatial features of the target are encoded into the spectrum of the broadband optical pulses, each representing a one-dimensional frame. The ultra-short optical pulse illumination freezes the motion of cells during high speed flow to achieve blur-free imaging with a throughput of 100,000 cells/s. The phase shift and intensity loss at each location within the field of view are embedded into the spectral interference patterns using a Michelson interferometer. Box 2: The interferogram pulses were then stretched in time so that spatial information could be mapped into time through time-stretch dispersive Fourier transform (TS-DFT), and then captured by a single pixel photodetector and an analog-to-digital converter (ADC). The loss of sensitivity at high shutter speed is compensated by stimulated Raman amplification during time stretch. Box 3: (a) Pulse synchronization; the time-domain signal carrying serially captured rainbow pulses is transformed into a series of one-dimensional spatial maps, which are used for forming line images. (b) The biomass density of a cell leads to a spatially varying optical phase shift. When a rainbow flash passes through the cells, the changes in refractive index at different locations will cause phase walk-off at interrogation wavelengths. Hilbert transformation and phase unwrapping are used to extract the spatial phase shift. (c) Decoding the phase shift in each pulse at each wavelength and remapping it into a pixel reveals the protein concentration distribution within cells. The optical loss induced by the cells, embedded in the pulse intensity variations, is obtained from the amplitude of the slowly varying envelope of the spectral interferograms. Thus, quantitative optical phase shift and intensity loss images are captured simultaneously. Both images are calibrated based on the regions where the cells are absent. Cell features describing morphology, granularity, biomass, etc are extracted from the images. (d) These biophysical features are used in a machine learning algorithm for high-accuracy label-free classification of the cells.
On another note, surface markers used to label cells, such as EpCAM21, are unavailable in some applications; for example, melanoma or pancreatic circulating tumor cells (CTCs) as well as some cancer stem cells are EpCAM-negative and will escape EpCAM-based detection platforms22. Furthermore, large-population cell sorting opens the doors to downstream operations, where the negative impacts of labels on cellular behavior and viability are often unacceptable23. Cell labels may cause activating/inhibitory signal transduction, altering the behavior of the desired cellular subtypes, potentially leading to errors in downstream analysis, such as DNA sequencing and subpopulation regrowth. In this way, quantitative phase imaging (QPI) methods24,25,26,27 that categorize unlabeled living cells with high accuracy are needed. Coherent time stretch imaging is a method that enables quantitative phase imaging at ultrahigh throughput for non-invasive label-free screening of large number of cells.
In this work, the information of quantitative optical loss and phase images are fused into expert designed features, leading to a record label-free classification accuracy when combined with deep learning. Image mining techniques are applied, for the first time, to time stretch quantitative phase imaging to measure biophysical attributes including protein concentration, optical loss, and morphological features of single cells at an ultrahigh flow rate and in a label-free fashion. These attributes differ widely28,29,30,31 among cells and their variations reflect important information of genotypes and physiological stimuli32. The multiplexed biophysical features thus lead to information-rich hyper-dimensional representation of the cells for label-free classification with high statistical precision.
We further improved the accuracy, repeatability, and the balance between sensitivity and specificity of our label-free cell classification by a novel machine learning pipeline, which harnesses the advantages of multivariate supervised learning, as well as unique training by evolutionary global optimization of receiver operating characteristics (ROC). To demonstrate sensitivity, specificity, and accuracy of multi-feature label-free flow cytometry using our technique, we classified (1) OT-IIhybridoma T-lymphocytes and SW-480 colon cancer epithelial cells, and (2) Chlamydomonas reinhardtii algal cells (herein referred to as Chlamydomonas) based on their lipid content, which is related to the yield in biofuel production. Our preliminary results show that compared to classification by individual biophysical parameters, our label-free hyperdimensional technique improves the detection accuracy from 77.8% to 95.5%, or in other words, reduces the classification inaccuracy by about five times. ……..
Feature Extraction
The decomposed components of sequential line scans form pairs of spatial maps, namely, optical phase and loss images as shown in Fig. 2 (see Section Methods: Image Reconstruction). These images are used to obtain biophysical fingerprints of the cells8,36. With domain expertise, raw images are fused and transformed into a suitable set of biophysical features, listed in Table 1, which the deep learning model further converts into learned features for improved classification.
The new technique combines two components that were invented at UCLA:
A “photonic time stretch” microscope, which is capable of quickly imaging cells in blood samples. Invented by Barham Jalali, professor and Northrop-Grumman Optoelectronics Chair in electrical engineering, it works by taking pictures of flowing blood cells using laser bursts (similar to how a camera uses a flash). Each flash only lasts nanoseconds (billionths of a second) to avoid damage to cells, but that normally means the images are both too weak to be detected and too fast to be digitized by normal instrumentation. The new microscope overcomes those challenges by using specially designed optics that amplify and boost the clarity of the images, and simultaneously slow them down enough to be detected and digitized at a rate of 36 million images per second.
A deep learning computer program, which identifies cancer cells with more than 95 percent accuracy. Deep learning is a form of artificial intelligence that uses complex algorithms to extract patterns and knowledge from rich multidimenstional datasets, with the goal of achieving accurate decision making.
The study was published in the open-access journal Nature Scientific Reports. The researchers write in the paper that the system could lead to data-driven diagnoses by cells’ physical characteristics, which could allow quicker and earlier diagnoses of cancer, for example, and better understanding of the tumor-specific gene expression in cells, which could facilitate new treatments for disease.
The optical loss images of the cells are affected by the attenuation of multiplexed wavelength components passing through the cells. The attenuation itself is governed by the absorption of the light in cells as well as the scattering from the surface of the cells and from the internal cell organelles. The optical loss image is derived from the low frequency component of the pulse interferograms. The optical phase image is extracted from the analytic form of the high frequency component of the pulse interferograms using Hilbert Transformation, followed by a phase unwrapping algorithm. Details of these derivations can be found in Section Methods. Also, supplementary Videos 1 and 2 show measurements of cell-induced optical path length difference by TS-QPI at four different points along the rainbow for OT-II and SW-480, respectively.
Table 1: List of extracted features.
Feature Name Description Category
Figure 3: Biophysical features formed by image fusion.
(a) Pairwise correlation matrix visualized as a heat map. The map depicts the correlation between all major 16 features extracted from the quantitative images. Diagonal elements of the matrix represent correlation of each parameter with itself, i.e. the autocorrelation. The subsets in box 1, box 2, and box 3 show high correlation because they are mainly related to morphological, optical phase, and optical loss feature categories, respectively. (b) Ranking of biophysical features based on their AUCs in single-feature classification. Blue bars show performance of the morphological parameters, which includes diameter along the interrogation rainbow, diameter along the flow direction, tight cell area, loose cell area, perimeter, circularity, major axis length, orientation, and median radius. As expected, morphology contains most information, but other biophysical features can contribute to improved performance of label-free cell classification. Orange bars show optical phase shift features i.e. optical path length differences and refractive index difference. Green bars show optical loss features representing scattering and absorption by the cell. The best performed feature in these three categories are marked in red.
Figure 4: Machine learning pipeline. Information of quantitative optical phase and loss images are fused to extract multivariate biophysical features of each cell, which are fed into a fully-connected neural network.
The neural network maps input features by a chain of weighted sum and nonlinear activation functions into learned feature space, convenient for classification. This deep neural network is globally trained via area under the curve (AUC) of the receiver operating characteristics (ROC). Each ROC curve corresponds to a set of weights for connections to an output node, generated by scanning the weight of the bias node. The training process maximizes AUC, pushing the ROC curve toward the upper left corner, which means improved sensitivity and specificity in classification.
…. How to cite this article: Chen, C. L. et al. Deep Learning in Label-free Cell Classification.
To better characterize the functional context of genomic variations in cancer, researchers developed a new computer algorithm called REVEALER. [UC San Diego Health]
Scientists at the University of California San Diego School of Medicine and the Broad Institute say they have developed a new computer algorithm—REVEALER—to better characterize the functional context of genomic variations in cancer. The tool, described in a paper (“Characterizing Genomic Alterations in Cancer by Complementary Functional Associations”) published in Nature Biotechnology, is designed to help researchers identify groups of genetic variations that together associate with a particular way cancer cells get activated, or how they respond to certain treatments.
REVEALER is available for free to the global scientific community via the bioinformatics software portal GenePattern.org.
“This computational analysis method effectively uncovers the functional context of genomic alterations, such as gene mutations, amplifications, or deletions, that drive tumor formation,” said senior author Pablo Tamayo, Ph.D., professor and co-director of the UC San Diego Moores Cancer Center Genomics and Computational Biology Shared Resource.
Dr. Tamayo and team tested REVEALER using The Cancer Genome Atlas (TCGA), the NIH’s database of genomic information from more than 500 human tumors representing many cancer types. REVEALER revealed gene alterations associated with the activation of several cellular processes known to play a role in tumor development and response to certain drugs. Some of these gene mutations were already known, but others were new.
For example, the researchers discovered new activating genomic abnormalities for beta-catenin, a cancer-promoting protein, and for the oxidative stress response that some cancers hijack to increase their viability.
REVEALER requires as input high-quality genomic data and a significant number of cancer samples, which can be a challenge, according to Dr. Tamayo. But REVEALER is more sensitive at detecting similarities between different types of genomic features and less dependent on simplifying statistical assumptions, compared to other methods, he adds.
“This study demonstrates the potential of combining functional profiling of cells with the characterizations of cancer genomes via next-generation sequencing,” said co-senior author Jill P. Mesirov, Ph.D., professor and associate vice chancellor for computational health sciences at UC San Diego School of Medicine.
Characterizing genomic alterations in cancer by complementary functional associations
Jong Wook Kim, Olga B Botvinnik, Omar Abudayyeh, Chet Birger, et al.
Systematic efforts to sequence the cancer genome have identified large numbers of mutations and copy number alterations in human cancers. However, elucidating the functional consequences of these variants, and their interactions to drive or maintain oncogenic states, remains a challenge in cancer research. We developed REVEALER, a computational method that identifies combinations of mutually exclusive genomic alterations correlated with functional phenotypes, such as the activation or gene dependency of oncogenic pathways or sensitivity to a drug treatment. We used REVEALER to uncover complementary genomic alterations associated with the transcriptional activation of β-catenin and NRF2, MEK-inhibitor sensitivity, and KRAS dependency. REVEALER successfully identified both known and new associations, demonstrating the power of combining functional profiles with extensive characterization of genomic alterations in cancer genomes
Figure 2: REVEALER results for transcriptional activation of β-catenin in cancer.close
(a) This heatmap illustrates the use of the REVEALER approach to find complementary genomic alterations that match the transcriptional activation of β-catenin in cancer. The target profile is a TCF4 reporter that provides an estimate of…
An imaging-based platform for high-content, quantitative evaluation of therapeutic response in 3D tumour models
Jonathan P. Celli, Imran Rizvi, Adam R. Blanden, Iqbal Massodi, Michael D. Glidden, Brian W. Pogue & Tayyaba Hasan
While it is increasingly recognized that three-dimensional (3D) cell culture models recapitulate drug responses of human cancers with more fidelity than monolayer cultures, a lack of quantitative analysis methods limit their implementation for reliable and routine assessment of emerging therapies. Here, we introduce an approach based on computational analysis of fluorescence image data to provide high-content readouts of dose-dependent cytotoxicity, growth inhibition, treatment-induced architectural changes and size-dependent response in 3D tumour models. We demonstrate this approach in adherent 3D ovarian and pancreatic multiwell extracellular matrix tumour overlays subjected to a panel of clinically relevant cytotoxic modalities and appropriately designed controls for reliable quantification of fluorescence signal. This streamlined methodology reads out the high density of information embedded in 3D culture systems, while maintaining a level of speed and efficiency traditionally achieved with global colorimetric reporters in order to facilitate broader implementation of 3D tumour models in therapeutic screening.
The attrition rates for preclinical development of oncology therapeutics are particularly dismal due to a complex set of factors which includes 1) the failure of pre-clinical models to recapitulate determinants of in vivo treatment response, and 2) the limited ability of available assays to extract treatment-specific data integral to the complexities of therapeutic responses1,2,3. Three-dimensional (3D) tumour models have been shown to restore crucial stromal interactions which are missing in the more commonly used 2D cell culture and that influence tumour organization and architecture4,5,6,7,8, as well as therapeutic response9,10, multicellular resistance (MCR)11,12, drug penetration13,14, hypoxia15,16, and anti-apoptotic signaling17. However, such sophisticated models can only have an impact on therapeutic guidance if they are accompanied by robust quantitative assays, not only for cell viability but also for providing mechanistic insights related to the outcomes. While numerous assays for drug discovery exist18, they are generally not developed for use in 3D systems and are often inherently unsuitable. For example, colorimetric conversion products have been noted to bind to extracellular matrix (ECM)19 and traditional colorimetric cytotoxicity assays reduce treatment response to a single number reflecting a biochemical event that has been equated to cell viability (e.g. tetrazolium salt conversion20). Such approaches fail to provide insight into the spatial patterns of response within colonies, morphological or structural effects of drug response, or how overall culture viability may be obscuring the status of sub-populations that are resistant or partially responsive. Hence, the full benefit of implementing 3D tumour models in therapeutic development has yet to be realized for lack of analytical methods that describe the very aspects of treatment outcome that these systems restore.
