Advertisements
Feeds:
Posts
Comments

Archive for the ‘Electrophysiology’ Category


Multiple Barriers Identified Which May Hamper Use of Artificial Intelligence in the Clinical Setting

Reporter: Stephen J. Williams, PhD.

From the Journal Science:Science  21 Jun 2019: Vol. 364, Issue 6446, pp. 1119-1120

By Jennifer Couzin-Frankel

 

In a commentary article from Jennifer Couzin-Frankel entitled “Medicine contends with how to use artificial intelligence  the barriers to the efficient and reliable adoption of artificial intelligence and machine learning in the hospital setting are discussed.   In summary these barriers result from lack of reproducibility across hospitals. For instance, a major concern among radiologists is the AI software being developed to read images in order to magnify small changes, such as with cardiac images, is developed within one hospital and may not reflect the equipment or standard practices used in other hospital systems.  To address this issue, lust recently, US scientists and government regulators issued guidance describing how to convert research-based AI into improved medical images and published these guidance in the Journal of the American College of Radiology.  The group suggested greater collaboration among relevant parties in developing of AI practices, including software engineers, scientists, clinicians, radiologists etc. 

As thousands of images are fed into AI algorithms, according to neurosurgeon Eric Oermann at Mount Sinai Hospital, the signals they recognize can have less to do with disease than with other patient characteristics, the brand of MRI machine, or even how a scanner is angled.  For example Oermann and Mount Sinai developed an AI algorithm to detect spots on a lung scan indicative of pneumonia and when tested in a group of new patients the algorithm could detect pneumonia with 93% accuracy.  

However when the group from Sinai tested their algorithm from tens of thousands of scans from other hospitals including NIH success rate fell to 73-80%, indicative of bias within the training set: in other words there was something unique about the way Mt. Sinai does their scans relative to other hospitals.  Indeed, many of the patients Mt. Sinai sees are too sick to get out of bed and radiologists would use portable scanners, which generate different images than stand alone scanners.  

The results were published in Plos Medicine as seen below:

PLoS Med. 2018 Nov 6;15(11):e1002683. doi: 10.1371/journal.pmed.1002683. eCollection 2018 Nov.

Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study.

Zech JR1, Badgeley MA2, Liu M2, Costa AB3, Titano JJ4, Oermann EK3.

Abstract

BACKGROUND:

There is interest in using convolutional neural networks (CNNs) to analyze medical imaging to provide computer-aided diagnosis (CAD). Recent work has suggested that image classification CNNs may not generalize to new data as well as previously believed. We assessed how well CNNs generalized across three hospital systems for a simulated pneumonia screening task.

METHODS AND FINDINGS:

A cross-sectional design with multiple model training cohorts was used to evaluate model generalizability to external sites using split-sample validation. A total of 158,323 chest radiographs were drawn from three institutions: National Institutes of Health Clinical Center (NIH; 112,120 from 30,805 patients), Mount Sinai Hospital (MSH; 42,396 from 12,904 patients), and Indiana University Network for Patient Care (IU; 3,807 from 3,683 patients). These patient populations had an age mean (SD) of 46.9 years (16.6), 63.2 years (16.5), and 49.6 years (17) with a female percentage of 43.5%, 44.8%, and 57.3%, respectively. We assessed individual models using the area under the receiver operating characteristic curve (AUC) for radiographic findings consistent with pneumonia and compared performance on different test sets with DeLong’s test. The prevalence of pneumonia was high enough at MSH (34.2%) relative to NIH and IU (1.2% and 1.0%) that merely sorting by hospital system achieved an AUC of 0.861 (95% CI 0.855-0.866) on the joint MSH-NIH dataset. Models trained on data from either NIH or MSH had equivalent performance on IU (P values 0.580 and 0.273, respectively) and inferior performance on data from each other relative to an internal test set (i.e., new data from within the hospital system used for training data; P values both <0.001). The highest internal performance was achieved by combining training and test data from MSH and NIH (AUC 0.931, 95% CI 0.927-0.936), but this model demonstrated significantly lower external performance at IU (AUC 0.815, 95% CI 0.745-0.885, P = 0.001). To test the effect of pooling data from sites with disparate pneumonia prevalence, we used stratified subsampling to generate MSH-NIH cohorts that only differed in disease prevalence between training data sites. When both training data sites had the same pneumonia prevalence, the model performed consistently on external IU data (P = 0.88). When a 10-fold difference in pneumonia rate was introduced between sites, internal test performance improved compared to the balanced model (10× MSH risk P < 0.001; 10× NIH P = 0.002), but this outperformance failed to generalize to IU (MSH 10× P < 0.001; NIH 10× P = 0.027). CNNs were able to directly detect hospital system of a radiograph for 99.95% NIH (22,050/22,062) and 99.98% MSH (8,386/8,388) radiographs. The primary limitation of our approach and the available public data is that we cannot fully assess what other factors might be contributing to hospital system-specific biases.