Motivated by these factors, we introduce a new platform for quantitative in situ treatment assessment (qVISTA) in 3D tumour models based on computational analysis of information-dense biological image datasets (bioimage-informatics)21,22. This methodology provides software end-users with multiple levels of complexity in output content, from rapidly-interpreted dose response relationships to higher content quantitative insights into treatment-dependent architectural changes, spatial patterns of cytotoxicity within fields of multicellular structures, and statistical analysis of nodule-by-nodule size-dependent viability. The approach introduced here is cognizant of tradeoffs between optical resolution, data sampling (statistics), depth of field, and widespread usability (instrumentation requirement). Specifically, it is optimized for interpretation of fluorescent signals for disease-specific 3D tumour micronodules that are sufficiently small that thousands can be imaged simultaneously with little or no optical bias from widefield integration of signal along the optical axis of each object. At the core of our methodology is the premise that the copious numerical readouts gleaned from segmentation and interpretation of fluorescence signals in these image datasets can be converted into usable information to classify treatment effects comprehensively, without sacrificing the throughput of traditional screening approaches. It is hoped that this comprehensive treatment-assessment methodology will have significant impact in facilitating more sophisticated implementation of 3D cell culture models in preclinical screening by providing a level of content and biological relevance impossible with existing assays in monolayer cell culture in order to focus therapeutic targets and strategies before costly and tedious testing in animal models.
Using two different cell lines and as depicted in Figure 1, we adopt an ECM overlay method pioneered originally for 3D breast cancer models23, and developed in previous studies by us to model micrometastatic ovarian cancer19,24. This system leads to the formation of adherent multicellular 3D acini in approximately the same focal plane atop a laminin-rich ECM bed, implemented here in glass-bottom multiwell imaging plates for automated microscopy. The 3D nodules resultant from restoration of ECM signaling5,8, are heterogeneous in size24, in contrast to other 3D spheroid methods, such as rotary or hanging drop cultures10, in which cells are driven to aggregate into uniformly sized spheroids due to lack of an appropriate substrate to adhere to. Although the latter processes are also biologically relevant, it is the adherent tumour populations characteristic of advanced metastatic disease that are more likely to be managed with medical oncology, which are the focus of therapeutic evaluation herein. The heterogeneity in 3D structures formed via ECM overlay is validated here by endoscopic imaging ofin vivo tumours in orthotopic xenografts derived from the same cells (OVCAR-5).
Figure 1: A simplified schematic flow chart of imaging-based quantitative in situ treatment assessment (qVISTA) in 3D cell culture.
(This figure was prepared in Adobe Illustrator® software by MD Glidden, JP Celli and I Rizvi). A detailed breakdown of the image processing (Step 4) is provided in Supplemental Figure 1.
A critical component of the imaging-based strategy introduced here is the rational tradeoff of image-acquisition parameters for field of view, depth of field and optical resolution, and the development of image processing routines for appropriate removal of background, scaling of fluorescence signals from more than one channel and reliable segmentation of nodules. In order to obtain depth-resolved 3D structures for each nodule at sub-micron lateral resolution using a laser-scanning confocal system, it would require ~ 40 hours (at approximately 100 fields for each well with a 20× objective, times 1 minute/field for a coarse z-stack, times 24 wells) to image a single plate with the same coverage achieved in this study. Even if the resources were available to devote to such time-intensive image acquisition, not to mention the processing, the optical properties of the fluorophores would change during the required time frame for image acquisition, even with environmental controls to maintain culture viability during such extended imaging. The approach developed here, with a mind toward adaptation into high throughput screening, provides a rational balance of speed, requiring less than 30 minutes/plate, and statistical rigour, providing images of thousands of nodules in this time, as required for the high-content analysis developed in this study. These parameters can be further optimized for specific scenarios. For example, we obtain the same number of images in a 96 well plate as for a 24 well plate by acquiring only a single field from each well, rather than 4 stitched fields. This quadruples the number conditions assayed in a single run, at the expense of the number of nodules per condition, and therefore the ability to obtain statistical data sets for size-dependent response, Dfrac and other segmentation-dependent numerical readouts.
We envision that the system for high-content interrogation of therapeutic response in 3D cell culture could have widespread impact in multiple arenas from basic research to large scale drug development campaigns. As such, the treatment assessment methodology presented here does not require extraordinary optical instrumentation or computational resources, making it widely accessible to any research laboratory with an inverted fluorescence microscope and modestly equipped personal computer. And although we have focused here on cancer models, the methodology is broadly applicable to quantitative evaluation of other tissue models in regenerative medicine and tissue engineering. While this analysis toolbox could have impact in facilitating the implementation of in vitro 3D models in preclinical treatment evaluation in smaller academic laboratories, it could also be adopted as part of the screening pipeline in large pharma settings. With the implementation of appropriate temperature controls to handle basement membranes in current robotic liquid handling systems, our analyses could be used in ultra high-throughput screening. In addition to removing non-efficacious potential candidate drugs earlier in the pipeline, this approach could also yield the additional economic advantage of minimizing the use of costly time-intensive animal models through better estimates of dose range, sequence and schedule for combination regimens.
Microscope Uses AI to Find Cancer Cells More Efficiently
Scientists at the California NanoSystems Institute at UCLA have developed a new technique for identifying cancer cells in blood samples faster and more accurately than the current standard methods.
In one common approach to testing for cancer, doctors add biochemicals to blood samples. Those biochemicals attach biological “labels” to the cancer cells, and those labels enable instruments to detect and identify them. However, the biochemicals can damage the cells and render the samples unusable for future analyses.
There are other current techniques that don’t use labeling but can be inaccurate because they identify cancer cells based only on one physical characteristic.
The new technique images cells without destroying them and can identify 16 physical characteristics — including size, granularity and biomass — instead of just one. It combines two components that were invented at UCLA: a photonic time stretch microscope, which is capable of quickly imaging cells in blood samples, and a deep learning computer program that identifies cancer cells with over 95 percent accuracy.
Deep learning is a form of artificial intelligence that uses complex algorithms to extract meaning from data with the goal of achieving accurate decision making.
The study, which was published in the journal Nature Scientific Reports, was led by Barham Jalali, professor and Northrop-Grumman Optoelectronics Chair in electrical engineering; Claire Lifan Chen, a UCLA doctoral student; and Ata Mahjoubfar, a UCLA postdoctoral fellow.
Photonic time stretch was invented by Jalali, and he holds a patent for the technology. The new microscope is just one of many possible applications; it works by taking pictures of flowing blood cells using laser bursts in the way that a camera uses a flash. This process happens so quickly — in nanoseconds, or billionths of a second — that the images would be too weak to be detected and too fast to be digitized by normal instrumentation.
The new microscope overcomes those challenges using specially designed optics that boost the clarity of the images and simultaneously slow them enough to be detected and digitized at a rate of 36 million images per second. It then uses deep learning to distinguish cancer cells from healthy white blood cells.
“Each frame is slowed down in time and optically amplified so it can be digitized,” Mahjoubfar said. “This lets us perform fast cell imaging that the artificial intelligence component can distinguish.”
Normally, taking pictures in such minuscule periods of time would require intense illumination, which could destroy live cells. The UCLA approach also eliminates that problem.
“The photonic time stretch technique allows us to identify rogue cells in a short time with low-level illumination,” Chen said.
The researchers write in the paper that the system could lead to data-driven diagnoses by cells’ physical characteristics, which could allow quicker and earlier diagnoses of cancer, for example, and better understanding of the tumor-specific gene expression in cells, which could facilitate new treatments for disease. ….. see also http://www.nature.com/article-assets/npg/srep/2016/160315/srep21471/images_hires/m685/srep21471-f1.jpg
Advances in indium antimonide (InSb) infrared detectors, detector cooling and system design have resulted in a new generation of application-specific cameras where either SWIR or broadband response is critical.
ELLIOTT RITTENBERG, LEN KAMLET AND ARNOLD ADAMS, IRCAMERAS LLC
http://www.photonics.com/Article.aspx?AID=58415Due to their high thermal sensitivity and favorable atmospheric transmission in the 3- to 5-µm spectral region, cameras based on InSb sensors have been used in traditional applications such as medium- to long-range surveillance, as well as for scientific and research requirements where detection of very small temperature differences or wavelength-specific imaging is beneficial. With recent developments, InSb focal plane arrays (FPAs), when properly passivated, have become a viable option to extend the wavelength response of today’s cameras below 3 µm, into the short-wave infrared (SWIR) spectral region, visible and even ultraviolet wavelength regions. This broadband response provides the ability to solve a variety of challenges with a single instrument, and to build application-specific systems that surpass limitations of other infrared imaging technologies.Three passivated but otherwise uncoated InSb focal plane arrays with spectral responses from <200 to >500 nm. FPAs are in leadless chip carriers. Courtesy of IRCameras.http://www.photonics.com/images/Web/Articles/2016/3/7/Detectors_Plane.jpgDetector advancements
Indium gallium arsenide (InGaAs) sensors are the most common choice for imaging in the SWIR from 900 to 1700 nm due to their high quantum efficiency, ability to operate at or near room temperature, and relatively low production and packaging costs. “Visible InGaAs” sensors, with a response from 400 to 1700 nm, have also proliferated in recent years. InGaAs sensors, however, have been limited to an upper wavelength cutoff at 1700 nm due to difficulties in achieving uniform pixel-to-pixel response in two-dimensional s at longer wavelengths above 1700 nm.
Unfortunately, this leaves these sensors blind in the wavelength region from 1700 to 2500 nm, which contains significant information for both military and civilian applications. Additionally, InGaAs sensors are often subject to image lag or retention when used in very low-light-level applications requiring high-gain states in the sensor.
With advancements in the production of InSb focal planes, infrared cameras today can be equipped with FPAs providing a wide broadband spectral response unparalleled by other production sensors. Advances in the passivation of InSb FPAs eliminate surface states that retain charge when the detector is exposed to higher energy photons, thus removing the possibility of after-images and image lag, which has been an issue for both InSb and InGaAs sensors.
Typically, the quantum efficiency of InSb focal planes prior to the application of an antireflection (AR) coating is about 60% from UV wavelengths to approximately 5000 nm. Improved AR coating processes increase the quantum efficiency to greater than 90% over the wavelength range from 1 to >5 µm. More specialized coatings are available that extend the lower wavelength cut to 400 nm and below, resulting in a single sensor that can be utilized for a broad range of imaging applications.
Finally, the use of digital readout integrated circuits for InSb FPAs results in performance improvements over more traditional analog-based FPAs. Digital InSb FPAs require lower power, eliminate pixel crosstalk, are resistant to blooming and simplify downstream electronics by digitizing data within the FPA instead of requiring an analog-to-digital converter external to the FPA.
InSb sensors with the capabilities
described above are available today in formats ranging from quarter video graphics array (VGA) (320 × 256) to high definition (1280 × 1024), with even higher-resolution (2520 × 2048 and beyond) FPAs just over the horizon.
Cooling InSb sensors
InSb sensors require cryogenic cooling to become photoconductive. Typically, InSb sensors are cooled to about 77 K, though for some applications, including imagers intended for SWIR applications, further cooling of the sensor to less than 70 K is desirable to reduce noise due to dark current.
To cool InSb sensors to the cryogenic temperatures necessary, the focal planes can be integrated into either liquid nitrogen (LN2)-cooled Dewar assemblies, or closed-cycle integrated Dewar cooler assemblies (IDCAs). There are, of course, advantages and drawbacks to each approach.
InSb cameras based on a LN2-cooled Dewar assembly offer the ultimate flexibility for users who may wish to modify the system as application requirements change, or to demonstrate a proof-of-concept before acquiring a closed-cycle cooled system that cannot easily be reconfigured. With an LN2-cooled system, it is possible for the user to easily change cold filters, apertures (ƒ-number) and even the optical back working distance to accommodate a variety of optics, or to define and test the interface to spectrometers and other instruments into which a camera may be integrated. It is also easy to replace FPAs in an LN2-cooled camera to allow upgrades to different format sensors or even different sensor materials.
LN2-cooled systems are also vibration-free, which can be an important consideration when used as an OEM component in a highly sensitive metrology product such as an interferometer. LN2-cooled cameras may also be equipped with motorized multi-position cold filter wheels, allowing the use of spectral filters cooled to cryogenic temperatures to limit their thermal emission and prevent the filter’s emissions from affecting measurements.
Image of Antares/Cygnus liftoff Sept. 18, 2013, using IRC912 mid-wave infrared camera. Courtesy of MARS Scientific. http://www.photonics.com/images/Web/Articles/2016/3/7/Detectors_Liftoff.jpgThe disadvantage of LN2-cooled InSb cameras is that they require replenishment of the LN2, either by filling the Dewar manually, or through the use of a bulky tank and autofill system. For those applications that require cooling below 77 K, the boiling point of LN2, it is possible to lower the temperature of the LN2 by increasing the vacuum above the LN2 reservoir in the Dewar. When doing so, however, the temperature of the LN2 may fluctuate with atmospheric temperature changes or with fluctuating vacuum pressure changes in the liquid nitrogen reservoir, resulting in measurement inconsistencies. For some applications, including astronomical observations, researchers will run InSb sensors at much lower temperatures, below the 63 K freezing point of nitrogen, obviating its use as a cooling medium.
Closed-cycle Stirling coolers provide a method for cooling InSb and other cryogenic cameras designed for continuous operation. These devices remove heat from the detector through the expansion and contraction of helium gas, and are intended for environments where the use of LN2 is not feasible, or for applications where the sensor temperature must be driven below 63 K. Until recently, most cameras so configured incorporated a rotary Stirling cooler, for which manufacturers would historically claim a limited mean time to failure (MTTF) on the order of 8,000 hours. Actual experience, however, has demonstrated this figure to be optimistic. Stirling cooler lifetime is impacted by operating conditions — continuous use in high ambient temperatures and repeated power cycling of the cooler tend to negatively impact MTTF.
Flexure bearing Stirling cooler. Courtesy of Thales Cryogenics. http://www.photonics.com/images/Web/Articles/2016/3/7/Detectors_Cooler.jpgOver the past several years, the introduction and maturation of linear Stirling coolers based on flexure bearing technology has resulted in significant improvements in cooler MTTF. Today, a Stirling cooler with a flexure bearing dual opposed piston compressor routinely provides 25,000 hours of operation. Additionally, linear coolers are significantly less prone to vibration than rotary coolers, making the linear cooler a more desirable choice for precision metrology instruments.
An important consideration for IR cameras based on closed-cycle Stirling coolers, whether rotary or linear, is that once the system is built, it is difficult to change its configuration. The detector, cold shield, cold finger and cold filter are housed in a metal Dewar that is welded closed, and must be cut open in order to make changes or effect repairs. Great care must be taken when cutting open a closed-cycle Dewar to ensure that damage is not inflicted on any of the internal components. Also, with the exception of highly specialized custom built systems, a typical closed-cycle IDCA does not support the use of a cold filter wheel assembly due to the increased cooling capacity required by the cold wheel and increased Dewar volume. This limits the spectral response of the system to that defined by the combination of the sensor and cold filter installed in the Dewar. The result is a purpose-built, continuously operable product that can be used either as a stand-alone product, or as an OEM component for integration into an end-user system.
Today’s infrared imaging systems based on high-performance InSb FPAs offer unmatched sensitivity for a variety of application-specific requirements where either SWIR or broadband spectral response is essential. Understanding available options for sensor resolution and coatings that drive the FPA spectral response; cooling technologies and their associated conveniences; optical filters that define system spectral response; and optical designs that describe how the system interacts with the outside world are critical considerations in the selection of these highly enabled SWIR/MWIR imaging cameras.
Meet the authors
Elliott Rittenberg serves as vice president of sales and marketing at IRCameras LLC in Santa Barbara, Calif.; email: elliott@ircameras.com. Len Kamlet, Ph.D., is product development manager at IRCameras, and has been working exclusively with infrared technology for 14 years; email: len.kamlet@ircameras.com. Arnold Adams, Ph.D., is chief technical officer at IRCameras; email: arn.adams@ircameras.com.
Berkeley Lab captures first high-res 3D images of DNA segments
DNA segments are targeted to be building blocks for molecular computer memory and electronic devices, nanoscale drug-delivery systems, and as markers for biological research and imaging disease-relevant proteins
In a Berkeley Lab-led study, flexible double-helix DNA segments (purple, with green DNA models) connected to gold nanoparticles (yellow) are revealed from the 3D density maps reconstructed from individual samples using a Berkeley Lab-developed technique called individual-particle electron tomography (IPET). Projections of the structures are shown in the green background grid. (credit: Berkeley Lab)
An international research team working at the Lawrence Berkeley National Laboratory (Berkeley Lab) has captured the first high-resolution 3D images of double-helix DNA segments attached at either end to gold nanoparticles — which could act as building blocks for molecular computer memory and electronic devices (see World’s smallest electronic diode made from single DNA molecule), nanoscale drug-delivery systems, and as markers for biological research and for imaging disease-relevant proteins.
The researchers connected coiled DNA strands between polygon-shaped gold nanoparticles and then reconstructed 3D images, using a cutting-edge electron microscope technique coupled with a protein-staining process and sophisticated software that provided structural details at the scale of about 2 nanometers.
“We had no idea about what the double-strand DNA would look like between the gold nanoparticles,” said Gang “Gary” Ren, a Berkeley Lab scientist who led the research. “This is the first time for directly visualizing an individual double-strand DNA segment in 3D,” he said.
The results were published in an open-access paper in the March 30 edition of Nature Communications.
The method developed by this team, called individual-particle electron tomography (IPET), had earlier captured the 3-D structure of a single protein that plays a key role in human cholesterol metabolism. By grabbing 2D images of an object from different angles, the technique allows researchers to assemble a 3D image of that object.
The team has also used the technique to uncover the fluctuation of another well-known flexible protein, human immunoglobulin 1, which plays a role in the human immune system.
https://youtu.be/lQrbmg9ry90 Berkeley Lab | 3-D Reconstructions of Double strand DNA and Gold Nanoparticle Structures
For this new study of DNA nanostructures, Ren used an electron-beam study technique called cryo-electron microscopy (cryo-EM) to examine frozen DNA-nanogold samples, and used IPET to reconstruct 3-D images from samples stained with heavy metal salts. The team also used molecular simulation tools to test the natural shape variations (“conformations”) in the samples, and compared these simulated shapes with observations.
First visualization of DNA strand dynamics without distorting x-ray crystallography
Ren explained that the naturally flexible dynamics of samples, like a man waving his arms, cannot be fully detailed by any method that uses an average of many observations.
A popular way to view the nanoscale structural details of delicate biological samples is to form them into crystals and zap them with X-rays, but that destroys their natural shape, especially fir the DNA-nanogold samples in this study, which the scientists say are incredibly challenging to crystallize. Other common research techniques may require a collection of thousands of near-identical objects, viewed with an electron microscope, to compile a single, averaged 3-D structure. But an averaged 3D image may not adequately show the natural shape fluctuations of a given object.
The samples in the latest experiment were formed from individual polygon gold nanostructures, measuring about 5 nanometers across, connected to single DNA-segment strands with 84 base pairs. Base pairs are basic chemical building blocks that give DNA its structure. Each individual DNA segment and gold nanoparticle naturally zipped together with a partner to form the double-stranded DNA segment with a gold particle at either end.
https://youtu.be/RDOpgj62PLU Berkeley Lab | These views compare the various shape fluctuations obtained from different samples of the same type of double-helix DNA segment (DNA renderings in green, 3D reconstructions in purple) connected to gold nanoparticles (yellow).
The samples were flash-frozen to preserve their structure for study with cryo-EM imaging. The distance between the two gold nanoparticles in individual samples varied from 20 to 30 nanometers, based on different shapes observed in the DNA segments.
Researchers used a cryo-electron microscope at Berkeley Lab’s Molecular Foundry for this study. They collected a series of tilted images of the stained objects, and reconstructed 14 electron-density maps that detailed the structure of individual samples using the IPET technique.
Sub-nanometer images next
Ren said that the next step will be to work to improve the resolution to the sub-nanometer scale.
“Even in this current state we begin to see 3-D structures at 1- to 2-nanometer resolution,” he said. “Through better instrumentation and improved computational algorithms, it would be promising to push the resolution to that visualizing a single DNA helix within an individual protein.”
In future studies, researchers could attempt to improve the imaging resolution for complex structures that incorporate more DNA segments as a sort of “DNA origami,” Ren said. Researchers hope to build and better characterize nanoscale molecular devices using DNA segments that can, for example, store and deliver drugs to targeted areas in the body.
“DNA is easy to program, synthesize and replicate, so it can be used as a special material to quickly self-assemble into nanostructures and to guide the operation of molecular-scale devices,” he said. “Our current study is just a proof of concept for imaging these kinds of molecular devices’ structures.”
The team included researchers at UC Berkeley, the Kavli Energy NanoSciences Institute at Berkeley Lab and UC Berkeley, and Xi’an Jiaotong University in China. This work was supported by the National Science Foundation, DOE Office of Basic Energy Sciences, National Institutes of Health, the National Natural Science Foundation of China, Xi’an Jiaotong University in China, and the Ministry of Science and Technology in China. View more about Gary Ren’s research group here.
Abstract of Three-dimensional structural dynamics and fluctuations of DNA-nanogold conjugates by individual-particle electron tomography
DNA base pairing has been used for many years to direct the arrangement of inorganic nanocrystals into small groupings and arrays with tailored optical and electrical properties. The control of DNA-mediated assembly depends crucially on a better understanding of three-dimensional structure of DNA-nanocrystal-hybridized building blocks. Existing techniques do not allow for structural determination of these flexible and heterogeneous samples. Here we report cryo-electron microscopy and negative-staining electron tomography approaches to image, and three-dimensionally reconstruct a single DNA-nanogold conjugate, an 84-bp double-stranded DNA with two 5-nm nanogold particles for potential substrates in plasmon-coupling experiments. By individual-particle electron tomography reconstruction, we obtain 14 density maps at ~2-nm resolution. Using these maps as constraints, we derive 14 conformations of dsDNA by molecular dynamics simulations. The conformational variation is consistent with that from liquid solution, suggesting that individual-particle electron tomography could be an expected approach to study DNA-assembling and flexible protein structure and dynamics.
By inserting a small “coralyne” molecule into DNA, scientists were able to create a single-molecule diode (connected here by two gold electrodes), which can be used as an active element in future nanoscale circuits. The diode circuit symbol is shown on the left. (credit: University of Georgia and Ben-Gurion University)
Nanoscale electronic components can be made from single DNA molecules, as researchers at the University of Georgia and at Ben-Gurion University in Israel have demonstrated, using a single molecule of DNA to create the world’s smallest diode.
DNA double helix with base pairs (credit: National Human Genome Research Institute)
A diode is a component vital to electronic devices that allows current to flow in one direction but prevents its flow in the other direction. The development could help stimulate development of DNA components for molecular electronics.
As noted in an open-access Nature Chemistry paper published this week, the researchers designed a 11-base-pair (bp) DNA molecule and inserted a small molecule named coralyne into the DNA.*
They found, surprisingly, that this caused the current flowing through the DNA to be 15 times stronger for negative voltages than for positive voltages, a necessary feature of a diode.
Electronic elements 1,00o times smaller than current components
“Our discovery can lead to progress in the design and construction of nanoscale electronic elements that are at least 1,000 times smaller than current components,” says the study’s lead author, Bingqian Xu an associate professor in the UGA College of Engineering and an adjunct professor in chemistry and physics.
The research team plans to enhance the performance of the molecular diode and construct additional molecular devices, which may include a transistor (similar to a two-layer diode, but with one additional layer).
A theoretical model developed by Yanantan Dubi of Ben-Gurion University indicated the diode-like behavior of DNA originates from the bias voltage-induced breaking of spatial symmetry inside the DNA molecule after the coralyne is inserted.
The research is supported by the National Science Foundation.
*“We prepared the DNA–coralyne complex by specifically intercalating two coralyne molecules into a custom-designed 11-base-pair (bp) DNA molecule (5′-CGCGAAACGCG-3′) containing three mismatched A–A base pairs at the centre,” according to the authors.
UPDATE April 6, 2016 to clarify the coralyne intercalation (insertion) into the DNA molecule.
Abstract of Molecular rectifier composed of DNA with high rectification ratio enabled by intercalation
The predictability, diversity and programmability of DNA make it a leading candidate for the design of functional electronic devices that use single molecules, yet its electron transport properties have not been fully elucidated. This is primarily because of a poor understanding of how the structure of DNA determines its electron transport. Here, we demonstrate a DNA-based molecular rectifier constructed by site-specific intercalation of small molecules (coralyne) into a custom-designed 11-base-pair DNA duplex. Measured current–voltage curves of the DNA–coralyne molecular junction show unexpectedly large rectification with a rectification ratio of about 15 at 1.1 V, a counter-intuitive finding considering the seemingly symmetrical molecular structure of the junction. A non-equilibrium Green’s function-based model—parameterized by density functional theory calculations—revealed that the coralyne-induced spatial asymmetry in the electron state distribution caused the observed rectification. This inherent asymmetry leads to changes in the coupling of the molecular HOMO−1 level to the electrodes when an external voltage is applied, resulting in an asymmetric change in transmission.
UNSW researchers say the therapy has enormous potential for treating spinal disc injury and joint and muscle degeneration and could also speed up recovery following complex surgeries where bones and joints need to integrate with the body (credit: UNSW TV)
A stem cell therapy system capable of regenerating any human tissue damaged by injury, disease, or aging could be available within a few years, say University of New South Wales (UNSW Australia) researchers.
Their new repair system*, similar to the method used by salamanders to regenerate limbs, could be used to repair everything from spinal discs to bone fractures, and could transform current treatment approaches to regenerative medicine.
The UNSW-led research was published this week in the Proceedings of the National Academy of Sciences journal.
Reprogramming bone and fat cells
The system reprograms bone and fat cells into induced multipotent stem cells (iMS), which can regenerate multiple tissue types and has been successfully demonstrated in mice, according to study lead author, haematologist, and UNSW Associate Professor John Pimanda.
“This technique is a significant advance on many of the current unproven stem cell therapies, which have shown little or no objective evidence they contribute directly to new tissue formation,” Pimanda said. “We have taken bone and fat cells, switched off their memory and converted them into stem cells so they can repair different cell types once they are put back inside the body.”
“We are currently assessing whether adult human fat cells reprogrammed into iMS cells can safely repair damaged tissue in mice, with human trials expected to begin in late 2017.”
There are different types of stem cells including embryonic stem (ES) cells, which during embryonic development generate every type of cell in the human body, and adult stem cells, which are tissue-specific, but don’t regenerate multiple tissue types. Embryonic stem cells cannot be used to treat damaged tissues because of their tumor forming capacity. The other problem when generating stem cells is the requirement to use viruses to transform cells into stem cells, which is clinically unacceptable, the researchers note.
Research shows that up to 20% of spinal implants either don’t heal or there is delayed healing. The rates are higher for smokers, older people and patients with diseases such diabetes or kidney disease.
Human trials are planned next year once the safety and effectiveness of the technique using human cells in mice has been demonstrated.
* The technique involves extracting adult human fat cells and treating them with the compound 5-Azacytidine (AZA), along with platelet-derived growth factor-AB (PDGF-AB) for about two days. The cells are then treated with the growth factor alone for a further two-three weeks.
AZA is known to induce cell plasticity, which is crucial for reprogramming cells. The AZA compound relaxes the hard-wiring of the cell, which is expanded by the growth factor, transforming the bone and fat cells into iMS cells. When the stem cells are inserted into the damaged tissue site, they multiply, promoting growth and healing.
The new technique is similar to salamander limb regeneration, which is also dependent on the plasticity of differentiated cells, which can repair multiple tissue types, depending on which body part needs replacing.
Along with confirming that human adult fat cells reprogrammed into iMS stem cells can safely repair damaged tissue in mice, the researchers said further work is required to establish whether iMS cells remain dormant at the sites of transplantation and retain their capacity to proliferate on demand.