CONCLUSION:

Pneumonia-screening CNNs achieved better internal than external performance in 3 out of 5 natural comparisons. When models were trained on pooled data from sites with different pneumonia prevalence, they performed better on new pooled data from these sites but not on external data. CNNs robustly identified hospital system and department within a hospital, which can have large differences in disease burden and may confound predictions.

PMID: 30399157 PMCID: PMC6219764 DOI: 10.1371/journal.pmed.1002683

[Indexed for MEDLINE] Free PMC Article

Images from this publication.See all images (3)Free text

 

 

Surprisingly, not many researchers have begun to use data obtained from different hospitals.  The FDA has issued some guidance in the matter but considers “locked” AI software or unchanging software as a medical device.  However they just announced development of a framework for regulating more cutting edge software that continues to learn over time.

Still the key point is that collaboration over multiple health systems in various countries may be necessary for development of AI software which is used in multiple clinical settings.  Otherwise each hospital will need to develop their own software only used on their own system and would provide a regulatory headache for the FDA.

 

Other articles on Artificial Intelligence in Clinical Medicine on this Open Access Journal include:

Top 12 Artificial Intelligence Innovations Disrupting Healthcare by 2020

The launch of SCAI – Interview with Gérard Biau, director of the Sorbonne Center for Artificial Intelligence (SCAI).

Real Time Coverage @BIOConvention #BIO2019: Machine Learning and Artificial Intelligence #AI: Realizing Precision Medicine One Patient at a Time

50 Contemporary Artificial Intelligence Leading Experts and Researchers

 

Advertisements

Read Full Post »


scPopCorn: A New Computational Method for Subpopulation Detection and their Comparative Analysis Across Single-Cell Experiments

Reporter and Curator: Dr. Sudipta Saha, Ph.D.

 

Present day technological advances have facilitated unprecedented opportunities for studying biological systems at single-cell level resolution. For example, single-cell RNA sequencing (scRNA-seq) enables the measurement of transcriptomic information of thousands of individual cells in one experiment. Analyses of such data provide information that was not accessible using bulk sequencing, which can only assess average properties of cell populations. Single-cell measurements, however, can capture the heterogeneity of a population of cells. In particular, single-cell studies allow for the identification of novel cell types, states, and dynamics.

 

One of the most prominent uses of the scRNA-seq technology is the identification of subpopulations of cells present in a sample and comparing such subpopulations across samples. Such information is crucial for understanding the heterogeneity of cells in a sample and for comparative analysis of samples from different conditions, tissues, and species. A frequently used approach is to cluster every dataset separately, inspect marker genes for each cluster, and compare these clusters in an attempt to determine which cell types were shared between samples. This approach, however, relies on the existence of predefined or clearly identifiable marker genes and their consistent measurement across subpopulations.