Abstract of PDGF-AB and 5-Azacytidine induce conversion of somatic cells into tissue-regenerative multipotent stem cells
Current approaches in tissue engineering are geared toward generating tissue-specific stem cells. Given the complexity and heterogeneity of tissues, this approach has its limitations. An alternate approach is to induce terminally differentiated cells to dedifferentiate into multipotent proliferative cells with the capacity to regenerate all components of a damaged tissue, a phenomenon used by salamanders to regenerate limbs. 5-Azacytidine (AZA) is a nucleoside analog that is used to treat preleukemic and leukemic blood disorders. AZA is also known to induce cell plasticity. We hypothesized that AZA-induced cell plasticity occurs via a transient multipotent cell state and that concomitant exposure to a receptive growth factor might result in the expansion of a plastic and proliferative population of cells. To this end, we treated lineage-committed cells with AZA and screened a number of different growth factors with known activity in mesenchyme-derived tissues. Here, we report that transient treatment with AZA in combination with platelet-derived growth factor–AB converts primary somatic cells into tissue-regenerative multipotent stem (iMS) cells. iMS cells possess a distinct transcriptome, are immunosuppressive, and demonstrate long-term self-renewal, serial clonogenicity, and multigerm layer differentiation potential. Importantly, unlike mesenchymal stem cells, iMS cells contribute directly to in vivo tissue regeneration in a context-dependent manner and, unlike embryonic or pluripotent stem cells, do not form teratomas. Taken together, this vector-free method of generating iMS cells from primary terminally differentiated cells has significant scope for application in tissue regeneration.
Because this process works at relatively low temperatures, many transistors can be made on a flexible backing at once. (credit: University of Pennsylvania)
University of Pennsylvania engineers have developed a simplified new approach for making transistors by sequentially depositing their components in the form of liquid nanocrystal “inks.” The new process open the door for transistors and other electronic components to be built into flexible or wearable applications. It also avoids the highly complex current process for creating transistors, which requires high-temperature, high-vacuum equipment. Also, the new lower-temperature process is compatible with a wide array of materials and can be applied to larger areas.
Transistors patterned on plastic backing
The researchers’ nanocrystal-based field effect transistors were patterned onto flexible plastic backings using spin coating, but could eventually be constructed by additive manufacturing systems, like 3D printers.
Kagan’s group developed four nanocrystal inks that comprise the transistor, then deposited them on a flexible backing. (credit: University of Pennsylvania)
The researchers began by dispersing a specific type of nanocrystals in a liquid, creating nanocrystal inks. They developed a library of four of these inks: a conductor (silver), an insulator (aluminum oxide), a semiconductor (cadmium selenide), and a conductor combined with a dopant (a mixture of silver and indium). (“Doping” the semiconductor layer of a transistor with impurities controls whether the device creates a positive or negative charge.)
“These materials are colloids just like the ink in your inkjet printer,” Kagan said, “but you can get all the characteristics that you want and expect from the analogous bulk materials, such as whether they’re conductors, semiconductors or insulators.” Although the electrical properties of several of these nanocrystal inks had been independently verified, they had never been combined into full devices. “Our question was whether you could lay them down on a surface in such a way that they work together to form functional transistors.”
Laying down patterns in layers
Such a process entails layering or mixing them in precise patterns.
First, the conductive silver nanocrystal ink was deposited from liquid on a flexible plastic surface that was treated with a photolithographic mask, then rapidly spun to draw it out in an even layer. The mask was then removed to leave the silver ink in the shape of the transistor’s gate electrode.
The researchers followed that layer by spin-coating a layer of the aluminum oxide nanocrystal-based insulator, then a layer of the cadmium selenide nanocrystal-based semiconductor and finally another masked layer for the indium/silver mixture, which forms the transistor’s source and drain electrodes. Upon heating at relatively low temperatures, the indium dopant diffused from those electrodes into the semiconductor component.
“The trick with working with solution-based materials is making sure that, when you add the second layer, it doesn’t wash off the first, and so on,” Kagan said. “We had to treat the surfaces of the nanocrystals, both when they’re first in solution and after they’re deposited, to make sure they have the right electrical properties and that they stick together in the configuration we want.”
Because this entirely ink-based fabrication process works at lower temperatures than existing vacuum-based methods, the researchers were able to make several transistors on the same flexible plastic backing at the same time.
The inks’ specialized surface chemistry allowed them to stay in configuration without losing their electrical properties. (credit: University of Pennsylvania)
“Making transistors over larger areas and at lower temperatures have been goals for an emerging class of technologies, when people think of the Internet of things, large area flexible electronics and wearable devices,” Kagan said. “We haven’t developed all of the necessary aspects so they could be printed yet, but because these materials are all solution-based, it demonstrates the promise of this materials class and sets the stage for additive manufacturing.”
Because this entirely ink-based fabrication process works at lower temperatures than existing vacuum-based methods, the researchers were able to make several transistors on the same flexible plastic backing at the same time.
3D-printing transistors for wearables
“This is the first work,” Choi said, “showing that all the components, the metallic, insulating, and semiconducting layers of the transistors, and even the doping of the semiconductor, could be made from nanocrystals.”
“Making transistors over larger areas and at lower temperatures have been goals for an emerging class of technologies, when people think of the Internet of things, large area flexible electronics and wearable devices,” Kagan said. “We haven’t developed all of the necessary aspects so they could be printed yet, but because these materials are all solution-based, it demonstrates the promise of this materials class and sets the stage for additive manufacturing.”
The research was supported by the National Science Foundation, the U.S. Department of Energy, the Office of Naval Research, and the Korea Institute of Geoscience and Mineral Resources funded by the Ministry of Science, ICT, and Future Planning of Korea.
Abstract of Exploiting the colloidal nanocrystal library to construct electronic devices
Synthetic methods produce libraries of colloidal nanocrystals with tunable physical properties by tailoring the nanocrystal size, shape, and composition. Here, we exploit colloidal nanocrystal diversity and design the materials, interfaces, and processes to construct all-nanocrystal electronic devices using solution-based processes. Metallic silver and semiconducting cadmium selenide nanocrystals are deposited to form high-conductivity and high-mobility thin-film electrodes and channel layers of field-effect transistors. Insulating aluminum oxide nanocrystals are assembled layer by layer with polyelectrolytes to form high–dielectric constant gate insulator layers for low-voltage device operation. Metallic indium nanocrystals are codispersed with silver nanocrystals to integrate an indium supply in the deposited electrodes that serves to passivate and dope the cadmium selenide nanocrystal channel layer. We fabricate all-nanocrystal field-effect transistors on flexible plastics with electron mobilities of 21.7 square centimeters per volt-second.
All four textile manufacturing processes and corresponding scaffold (structure) types studied exhibited the presence of lipid vacuoles (small red spheres, right column, indicating stem cells undergoing random differentiation), compared to control (left). Electrospun scaffolds (row a) exhibited only a monolayer of lipid vacuoles in a single focal plane, while meltblown, spunbond, and carded scaffolds (rows b, c, d) exhibited vacuoles in multiple planes throughout the fabric thickness. Scale bars: 100 μm (credit: S. A. Tuin et al./Biomedical Materials)
Elizabeth Loboa, dean of the Missouri University College of Engineering, and her team have tested new tissue- engineering methods (based on textile manufacturing) to find ones that are most cost-effective and can be produced in larger quantities.
Tissue engineering is a process that uses novel biomaterials seeded with stem cells to grow and replace missing tissues. When certain types of materials are used, the “scaffolds” that are created to hold stem cells eventually degrade, leaving natural tissue in its place. The new tissues could help patients suffering from wounds caused by diabetes and circulation disorders, patients in need of cartilage or bone repair, and women who have had mastectomies by replacing their breast tissue. The challenge is creating enough of the material on a scale that clinicians need to treat patients.
Electrospinning experiment: nanofibers are collected into an ethanol bath and removed at predefined time intervals (credit: J. M. Coburn et al./The Johns Hopkins University/PNAS)
However, large-scale production with electrospinning is not cost-effective. “Electrospinning produces weak fibers, scaffolds that are not consistent, and pores that are too small,” Loboa said. “The goal of ‘scaling up’ is to produce hundreds of meters of material that look the same, have the same properties, and can be used in clinical settings. So we investigated the processes that create textiles, such as clothing and window furnishings like drapery, to scale up the manufacturing process.”
The group published two papers using three industry-standard, high-throughput manufacturing techniques — meltblowing, spunbonding, and carding — to determine if they would create the materials needed to mimic native tissue.
Meltblowing is a technique during which nonwoven materials are created using a molten polymer to create continuous fibers. Spunbond materials are made much the same way but the fibers are drawn into a web while in a solid state instead of a molten one. Carding involves the separation of fibers through the use of rollers, forming the web needed to hold stem cells in place.
Schematic of gilled fiber multifilament spinning and carded scaffold fabrication (credit: Stephen A. Tuin et al./Acta Biomaterialia)
Cost-effective methods
Loboa and her colleagues tested these techniques to create polylactic acid (PLA) scaffolds (a Food and Drug Administration-approved material used as collagen fillers), seeded with human stem cells. They then spent three weeks studying whether the stem cells remained healthy and if they began to differentiate into fat and bone pathways, which is the goal of using stem cells in a clinical setting when new bone and/or new fat tissue is needed at a defect site. Results showed that the three textile manufacturing methods proved as viable if not more so than electrospinning.
“These alternative methods are more cost-effective than electrospinning,” Loboa said. “A small sample of electrospun material could cost between $2 to $5. The cost for the three manufacturing methods is between $.30 to $3.00; these methods proved to be effective and efficient. Next steps include testing how the different scaffolds created in the three methods perform once implanted in animals.”
Researchers at North Carolina State University and the University of North Carolina at Chapel Hill were also involved in the two studies, which were published in Biomedical Materials (open access) and Acta Biomaterialia. The National Science Foundation, the National Institutes of Health, and the Nonwovens Institute provided funding for the studies.
Abstract of Creating tissues from textiles: scalable nonwoven manufacturing techniques for fabrication of tissue engineering scaffolds
Electrospun nonwovens have been used extensively for tissue engineering applications due to their inherent similarities with respect to fibre size and morphology to that of native extracellular matrix (ECM). However, fabrication of large scaffold constructs is time consuming, may require harsh organic solvents, and often results in mechanical properties inferior to the tissue being treated. In order to translate nonwoven based tissue engineering scaffold strategies to clinical use, a high throughput, repeatable, scalable, and economic manufacturing process is needed. We suggest that nonwoven industry standard high throughput manufacturing techniques (meltblowing, spunbond, and carding) can meet this need. In this study, meltblown, spunbond and carded poly(lactic acid) (PLA) nonwovens were evaluated as tissue engineering scaffolds using human adipose derived stem cells (hASC) and compared to electrospun nonwovens. Scaffolds were seeded with hASC and viability, proliferation, and differentiation were evaluated over the course of 3 weeks. We found that nonwovens manufactured via these industry standard, commercially relevant manufacturing techniques were capable of supporting hASC attachment, proliferation, and both adipogenic and osteogenic differentiation of hASC, making them promising candidates for commercialization and translation of nonwoven scaffold based tissue engineering strategies.
Abstract of Fabrication of novel high surface area mushroom gilled fibers and their effects on human adipose derived stem cells under pulsatile fluid flow for tissue engineering applications
The fabrication and characterization of novel high surface area hollow gilled fiber tissue engineering scaffolds via industrially relevant, scalable, repeatable, high speed, and economical nonwoven carding technology is described. Scaffolds were validated as tissue engineering scaffolds using human adipose derived stem cells (hASC) exposed to pulsatile fluid flow (PFF). The effects of fiber morphology on the proliferation and viability of hASC, as well as effects of varied magnitudes of shear stress applied via PFF on the expression of the early osteogenic gene marker runt related transcription factor 2 (RUNX2) were evaluated. Gilled fiber scaffolds led to a significant increase in proliferation of hASC after seven days in static culture, and exhibited fewer dead cells compared to pure PLA round fiber controls. Further, hASC-seeded scaffolds exposed to 3 and 6 dyn/cm2 resulted in significantly increased mRNA expression of RUNX2 after one hour of PFF in the absence of soluble osteogenic induction factors. This is the first study to describe a method for the fabrication of high surface area gilled fibers and scaffolds. The scalable manufacturing process and potential fabrication across multiple nonwoven and woven platforms makes them promising candidates for a variety of applications that require high surface area fibrous materials.
Statement of Significance
We report here for the first time the successful fabrication of novel high surface area gilled fiber scaffolds for tissue engineering applications. Gilled fibers led to a significant increase in proliferation of human adipose derived stem cells after one week in culture, and a greater number of viable cells compared to round fiber controls. Further, in the absence of osteogenic induction factors, gilled fibers led to significantly increased mRNA expression of an early marker for osteogenesis after exposure to pulsatile fluid flow. This is the first study to describe gilled fiber fabrication and their potential for tissue engineering applications. The repeatable, industrially scalable, and versatile fabrication process makes them promising candidates for a variety of scaffold-based tissue engineering applications.
Update on FDA Policy Regarding 3D Bioprinted Material
Curator: Stephen J. Williams, Ph.D.
Last year (2015) in late October the FDA met to finalize a year long process of drafting guidances for bioprinting human tissue and/or medical devices such as orthopedic devices. This importance of the development of these draft guidances was highlighted in a series of articles below, namely that
there were no standards as a manufacturing process
use of human tissues and materials could have certain unforseen adverse events associated with the bioprinting process
In the last section of this post a recent presentation by the FDA is given as well as an excellent pdf here BioprintingGwinnfinal written by a student at University of Kentucky James Gwinn on regulatory concerns of bioprinting.
Bio-Printing Could Be Banned Or Regulated In Two Years
Cross-section of multi-cellular bioprinted human liver tissue Credit: organovo.com
Bio-printing has been touted as the pinnacle of additive manufacturing and medical science, but what if it might be shut down before it splashes onto the medical scene. Research firm, Gartner Inc believes that the rapid development of bio-printing will spark calls to ban the technology for human and non-human tissue within two years.
A report released by Gartner predicts that the time is drawing near when 3D-bioprinted human organs will be readily available, causing widespread debate. They use an example of 3D printed liver tissue by a San Diego-based company named Organovo.
“At one university, they’re actually using cells from human and non-human organs,” said Pete Basiliere, a Gartner Research Director. “In this example, there was human amniotic fluid, canine smooth muscle cells, and bovine cells all being used. Some may feel those constructs are of concern.”
Bio-printing
Bio-printing uses extruder needles or inkjet-like printers to lay down rows of living cells. Major challenges still face the technology, such as creating vascular structures to support tissue with oxygen and nutrients. Additionally, creating the connective tissue or scaffolding-like structures to support functional tissue is still a barrier that bio-printing will have to overcome.
Organovo has worked around a number of issues and they hope to print a fully functioning liver for pharmaceutical industry by the end of this year. “We have achieved thicknesses of greater than 500 microns, and have maintained liver tissue in a fully functional state with native phenotypic behavior for at least 40 days,” said Mike Renard, Organovo’s executive vice president of commercial operations.
clinical trails and testing of organs could take over a decade in the U.S. This is because of the strict rules the U.S. Food and Drug Administration (FDA) places on any new technology. Bio-printing research could outplace regulatory agencies ability to keep up.
“What’s going to happen, in some respects, is the research going on worldwide is outpacing regulatory agencies ability to keep up,” Basiliere said. “3D bio-printing facilities with the ability to print human organs and tissue will advance far faster than general understanding and acceptance of the ramifications of this technology.”
Other companies have been successful with bio-printing as well. Munich-based EnvisionTEC is already selling a printer called a Bioplotter that sells for $188,000 and can print 3D pieces of human tissue. China’s Hangzhou Dianzi University has developed a printer called Regenovo, which printed a small working kidney that lasted four months.
“These initiatives are well-intentioned, but raise a number of questions that remain unanswered. What happens when complex enhanced organs involving nonhuman cells are made? Who will control the ability to produce them? Who will ensure the quality of the resulting organs?” Basiliere said.
Gartner believes demand for bio-printing will explode in 2015, due to a burgeoning population and insufficient levels of healthcare in emerging markets. “The overall success rates of 3D printing use cases in emerging regions will escalate for three main reasons: the increasing ease of access and commoditization of the technology; ROI; and because it simplifies supply chain issues with getting medical devices to these regions,” Basiliere said. “Other primary drivers are a large population base with inadequate access to healthcare in regions often marred by internal conflicts, wars or terrorism.”
It’s interesting to hear Gartner’s bold predictions for bio-printing. Some of the experts we have talked to seem to think bio-printing is further off than many expect, possibly even 20 or 30 years away for fully functioning organs used in transplants on humans. However, less complicated bio-printing procedures and tissue is only a few years away.
FDA examining regulations for 3‑D printed medical devices
The official purpose of a recent FDA-sponsored workshop was “to provide a forum for FDA, medical device manufacturers, additive manufacturing companies and academia to discuss technical challenges and solutions of 3-D printing.” The FDA wants “input to help it determine technical assessments that should be considered for additively manufactured devices to provide a transparent evaluation process for future submissions.”
Simply put, the FDA is trying to stay current with advanced manufacturing technologies that are revolutionizing patient care and, in some cases, democratizing its availability. When a next-door neighbor can print a medical device in his or her basement, it clearly has many positive and negative implications that need to be considered.
Ignoring the regulatory implications for a moment, the presentations at the workshop were fascinating.
STERIS representative Dr. Bill Brodbeck cautioned that the complex designs and materials now being created with additive manufacturing make sterilization practices challenging. For example, how will the manufacturer know if the implant is sterile or if the agent has been adequately removed? Also, some materials and designs cannot tolerate acids, heat or pressure, making sterilization more difficult.
Dr. Thomas Boland from the University of Texas at El Paso shared his team’s work on 3-D-printed tissues. Using inkjet technology, the researchers are evaluating the variables involved in successfully printing skin. Another bio-printing project being undertaken at Wake Forest by Dr. James Yoo involves constructing bladder-shaped prints using bladder cell biopsies and scaffolding.
Dr. Peter Liacouras at Walter Reed discussed his institution’s practice of using 3-D printing to create surgical guides and custom implants. In another biomedical project, work done at Children’s National Hospital by Drs. Axel Krieger and Laura Olivieri involves the physicians using printed cardiac models to “inform clinical decisions,” i.e. evaluate conditions, plan surgeries and reduce operating time.
As interesting as the presentations were, the subsequent discussions were arguably more important. In an attempt to identify and address all significant impacts of additive manufacturing on medical device production, the subject was organized into preprinting (input), printing (process) and post-printing (output) considerations. Panelists and other stakeholders shared their concerns and viewpoints on each topic in an attempt to inform and persuade FDA decision-makers.
An interesting (but expected) outcome was the relative positions of the various stakeholders. Well-established and large manufacturers proposed validation procedures: material testing, process operating guidelines, quality control, traceability programs, etc. Independent makers argued that this approach would impede, if not eliminate, their ability to provide low-cost prosthetic devices.
Comparing practices to the highly regulated food industry, one can understand and accept the need to adopt similar measures for some additively manufactured medical devices. An implant is going into someone’s body, so the manufacturer needs to evaluate and assure the quality of raw materials, processing procedures and finished product.
But, as in the food industry, this means the producer needs to know the composition of materials. Suppliers cannot hide behind proprietary formulations. If manufacturers are expected to certify that a device is safe, they need to know what ingredients are in the materials they are using.
Many in the industry are also lobbying the FDA to agree that manufacturers should be expected to certify the components and not the additive manufacturing process itself. They argue that what matters is whether the device is safe, not what process was used to make it.
Another distinction should be the product’s risk level. Devices should continue to be classified as I, II or III and that classification, not the process used, should determine its level of regulation.
(3DPrintingChannel) The FDA currently assesses 3D printed medical devices and conventionally made products under the same guidelines, despite the different manufacturing methods involved. To receive device approval, manufacturers must prove that the device is equivalent to a product already on the market for the same use, or the device must undergo the process of attaining pre-market approval. However, the approval process for 3D printed devices could become complicated because the devices are manufactured differently and can be customizable. Two teams at the agency are now trying to determine how approval process should be tweaked to account for the changes.
3D Printing and 3D Bioprinting – Will the FDA Regulate Bioprinting?
This entry was posted by Bill Decker on May 20, 2014 at 8:52 am
The 3d printing revolution came to medicine and is making people happy while scaring them at the same time!
3-D printing—the process of making a solid object of any shape from a digital model—has grown increasingly common in recent years, allowing doctors to craft customized devices like hearing aids, dental implants, and surgical instruments. For example, University of Michigan researchers last year used a 3-D laser printer to create an airway splint out of plastic particles. In another case, a patient had 75% of his skull replaced with a 3-D printed implant customized to fit his head. The 3d printing revolution came to medicine and is making people happy while scaring them at the same time!
Printed hearts? Doctors are getting there
FDA currently treats assesses 3-D printed medical devices and conventionally made products under the same guidelines, despite the different manufacturing methods involved. To receive device approval, manufacturers must prove that the device is equivalent to a product already on the market for the same use, or the device must undergo the process of attaining pre-market approval.
“We evaluate all devices, including any that utilize 3-D printing technology, for safety and effectiveness, and appropriate benefit and risk determination, regardless of the manufacturing technologies used,” FDA spokesperson Susan Laine said.
However, the approval process for 3-D printed devices could become complicated because the devices are manufactured differently and can be customizable. Two teams at the agency now are trying to determine how approval process should be tweaked to account for the changes:
With 3D printers, what used to exist only in the realm of science fiction — who doesn’t remember the Star Trek food replicator that could materialize a drink or meal with the mere press of a button — is now becoming more widely available with food on demand, prosthetic devices, tracheal splints, skull implants, and even liver tissue all having recently been printed, used, implanted or consumed. 3D printing, while exciting, also presents a unique hybrid of technology and biology, making it a potentially unique and difficult area to regulate and oversee. With all of the recent technological advances surround 3D printer technology, the FDA recently announced in a blog post that it too was going 3D, using it to “expand our research efforts and expand our capabilities to review innovative medical products.” In addition, the agency will be investigating how 3D printing technology impacts medical devices and manufacturing processes. This will, in turn, raise the additional question of how such technology — one of the goals of which, at least in the medical world, is to create unique and custom printed devices, tissue and other living organs for use in medical procedures — can be properly evaluated, regulated and monitored.
In medicine, 3D printing is known as “bioprinting,” where so-called bioprinters print cells in liquid or gel format in an attempt to engineer cartilage, bone, skin, blood vessels, and even small pieces of liver and other human tissues [see a recent New York Times article here]. Not to overstate the obvious, but this is truly cutting edge science that could have significant health and safety ramifications for end users. And more importantly for regulatory purposes, such bioprinting does not fit within the traditional category of a “device” or a “biologic.” As was noted in Forbes, “more of the products that FDA is tasked with regulating don’t fit into the traditional categories in which FDA has historically divided its work. Many new medical products transcend boundaries between drugs, devices, and biologics…In such a world, the boundaries between FDA’s different centers may no longer make as much sense.” To that end, Forbes reported that FDA Commissioner Peggy Hamburg announced Friday the formation of a “Program Alignment Group” at the FDA whose goal is to identify and develop plans “to best adapt to the ongoing rapid changes in the regulatory environment, driven by scientific innovation, globalization, the increasing complexity of regulated products, new legal authorities and additional user fee programs.”
It will be interesting to see if the FDA can retool the agency to make it a more flexible, responsive, and function-specific organization. In the short term, the FDA has tasked two laboratories in the Office of Science and Engineering Laboratories with investigating how the new 3D technology can impact the safety and efficacy of devices and materials manufactured using the technology. The Functional Performance and Device Use Laboratory is evaluating “the effect of design changes on the safety and performance of devices when used in different patient populations” while the Laboratory for Solid Mechanics is assessing “how different printing techniques and processes affect the strength and durability of the materials used in medical devices.” Presumably, all of this information will help the FDA evaluate at some point in the future whether a 3D printed heart is safe and effective for use in the patient population.
In any case, this type of hybrid technology can present a risk for companies and manufacturers creating and using such devices. It remains to be seen what sort of regulations will be put in place to determine, for example, what types of clinical trials and information will have to be provided before a 3D printer capable of printing a human heart is approved for use by the FDA. Or even on a different scale, what regulatory hurdles (and on-going monitoring, reporting, and studies) will be required before bioprinted cartilage can be implanted in a patient’s knee. Are food replicators and holodecks far behind?
The US Food and Drug Administration (FDA) plans to soon hold a meeting to discuss the future of regulating medical products made using 3D printing techniques, it has announced.
Background
3D printing is a manufacturing process which layers printed materials on top of one another, creating three-dimensional parts (as opposed to injection molding or routing materials).
The manufacturing method has recently come into vogue with hobbyists, who have been driven by several factors only likely to accelerate in the near future:
The cost of 3D printers has come down considerably.
Electronic files which automate the printing process are shareable over the Internet, allowing anyone with the sufficient raw materials to build a part.
The technology behind 3D printing is becoming more advanced, allowing for the manufacture of increasingly durable parts.
While the technology has some alarming components—the manufacture of untraceable weapons, for example—it’s increasingly being looked at as the future source of medical product innovation, and in particular for medical devices like prosthetics.
Promise and Problems
But while 3D printing holds promise for patients, it poses immense challenges for regulators, who must assess how to—or whether to—regulate the burgeoning sector.
In a recent FDA Voice blog posting, FDA regulators noted that 3D-printed medical devices have already been used in FDA-cleared clinical interventions, and that it expects more devices to emerge in the future.
Already, FDA’s Office of Science and Engineering laboratories are working to investigate how the technology will affect the future of device manufacturing, and CDRH’s Functional Performance and Device Use Laboratory is developing and adapting computer modeling methods to help determine how small design changes could affect the safety of a device. And at the Laboratory for Solid Mechanics, FDA said it is investigating the materials used in the printing process and how those might affect durability and strength of building materials.
And as Focus noted in August 2013, there are myriad regulatory challenges to confront as well. For example: If a 3D printer makes a medical device, will that device be considered adulterated since it was not manufactured under Quality System Regulation-compliant conditions? Would each device be required to be registered with FDA? And would FDA treat shared design files as unauthorized promotion if they failed to make proper note of the device’s benefits and risks? What happens if a device was never cleared or approved by FDA?
The difficulties for FDA are seemingly endless.
Plans for a Guidance Document
But there have been indications that FDA has been thinking about this issue extensively.
In September 2013, Focus first reported that CDRH Director Jeffery Shuren was planning to release a guidance on 3D printing in “less than two years.”
Responding to Focus, Shuren said the guidance would be primarily focused on the “manufacturing side,” and probably on how 3D printing occurs and the materials used rather than some of the loftier questions posed above.
“What you’re making, and how you’re making it, may have implications for how safe and effective that device is,” he said, explaining how various methods of building materials can lead to various weaknesses or problems.
“Those are the kinds of things we’re working through. ‘What are the considerations to take into account?'”
“We’re not looking to get in the way of 3D printing,” Shuren continued, noting the parallel between 3D printing and personalized medicine. “We’d love to see that.”
Guidance Coming ‘Soon’
In recent weeks there have been indications that the guidance could soon see a public release. Plastics News reported that CDRH’s Benita Dair, deputy director of the Division of Chemistry and Materials Science, said the 3D printing guidance would be announced “soon.”
“In terms of 3-D printing, I think we will soon put out a communication to the public about FDA’s thoughts,” Dair said, according to Plastics News. “We hope to help the market bring new devices to patients and bring them to the United States first. And we hope to play an integral part in that.”
Public Meeting
But FDA has now announced that it may be awaiting public input before it puts out that guidance document. In a 16 May 2014 Federal Register announcement, the agency said it will hold a meeting in October 2014 on the “technical considerations of 3D printing.”
“The purpose of this workshop is to provide a forum for FDA, medical device manufacturers, additive manufacturing companies, and academia to discuss technical challenges and solutions of 3-D printing. The Agency would like input regarding technical assessments that should be considered for additively manufactured devices to provide a transparent evaluation process for future submissions.”
That language—”transparent evaluation process for future submissions”—indicates that at least one level, FDA plans to treat 3D printing no differently than any other medical device, subjecting the products to the same rigorous premarket assessments that many devices now undergo.
FDA’s notice seems to focus on industrial applications for the technology—not individual ones. The agency notes that it has already “begun to receive submissions using additive manufacturing for both traditional and patient-matched devices,” and says it sees “many more on the horizon.”
Among FDA’s chief concerns, it said, are process verification and validation, which are both key parts of the medical device quality manufacturing regulations.
But the notice also indicates that existing guidance documents, such as those specific to medical device types, will still be in effect regardless of the 3D printing guidance.
Discussion Points
FDA’s proposed list of discussion topics include:
Preprinting considerations, including but not limited to:
material chemistry
physical properties
recyclability
part reproducibility
process validation
Printing considerations, including but not limited to:
The official purpose of a recent FDA-sponsored workshop was “to provide a forum for FDA, medical device manufacturers, additive manufacturing companies and academia to discuss technical challenges and solutions of 3-D printing.” The FDA wants “input to help it determine technical assessments that should be considered for additively manufactured devices to provide a transparent evaluation process for future submissions.”
Simply put, the FDA is trying to stay current with advanced manufacturing technologies that are revolutionizing patient care and, in some cases, democratizing its availability. When a next-door neighbor can print a medical device in his or her basement, it clearly has many positive and negative implications that need to be considered.
Ignoring the regulatory implications for a moment, the presentations at the workshop were fascinating.
STERIS representative Dr. Bill Brodbeck cautioned that the complex designs and materials now being created with additive manufacturing make sterilization practices challenging. For example, how will the manufacturer know if the implant is sterile or if the agent has been adequately removed? Also, some materials and designs cannot tolerate acids, heat or pressure, making sterilization more difficult.
Dr. Thomas Boland from the University of Texas at El Paso shared his team’s work on 3-D-printed tissues. Using inkjet technology, the researchers are evaluating the variables involved in successfully printing skin. Another bio-printing project being undertaken at Wake Forest by Dr. James Yoo involves constructing bladder-shaped prints using bladder cell biopsies and scaffolding.
Dr. Peter Liacouras at Walter Reed discussed his institution’s practice of using 3-D printing to create surgical guides and custom implants. In another biomedical project, work done at Children’s National Hospital by Drs. Axel Krieger and Laura Olivieri involves the physicians using printed cardiac models to “inform clinical decisions,” i.e. evaluate conditions, plan surgeries and reduce operating time.
As interesting as the presentations were, the subsequent discussions were arguably more important. In an attempt to identify and address all significant impacts of additive manufacturing on medical device production, the subject was organized into preprinting (input), printing (process) and post-printing (output) considerations. Panelists and other stakeholders shared their concerns and viewpoints on each topic in an attempt to inform and persuade FDA decision-makers.
An interesting (but expected) outcome was the relative positions of the various stakeholders. Well-established and large manufacturers proposed validation procedures: material testing, process operating guidelines, quality control, traceability programs, etc. Independent makers argued that this approach would impede, if not eliminate, their ability to provide low-cost prosthetic devices.
Comparing practices to the highly regulated food industry, one can understand and accept the need to adopt similar measures for some additively manufactured medical devices. An implant is going into someone’s body, so the manufacturer needs to evaluate and assure the quality of raw materials, processing procedures and finished product.
But, as in the food industry, this means the producer needs to know the composition of materials. Suppliers cannot hide behind proprietary formulations. If manufacturers are expected to certify that a device is safe, they need to know what ingredients are in the materials they are using.
Many in the industry are also lobbying the FDA to agree that manufacturers should be expected to certify the components and not the additive manufacturing process itself. They argue that what matters is whether the device is safe, not what process was used to make it.
Another distinction should be the product’s risk level. Devices should continue to be classified as I, II or III and that classification, not the process used, should determine its level of regulation.
If you are interested in submitting comments to the FDA on this topic, post them by Nov. 10.
Warfarin and Dabigatran, Similarities and Differences
Author and Curator: Danut Dragoi, PhD
What anticoagulants do?
An anticoagulant helps your body control how fast your blood clots; therefore, it prevents clots from forming inside your arteries, veins or heart during certain medical conditions.
If you have a blood clot, an anticoagulant may prevent the clot from getting larger. It also may prevent a piece of the clot from breaking off and traveling to your lungs, brain or heart. The anticoagulant medication does not dissolve the blood clot. With time, however, this clot may dissolve on its own.
Blood tests you will need
The blood tests for clotting time are called prothrombin time (Protime, PT) and international normalized ratio (INR). These tests help determine if your medication is working. The tests are performed at a laboratory, usually once a week to once a month, as directed by your doctor. Your doctor will help you decide which laboratory you will go to for these tests.
The test results help the doctor decide the dose of warfarin (Coumadin) that you should take to keep a balance between clotting and bleeding.
Important things to keep in mind regarding blood tests include:
Have your INR checked when scheduled.
Go to the same laboratory each time. (There can be a difference in results between laboratories).
If you are planning a trip, talk with your doctor about using another laboratory while traveling.
Dosage
The dose of medication usually ranges from 1 mg to 10 mg once daily. The doctor will prescribe one strength and change the dose as needed (your dose may be adjusted with each INR).
The tablet is scored and breaks in half easily. For example: if your doctor prescribes a 5 mg tablet and then changes the dose to 2.5 mg (2½ mg), which is half the strength, you should break one of the 5 mg tablets in half and take the half-tablet. If you have any questions about your dose, talk with your doctor or pharmacist.
What warfarin (Coumadin) tablets look like
Warfarin is made by several different drug manufacturers and is available in many different shapes. Each color represents a different strength, measured in milligrams (mg). Each tablet has the strength imprinted on one side, and is scored so you can break it in half easily to adjust your dose as your doctor instructed.
Today, on the basis of 4 clinical trials involving over 9,000 patients, PRADAXA is approved to treat blood clots in the veins of your legs(deep vein thrombosis, or DVT) or lungs (pulmonary embolism, or PE)in patients who have been treated with blood thinner injections, and to reduce the risk of them occurring again.
In these trials, PRADAXA was compared to warfarin or to placebo (sugar pills) for the treatment of DVT and PE patients.
Warfarin (NB-which goes by the brand name Coumadin, see link in here) reduces the risk of stroke in patients with atrial fibrillation (NB- atrial fibrillation (also called AFib or AF) is a quivering or irregular heartbeat (arrhythmia) that can lead to blood clots, stroke, heart failure and other heart-related complications. Some people refer to AF as a quivering heart, see link here) but increases the risk of hemorrhage and is difficult to use.
Dabigatran is a new oral direct thrombin inhibitor (NB-direct thrombin inhibitors are a class of medication that act as anticoagulants by directly inhibiting the enzyme thrombin). Some are in clinical use, while others are undergoing clinical development), see link in here.
Some international large clinical trials, see link in here, show results for patients with atrial fibrillation, dabigatran given at a dose of 110 mg was associated with rates of stroke and systemic embolism that were similar to those associated with warfarin, as well as lower rates of major hemorrhage. Dabigatran administered at a dose of 150 mg, as compared with warfarin, was associated with lower rates of stroke and systemic embolism but similar rates of major hemorrhage.
Picture below shows a deep vein thrombosis which is a blood clot that forms inside a vein, usually deep within the leg. About half a million Americans every year get one, and up to 100,000 die because of it. The danger is that part of the clot can break off and travel through your bloodstream. It could get stuck in your lungs and block blood flow, causing organ damage or death, see link in here.
The behaviour of blood thinning drugs is dependent on their physico-chemical properties and since a significant proportion of drugs contain ionisable centers a knowledge of their pKa (NB-pKa was introduced as an index to express the acidity of weak acids, where pKa is defined as follows. For example, the Ka constant for acetic acid (CH3C00H) is 0.0000158 (= 10-4.8), but the pKa constant is 4.8, which is a simpler expression. In addition, the smaller the pKa value, the stronger the acid, see link in here ) is essential, see link in here. The pKa is defined as the negative log of the dissociation constant, see link in here:
pka=-log10(Ka) (1)
where the dissociation constant is defined thus:
Ka=[A][H+]/[AH]
Most drugs have pKa in the range 0-12, and whilst it is possible to calculate pKa it is desirable to experimentally measure the value for representative examples. There are a number of instruments that are capable of measuring pKa utilising Sirius T3 instrument, see link in here .
Table 1 below shows the pka values for warfarin, see link in here and dabigatran, see link in here.
Table 1
==========================
Anticoagulant pka
warfarin 4.99
dabigatran 4.24 11.51*
==========================
* dabigatran possess both acidic and basic functionality.
Both groups are at ionized at blood pH and exist as zwitterionic
Adding physico-chemical features of anticoagulants utilized in “dissolving” blood clots is important for better understanding the de-blocking process within the veins utilizing anticoagulants.
Broadening the productivity spectrum with middleware
Anne Paxton
March 2016—As James Beck, MT(ASCP), remembers it, middleware was introduced at his institution about the same time that the nursing department decided connectivity should be the province of the laboratory.
When the concept of docking and interfacing glucose testing devices came on the scene around the turn of the millennium, that was a turning point, says Beck, who is point-of-care testing coordinator for the University of Pittsburgh Medical Center–St. Margaret, which uses the Telcor middleware solution QML. “What had previously been a nursing-run program turned into a lab-run program just because of their unfamiliarity with the electronics and connectivity issues. Our nursing department, at least, felt, okay, this is the end of line for us—you take it on.”
At that point, middleware was considered revolutionary in its ability just to handle getting point-of-care data to the laboratory information system, never mind to the hospital information system. But between its beginnings a couple of decades ago and today, middleware has charted a more multidimensional role. It is increasingly being called upon as an agile and resourceful productivity manager, software vendors and users say.
It was about 2006–2007 that middleware’s role started to evolve from connectivity to more “productivity opportunities,” Maureen Marentette recalls. Middleware vendor Data Innovations (DI), where Marentette is director of North American sales, started in 1989 with the interfacing of instruments to the LIS as its focus. Now, products like DI’s Instrument Manager take the information that instruments provide plus what the LIS provides, and “they combine those two pieces to provide powerful management of results at a completely different level.”
With drivers available from more than 1,000 different instruments in anatomic pathology, molecular, microbiology, immunoassay, and more, DI has vastly multiplied the number of parameters it can manage, Marentette says. “We’ve got over 473 data fields that we can use to create very specific rules that are patient- and physician-specific as well as specimen-specific.”
For example, if an instrument is giving a bilirubin result, depending on the hospital, it might be a result on a newborn, a 10-year-old child, or an adult. “Whereas most LISs only handle reference ranges in years, using a middleware product you’re able to drill down to a reference range based on number of hours old, because that parameter can be significant for a newborn from a diagnostic perspective.”
Some LIS companies still use DI for their interface engine for point-to-point interfaces from each new instrument a hospital purchases into their LIS. But when a laboratory purchases Instrument Manager, Marentette notes, “if you switch from, say, a Beckman to a Sysmex hematology instrument, you can simply repurpose the connection you had for the first instrument and move it over to the new one, then just do the editing needed to recognize the codes that are unique to each instrument. It’s really just a tweaking rather than having to recreate the entire interface. So we’ve evolved from connectivity to being able to help labs with productivity and optimizing their workflow.”
These capabilities become particularly important for laboratories dealing with staffing constraints. Some Instrument Manager clients, for example, have 95 percent autoverification of their core chemistry lab results, she says, so those tests go straight into the LIS once released, and then into the HIS for interpretation by the physician.
Many labs are still using LIS autoverification protocols only about 40 percent of the time but could be doing much more, Marentette believes. “People kind of look at it over the short term. There is change management that goes into moving into a middleware product initially. It’s a cost thing, and there’s a knowledge piece, and it takes time to implement and get people to understand how much better off they would be with middleware.”
Focused specialization is what Sysmex’s middleware product WAM (Work Area Management) offers for hematology labs, says Anne Tate, MT(ASCP), IT/automation group manager at Sysmex America. “We concentrate on managing everything around the lavender top.” As a best-of-breed solution, WAM provides hematology-specific rules honed over the past 10 years and now used by more than 300 hematology labs serving 1,100 hospital sites, Tate says.
Sixty percent of Sysmex WAM clients are large integrated health networks, many of which use Sysmex instruments such as the DI-60 and its slidemaker/stainers and sorters, as well as Bio-Rad and Cellavision instruments. “We have instrument lines with one or two instruments on them, up to large systems with two automation lines. So we support both complex and the high-end markets.”
Multisite capability is one of WAM’s key features. “It’s a server-based system, so whatever results you have on site A, you’ll be able to see them on site B. So you can get delta checking and previous results and be able to correlate data from one site to another, but also apply the same standard rules. Doctors and other professionals are assured predictability in how the results are looked at and managed between sites,” Tate says.
She considers middleware as the “production floor” of the lab. “We manage all the results coming out of the instruments, applying rules and logic to either autovalidate to the LIS, or hold up results to do something further such as rerun, reflex, review, or maybe move to another Work Area Manager.”
More and more customers are demanding middleware when they purchase automation, Tate points out, and middleware is increasingly becoming an essential part of the lab. “We’ve always been important. But the LIS really depends on us to manage these combinations and multiple sites because that’s where our expertise is.”
Tate believes LISs have to concentrate on other issues. “LIS vendors really can’t know everything about vendors’ automation and all the changes and parts that go with it. LISs are concentrating their dollars on interfacing more with the EMR and medical necessity and other modules, but they depend on us for decision logic to process results for small to large automation.”
Keeping track of operator competency is another WAM capability. “With our management reports, you can search by user and track almost everything users are doing on the system, with competency based on whatever your criteria are.”
What Sysmex customers most like about the WAM middleware is “they don’t have to look at every result. It automates the process so they are assured that all the logic they would have done manually is being done consistently without a verification.”
At Dartmouth-Hitchcock Medical Center in Lebanon, NH, which includes a main hospital, several clinics, and a network of regional lab partners, Sysmex’s WAM is what allowed the laboratory to bring on the clinics and increase its volume in 2009, says Dorothy Martin, MT(ASCP), hematology supervisor.
The facility started with the Sysmex WAM 3.0, which communicates to Dartmouth-Hitchcock’s Cerner LIS and its Epic HIS. After upgrading to WAM 5.0 in November 2015, “we’ve increased our autovalidation to 87.5 percent, so that’s another 2.5 percent that we can autovalidate. Not only that, but we decreased our manual differential rate. Before, it was a little over four percent, then we brought it down to two percent. Because those manual diffs take longer, that’s less tech time we’re spending at the scope.”“Originally we were sending many tests out to a reference lab because we could not handle the volume, but Sysmex allowed us to expand and then bring on the regional partners. This is helping us to drive cost out of the system for our patients.” Her lab also has middleware for urinalysis and is exploring use of middleware in coagulation testing as a 2017 goal.
The 5.0 version also offers manager reports. “Before, in order to know my autoverification rate, I had to use my Cerner system and pull a number of different reports and export that data into Excel spreadsheets. Now, with a click of a button or two, I can mine all of that data out of my WAM system. We can create reports based on rules, based on our turnaround time, based on technologist. So it’s great. It’s quite amazing.”
Martin works with database coordinator Kari Agan who does the “rules building” and integration work with Cerner. “We actually work with Sysmex to create our rules so they do exactly what we want. We want them customized to be based on critical values in our facility.” For example, “I have different platforms in the main lab than we have in the smaller facilities. I have rules for our hematology-oncology population that are different than rules we would have for a patient who comes in for their primary care visit. WAM allows us that flexibility.”
The WAM has been a key contributor to Dartmouth-Hitchcock’s reputation as an innovator, Martin believes. “We’ve been on this Lean Six Sigma journey for a number of years, and I think that middleware and automation have helped us get to those Lean processes.” The distance among the Lebanon, Nashua, and Manchester laboratories has been a non-issue. “All the labs, even though they are as much as 70 miles apart, are tied into the same middleware on the same server and perform the same way. There’s no issue with networking at long distances and it’s done through the Dartmouth-Hitchcock secure network, so all the patient data is protected.”
With the standardization the middleware has allowed among the technologists, and the growth of the network and technical staff that it has facilitated, “we are really proud of how much Sysmex has helped us improve the processes in our lab,” Martin says.
Kerstin Halverson, point-of-care coordinator for Children’s Hospitals and Clinics of Minnesota, uses Telcor’s QML Data Management and Connectivity Solution to interface five instruments and seven manual device types for the roughly 120,000 POC tests per year, encompassing glucose, hemoglobin, rapid strep, pregnancy, blood gases, and urine dipsticks. “We have Sunquest as our LIS and Cerner as our HIS, and everything point of care flows through Telcor from either a device or via Telcor’s offshoot product called WebMRE for manual result entry.”
She handles most every issue involving the connection between point of care, middleware, and the LIS. “I’m the point person when we have to upgrade servers or add devices. I work pretty specifically with the vendors to make sure everything gets established and connected properly.” Over the years, her lab has gone from having only glucoses and blood gases interfaced to having five device types interfaced, and manual testing is electronically entered now too.
Halverson
Children’s Hospitals and Clinics, with its 2,200 credentialed operators, was the first of Telcor’s customers to go live with an e-learning interface between Telcor and the hospital’s e-learning software, PeopleSoft. “As a piece of this interface, tracking attendance at the annual competency fair that all staff must attend is now done electronically. This allows us to monitor point-of-care testing competency efficiently in one spot,” Halverson says.
This project required considerable lead time. “It was not something we switched on one day. It was probably a three-and-a-half-year project with Telcor to get all my POC courses in a table and then on a daily basis, using PeopleSoft, have that information be updated with all the additions, deletions, department changes, course passing and failing, and so on.” The whole process is now automatic, Halverson says, which is useful for dealing with so many operators over 11 different device types.
In her previous job, more than 13 years ago, “we did not have any middleware,” she notes. “At that time, we had glucometers connected through fax machines, not on the hospital’s network. But results weren’t electronically transferred to charts yet; the nurses would just see results on the device and chart them. There was still a lot of discussion about whether to bill for point of care back then, so some institutions probably did not worry about interfacing and getting things electronically recorded the way things have evolved now.” These days, “IT is one of my major focuses. I can’t do point of care without it.”
In her work as POC coordinator, competency takes a huge chunk of her time—for example, keeping records of diplomas for anyone who is doing moderately complex testing. She says Telcor is working on a way to store those records centrally in its database. “After competency, I would say I spend the most time making sure we’re meeting all the regulations on a daily basis, and by that I mean monitoring QC, making sure the instrumentation is up and functioning properly, and troubleshooting issues.”
A change she has greatly appreciated was moving the Telcor software from a single PC at her desk to a server. “I had two different campuses to travel between, and as much as I try to clone myself, I can’t. So having it moved to a server and having access to the server if I’m somewhere else has been a huge jump forward. I can do troubleshooting pretty much anywhere I need to go now.”
Halverson likens POC to a spiderweb with middleware at the center. “For me, the Telcor middleware ends up being the center of the web to help pull everything into one place where I can manage it. It passes everything along to the LIS, then eventually to the HIS, but it helps cut down on a lot of potential problems by locking people out who aren’t trained and keeping the interfaces up. It’s a one-stop shop.”
Today, some POC devices have their own data-management systems and others don’t, but Telcor is able to interface with them either way, says UPMC’s James Beck. “I can’t say that Telcor makes those systems obsolete. But from my perspective, the fewer systems I need to get into, the easier it is to do what I need to do. Which is basically oversee what’s going on in the system, make sure quality checks are being done, and make sure data is moving through the system and getting to the end users. The fewer systems there are for that data, the more likely the data is going to end up on the chart so that clinical decisions can be made on it.”
Most devices, he has found, are interfaceable through Telcor, which is installed in about 1,900 U.S. and Canadian hospitals. UPMC uses Abbott’s Precision Xceed Pro for glucose, and has seen unprecedented growth in the number of tests it conducts using Abbott i-Stat in the past nine months. The connectivity issues have been for the most part already resolved—“we have just been adding a lot more tests to the menu.”
The ABL blood gas analyzers, Beck says, are considered by many to be laboratory instruments, not point-of-care devices, because they are tabletops. But “this is another example of Telcor flexibility, because before, many of the instruments were being run by respiratory therapists and they had a hard time internalizing how to use that lab interface. They didn’t get enough repeat experience with it, the way a lab person would, and the standard lab interface was really failing them. They could not seem to get the hang of it as far as the flow of data.”Siemens’ Clinitek automatically uploads standard urinalysis dipstick and urine hCG testing results to Telcor and in turn to the EMR, Beck says. “So a person who is multitasking does not have to stay there. They can load up a device and walk away and know that the result is going to automatically upload.” Similarly, Accriva’s Hemochron Signature Elite is used for activated clotting time, and Avoximeter 1000E, another Accriva device, is used for co-oximetry testing and cardiac catheterization labs. Both also connect to Telcor.
Beck approached Telcor about taking over the interface, and the company appointed a dedicated analyst to work it out. “Then our next hurdle was actually to convince the LIS folks to allow us to do it. I said, ‘I can make life so much easier for these respiratory therapists if we can send the information through a middleware system like Telcor as opposed to a standard lab interface, which is full of rules and things to remember. I can cut 20-plus steps and mouse clicks and verifications out of their process.’ So it took a lot of convincing but finally they said, okay, go ahead and try it. And that actually got us a patient safety award for our facility improvement process. It was copied at other hospitals within our health system because they had the same dilemmas and we had a proven model that worked.”
There has been downtime, he says, but not due to the middleware. “Recently, our centralized information support actually disconnected our Telcor server so it stopped transmission unexpectedly. It ended up getting reported by our ED where a nurse was doing a urine dip and knew it should have crossed by a certain amount of time.” So temporarily, no data were moving.
But, he says, “The beauty of the computer age is that everything is basically held in a buffer so there’s no data loss. Everything performed during a downtime does get captured ultimately.” However, with one system processing information from 19 hospitals, catching up can take quite a while. “In fact, one of the things we requested was that they expand memory so the catch-up is quicker when there is some backlog of processing that needs to be done.”
Some operators are actually physicians—typically anesthesiologists—and Beck has found there is better appreciation among physicians generally about middleware. “Definitely five or 10 years ago, they could afford to be more distant from middleware awareness, but now they are aware in making sure the data gets to the chart.”
It’s a good thing, too, because point of care is growing. In fact, “we’re seeing explosions in point of care that are often exceeding the ability of one person to oversee.” For this reason, Beck is pleased that his middleware vendor is able to keep pace with the state of technology and even stay one step ahead. Middleware capability like that has become necessary, he says.
The potential customer base for middleware is wide-ranging, says Data Innovations’ Marentette. DI has diverse customers that include the Department of Defense and Department of Veterans Affairs, large academic health science centers, and reference labs. But the company is finding a market as well in smaller and midrange hospitals, which are increasingly looking at DI’s productivity and quality suite modules and its “management through metrics” data mining.
On the quality assurance side, Marentette points to Instrument Manager’s “moving averages” program which helps monitor how instruments are performing between quality control challenges. “If there is a problem when the moving averages go out, you can first of all stop testing and prevent those results from being released to the LIS. Therefore you have fewer edited reports. That really helps on the quality side. And using our specimen management workspace for autoverification, you can put in instructions on what to do when a particular value comes up.”The data mining involves pulling data out of the specimen management database to conduct population or research studies. “We’ve now added our laboratory intelligence, which enables real-time metrics monitoring, so you can look at what’s going on in your lab five minutes ago, identify issues, and drill down to what the challenge is.”
Middleware’s return on investment is twofold, Marentette believes. “It’s not just within the lab itself, but also what it means to the rest of the hospital organization and the ability to enhance the patient experience. For example, quicker turnaround time on tests for the emergency department lets physicians make faster patient diagnostic decisions, bringing better patient experience and contributing to the overall revenue of the organization. Then, within the lab itself, if you are using the autoverification, you can redeploy your staff and take on more work by bringing on new tests or doing some of the centralized testing you maybe sent to another lab before.”
Through its JResultNet software, which DI purchased from Dawning Technologies more than two years ago, the same middleware benefits are available to physician office labs, and Laboratory Production Manager is DI’s parallel product in the European marketplace. DI itself was acquired in 2015 by Roper Technologies, which also owns LIS giant Sunquest Information Systems. But DI is still a standalone company, Marentette emphasizes, noting that a range of LISs (including Epic’s Beaker) as well as IVD manufacturers use Instrument Manager as their connectivity and middleware solution.
Instrument Manager’s usefulness also becomes apparent when hospitals or integrated delivery networks do acquisitions. “We have the ability to interface multiple LISs into the same IM database, so if a hospital acquires another hospital and wants to put things together on the same system, you can do that seamlessly through one of our interfaces. The middleware is able to keep orders and results reporting management straight between systems. So it does make consolidation much simpler and streamlined.”
With the changing configuration of IT in health care, Sysmex’s Anne Tate sees a changing role for middleware ahead. “WAM provides everything the LIS needs to report immediately to the EMR. In some instances, we’re seeing where the WAM can go into clinics or small labs that don’t have an LIS, so eventually we may have to connect the WAM directly to the EMR. We’ve never done it yet, but we’re seeing movement in that respect.” She doesn’t expect it in the hospital environment. “But in standalone clinics, in nontraditional or non-hospital–supported environments, I do see it happening. We’ve already had requests for that.”
Middleware does come with a cost, but customers understand middleware’s value, Tate says. “We don’t get much pushback anymore on the cost of middleware because they see the value and return on investment: They’ll get vastly improved turnaround time, they can reallocate FTEs, and they get standardized results and less error.”
Leading analytical and clinical diagnostics instrumentation OEMs rely on integrated light emission and detection solutions to unlock the mysteries around disease and treatment. Photonic solutions encompassing optical, illumination, sensing and optomechanical technologies provide these OEMs with not only convenience and simplicity, but also an accelerated path to market for the development of highly complex life science and analytical instruments. Even standard, off-the-shelf components, when coupled with a modular and expandable architecture design, can be used to create a fully functional flow cytometer that achieves levels of throughput and cell analysis sensitivity comparable to commercial products.
A good example of an analytical instrument is the flow cytometer, which will be used in this article to highlight the development of the optical heart of the analytical instrument system. Flow cytometers are instruments that measure features of cells in a liquid suspension, characterizing and analyzing cell populations using light scatter and fluorescence parameters. These life science instruments began as research laboratory instruments, and they still play an important role in research into immunology and cell biology. However, they are now also used in the clinical laboratory for diagnosis and monitoring of diseases such as HIV/AIDS and blood cancers. The next phase would be for them to become diagnostic tools used in the clinic or doctor’s office lab, right at the point-of-care.
Laser-induced fluorescence in the heart of the flow cytometer. The laser beam is the violet beam, entering from left, inducing fluorescence in the target flow in the center of the flow channel, entering from top.
Life science and analytical instruments based on optical techniques have moved from university and corporate research laboratories, through central diagnostic and pathology laboratories, to local hospital laboratories. They are now becoming more and more commonplace as diagnostic tools at or close to the point-of-care delivery. Enabling this development is the evolution of optical components and subsystems used within these instruments for the emission, manipulation and detection of light. The result has been increased capability and usability, along with decreased size and cost. For any optical instrument, the beam path can be summarized as follows in Figure 1.
Traditionally, the instrument manufacturer would have had to begin by mastering the photochemical reactions key to the instrument function. The manufacturer would also need to design the complete optical system within the instrument and then manage the sourcing of the necessary components from a large number of other manufacturers. The lasers originally used in the instrument would have been large, powerful argon or krypton ion lasers, needing water cooling and three-phase high-voltage electrical supplies. The beam would have been delivered through a free-space optical system of lenses, filters and microscope objectives to shape the beam and deliver it to the flow cell. Collecting the fluorescent or scattered signal would also have been done using free-space components, including lenses and mirrors, bandpass filters, and complex optomechanical mounts for alignment, allowing the signal to be delivered to the detectors. These detectors would typically have been a photomultiplier tube, a fragile and complex structure of electrodes sealed in glass vacuum tube, requiring over 1000 V to convert the photons entering the detector into a useful electronic output.
Qioptiq iRIS fiber-coupled laser module from Excelitas Technologies Corp. Photo courtesy of Excelitas Technologies Corp.
Today’s instrument designer can now use significantly smaller laser sources, using diode lasers and diode-pumped, solid-state lasers. These lasers can generate the wavelengths used to excite the fluorescence and provide the illumination for the scattering measurements. Usually, each excitation laser source is dedicated to one probe on the sample per run. Running parallel processes by using multiple illumination wavelengths fired in series on the sample can speed up the analysis.
Challenges in flow cytometry
In flow cytometry the primary challenge is to focus the light onto a moving sample in the flow stream, which is usually less than 100 microns wide. To gather meaningful data from a moving target, both the detector and illumination source need to be as stable and stationary as possible; otherwise movement from either one will cause image jitter and reduce resolution. The second physical challenge is positioning the beams from multiple light sources in sufficient proximity to generate parallel illumination spots focused in the flow cell, while sufficiently separating these spots to prevent cross-talk in the detection channels, which may collect emitted and fluorescent light simultaneously.
LynX silicon photomultiplier modules, as used in the flow cytometer demo system. Photo courtesy of Excelitas Technologies Corp.
Flow cytometers have traditionally used an optical system of lenses, filters and microscope objectives to shape the beam and deliver it to the flow cell. The principles are well-understood, although this can often result in a long optical beam train, making it more susceptible to the effects of laser jitter. An open beam path with optics mounted at various locations is also more susceptible to thermal temperature differences that will cause beam movement on the sample. Careful alignment of the optics is necessary as small movements can cause large changes in the beam pointing on the sample. As a result, instruments that use free-space optics can be susceptible to physical knocks and changes in environmental conditions, including heat generated by the laser itself.
The single-mode optical fiber solution
One way to guarantee a stable high-quality beam delivered to the sample without the use of complex optical components is to use single-mode optical fiber. This offers several advantages in functionality, from increased laser stability and image resolution to reduced instrument size and greater ruggedness.
Photo courtesy of Excelitas Technologies Corp.
Coupled directly from the laser to the fiber, the beam is much less sensitive to movement and thermal temperature changes, which reduces the need for instrument alignment service visits and creates a more robust instrument design. The fiber also acts as a spatial filter and eliminates any beam astigmatism to create a near-perfect Gaussian profile. The resulting beam is much more stable over time than those in free-space systems, and it is without accumulated errors from multiple optical interfaces. That helps ensure a more reliable and more stable measurement. Additionally, the use of fiber allows the laser source, which generates heat, to be located away from the flow cell. This means more flexibility for instrument layout, including the option of mounting the laser externally to the instrument head. It also simplifies servicing. That is because the optical alignment of the instrument is fixed and is independent of the laser — so there is no need to perform a lengthy realignment process when the laser is serviced.
Collecting the fluorescent and scattering signals from the sample requires the light to be gathered and collimated prior to using filters to select the wavelength area of interest. Again, this can be done completely in free space, or with the use of fiber, to simplify the optical set-up and to provide a stable delivery route to the detectors.
Photo courtesy of Excelitas Technologies Corp.
For detection of these signals, today’s designers can still use photomultiplier tubes (PMTs) for the ultimate in sensitivity. However, the advent of large area avalanche photodiodes (APDs) and silicon photomultipliers (SiPMs), allows a single detector type to support a wider range of scattered and fluorescent wavelengths. The result is a greater number of dyes can be used. An SiPM is a monolithic array of silicon micro-APDs offering high photon detection efficiencies from the near UV through to red and NIR wavelengths, coupled with relatively low dark counts. With low operating voltages compared to PMTs, as well as greater ruggedness and lower cost due to the solid-state nature of a silicon detector, many flow cytometer vendors are switching away from PMTs as their detection solution.
Simplifying the photonic engine design process
Many new and innovative analytical and diagnostic companies are using optical technologies such as flow cytometry to provide solutions to their customers. The goal of modern optical component suppliers is to make the manufacturer’s life easier by shouldering the burden of the complete photonic engine design, allowing the manufacturer to focus on its core competencies. To that end, demonstrating the ease and expediency of developing life science and analytical instruments from the ground up, Excelitas Technologies and its subsidiary Qioptiq decided to build a fully functional modular flow cytometry prototype. This prototype uses off-the-shelf products from Excelitas’ various divisions.
The design process began with lasers as excitation sources, incorporating two compact solid-state Qioptiq iFLEX-iRIS lasers, with one being blue and the other being violet. The violet laser took advantage of novel violet-excitable fluorescent dyes recently introduced for flow cytometry. The units were coupled with fiber optic delivery systems, allowing them to be positioned some distance from the sample for accessibility and an efficient design. They also simplified the alignment process and made for plug-and-play laser head replacements.
To capture and collimate the scattered and fluorescent light, two Optem High Resolution 20× microscope objectives were used. One was used for forward-scattered light (indicative of cell size) and the other was used for side-scattered light and fluorescence. Sets of longpass and bandpass filters ensured transmission of the appropriate light wavelengths to the detectors. Six high-performance SiPM modules were also used for photon detection, with one being used for the forward-scattered light, and a second for the side-scattered light. The other four detected the spectral emission wavelengths for fluorescence intensity, based on the lasers selected. The modules were SiPMs integrated with a power supply and amplifier, and they were simpler to integrate than conventional photomultiplier tubes or silicon components.
LINOS Microbench parts were used for organizing the optical bench. Anodized aluminum cubes with stainless steel posts, along with mounting plates, adaptors and connectors, were used with a compatible stainless steel rail system to secure small optical components and create the system layout. Microbench parts assemble and disassemble easily, yet they can be stably aligned, making impromptu adjustments to the layout practical. Cubes could be placed next to one another or stacked, simplifying the layout of components on all three axes. The system was also used to mount the fluidics and flow cell needed for a fully functional flow cytometer.
The system alignment was fine-tuned after assembly using fluorescence dyes and calibration particles. To show that the instrument met specifications, standard fluorescence reference particles were run on the completed system. This successful project clearly demonstrated the power of using standard, off-the-shelf components and of a design approach based on a modular and expandable architecture. This fully functional flow cytometer, designed and built from scratch in less than two months, achieved levels of throughput and cell analysis sensitivity comparable to commercial products.
From modularity and flexibility follows customization for reduced size and cost. That is what manufacturers must strive to achieve when striving to provide instruments — and not only cytometers — to users worldwide.
Meet the author
Richard Simons is the senior applications specialist with Excelitas Technologies Corp. in Montreal; e-mail: richard.simons@excelitas.com
A research collaboration involving the University of Würzburg (Würzburg, Germany), the University of Göttingen (Göttingen, Germany), and PicoQuant (Berlin, Germany) has discovered a novel strategy for fluorescence lifetime imaging (FLIM) that allows visualizing nine different target molecules at the same time.
As a fundamental imaging technique in life sciences, FLIM makes it possible to visualize structures and processes in cells by exciting molecules with light and employing the lifetime information of the triggered fluorescence. The newly devised approach, named spectrally resolved FLIM (sFLIM), uses three lasers with different wavelengths pulsing in an alternating pattern for exciting the molecular labels. As these labels exhibit subtle differences in their emission spectra and fluorescence decay patterns, the collected data can be analyzed through a software algorithm, allowing distinction between them with unparalleled precision.
Part of a cell labeled for tubulin with Abberior STAR 635p (green) and giantin with Atto647N (red) utilizing a single stimulated emission depletion (STED) laser wavelength, where the two species have been separated by fast pattern matching. (Sample courtesy of Markus Sauer, University of Würzburg, Germany)
The high sensitivity of sFLIM also allows using the same fluorescent dye to label three different cell structures at once, with the ability to still clearly distinguish them because of slight fluorescence lifetime variations induced by the chemical environment.
We introduce a pattern-matching technique for efficient identification of fluorophore ratios in complex multidimensional fluorescence signals using reference fluorescence decay and spectral signature patterns of individual fluorescent probes. Alternating pulsed laser excitation at three different wavelengths and time-resolved detection on 32 spectrally separated detection channels ensures efficient excitation of fluorophores and a maximum gain of fluorescence information. Using spectrally resolved fluorescence lifetime imaging microscopy (sFLIM), we were able to visualize up to nine different target molecules simultaneously in mouse C2C12 cells. By exploiting the sensitivity of fluorescence emission spectra and the lifetime of organic fluorophores on environmental factors, we carried out fluorescence imaging of three different target molecules in human U2OS cells with the same fluorophore. Our results demonstrate that sFLIM can be used for super-resolution multi-target imaging by stimulated emission depletion (STED).
(a) Total fluorescence intensity image of an U2OS cell labeled with five fluorescent probes. (b–g) Resulting sFLIM composite image (b) showing F-actin stained with ATTO 488 phalloidin (green; c), Golgi stained with primary rabbit antibo…
IR spectroscopy method is promising for clinical test that could detect Alzheimer’s early
Recognizing that current Alzheimer’s disease diagnosis methods only address symptomatic treatment, researchers at Ruhr University Bochum (RUB; Germany) and collaborators at the University of Göttingen and the German Center for Neurogenerative Diseases (DZNE; Göttingen, Germany) have developed a clinical test based on immuno-chemical analysis using infrared (IR)-induced difference spectroscopy that may lead to early detection of Alzheimer’s.
The promising test involves an IR sensor, whose surface is coated with highly specific antibodies that extract biomarkers for Alzheimer’s from the blood or the cerebrospinal fluid taken from the lower part of the back (lumbar liquor). The IR sensor then performs spectroscopic analysis to determine if the biomarkers show already pathological changes, which can take place more than 15 years before any clinical symptoms appear.
“If we wish to have a drug at our disposal that can significantly inhibit the progress of the disease, we need blood tests that detect Alzheimer’s in its pre-dementia stages,” says Prof. Dr. Klaus Gerwert, Head of the Department of Biophysics at RUB. “By applying such drugs at an early stage, we could prevent dementia, or at the very least delay its onset,” adds Prof. Dr. med. Jens Wiltfang, Head of the Department for Psychiatry and Psychotherapy at the University of Göttingen and Clinical Research Coordinator at DZNE Göttingen.
For the test, the secondary structure of beta amyloid (Aß) peptides serves as a biomarker. This structure changes in Alzheimer’s patients. In the misfolded, pathological structure, more and more Aß peptides can accumulate, gradually forming visible plaque deposits in the brain that are typical for Alzheimer’s disease. This, as mentioned previously, happens more than 15 years before first clinical symptoms are present. The pathological Aß plaques can be temporarily detected by positron emission tomography (PET), but this procedure is comparatively expensive and is accompanied by radiation exposure.
The IR sensor developed by the research team that detects misfolding of Aß peptides involves extracting the Aß peptide from body fluids. After initially working with cerebrospinal fluid, the researchers subsequently expanded the method towards blood analysis. “We do not merely select one single possible folding arrangement of the peptide; rather, we detect how all existing Aß secondary structures are distributed, in their healthy and in their pathological forms,” says Gerwert. Precise diagnostics is not possible until the distribution of all secondary structures is evaluated. Tests that analyze Aß peptide are already available with enzyme-linked immunosorbent assays (ELISA). They identify the total concentration, percentage of forms of different length, as well as the concentration of individual conformations in body fluids, but they do not provide information on the diagnostically relevant distribution of the secondary structures at once. “This is why ELISA tests have not been proven very effective when applied in blood sample analysis in practice,” he explains.
The research team used the new method to analyze samples from 141 patients, achieving diagnostic precision of 84% in the blood and 90% in cerebrospinal fluid. The test revealed an increase of misfolded biomarkers as spectral shift of Aß band below threshold, allowing the researchers to determine Alzheimer’s.
An infrared sensor analysing label-free the secondary structure of the Abeta peptide in presence of complex fluids
An immunologic ATR-FTIR sensor for Abeta peptide secondary structure analysis in complex fluids is presented.
The secondary structure change of the Abeta peptide to beta-sheet was proposed as an early event in Alzheimer’s disease. The transition may be used for diagnostics of this disease in an early state. We present an Attenuated Total Reflection (ATR) sensor modified with a specific antibody to extract minute amounts of Abeta peptide out of a complex fluid. Thereby, the Abeta peptide secondary structure was determined in its physiological aqueous environment by FTIR-difference-spectroscopy. The presented results open the door for label-free Alzheimer diagnostics in cerebrospinal fluid or blood. It can be extended to further neurodegenerative diseases.
2012pharmaceutical
sjwilliamspa