 

Although the aligned data can then be clustered to reveal subpopulations and their correspondence, solving the subpopulation-mapping problem by performing global alignment first and clustering second overlooks the original information about subpopulations existing in each experiment. In contrast, an approach addressing this problem directly might represent a more suitable solution. So, keeping this in mind the researchers developed a computational method, single-cell subpopulations comparison (scPopCorn), that allows for comparative analysis of two or more single-cell populations.

 

The performance of scPopCorn was tested in three distinct settings. First, its potential was demonstrated in identifying and aligning subpopulations from single-cell data from human and mouse pancreatic single-cell data. Next, scPopCorn was applied to the task of aligning biological replicates of mouse kidney single-cell data. scPopCorn achieved the best performance over the previously published tools. Finally, it was applied to compare populations of cells from cancer and healthy brain tissues, revealing the relation of neoplastic cells to neural cells and astrocytes. Consequently, as a result of this integrative approach, scPopCorn provides a powerful tool for comparative analysis of single-cell populations.

 

This scPopCorn is basically a computational method for the identification of subpopulations of cells present within individual single-cell experiments and mapping of these subpopulations across these experiments. Different from other approaches, scPopCorn performs the tasks of population identification and mapping simultaneously by optimizing a function that combines both objectives. When applied to complex biological data, scPopCorn outperforms previous methods. However, it should be kept in mind that scPopCorn assumes the input single-cell data to consist of separable subpopulations and it is not designed to perform a comparative analysis of single cell trajectories datasets that do not fulfill this constraint.

 

Several innovations developed in this work contributed to the performance of scPopCorn. First, unifying the above-mentioned tasks into a single problem statement allowed for integrating the signal from different experiments while identifying subpopulations within each experiment. Such an incorporation aids the reduction of biological and experimental noise. The researchers believe that the ideas introduced in scPopCorn not only enabled the design of a highly accurate identification of subpopulations and mapping approach, but can also provide a stepping stone for other tools to interrogate the relationships between single cell experiments.

 

References:

 

https://www.sciencedirect.com/science/article/pii/S2405471219301887

 

https://www.tandfonline.com/doi/abs/10.1080/23307706.2017.1397554

 

https://ieeexplore.ieee.org/abstract/document/4031383

 

https://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0927-y

 

https://www.sciencedirect.com/science/article/pii/S2405471216302666

 

 

Read Full Post »


@ClevelandClinic – Cardiac Consult: Catheter Ablation vs Antiarrhythmic Drug Therapy in Atrial Fibrillation: CABANA – What Did We Learn?

Reporter: Aviva Lev-Ari, PhD, RN

 

AUDIT PODCAST

https://my.clevelandclinic.org/podcasts/cardiac-consult/catheter-ablation-vs-antiarrhythmic-drug-therapy-in-atrial-fibrillation-cabana?_ga=2.88658141.711601484.1558922695-amp-RRJ7UwWd4zu5JL6IeLrcYA

 

The international CABANA trial (Catheter Ablation versus Arrhythmia Drug Therapy for Atrial Fibrillation) was the biggest buzz at the Heart Rhythm Society Scientific Sessions earlier this year, and it’s still making waves several months later.

Cleveland Clinic is among the 120 centers participating in the trial, and electrophysiologist Bruce Lindsay, MD, is the site’s principal investigator for the study. He recently sat down with Oussama Wazni, MD, Cleveland Clinic’s Section Head of Cardiac Electrophysiology and Pacing, to discuss the CABANA trial’s findings and implications. Below is an edited transcript of their conversation.

The problem was this: About 9 percent of the patients who were supposed to get ablations never did, and it’s not clear why. The reasons could have been financial issues or patients merely changing their mind or perhaps being too sick. If it was the latter reason, that would of course bias the results. But the problem is we don’t know.

On the other side, a substantial number of patients assigned to drug therapy — 27.5 percent — crossed over and received ablation. That rate of crossover was a bit higher than anticipated.

It’s difficult to use an intention-to-treat analysis when there’s a large crossover and a lot of people don’t get the treatment they were supposed to get. Nonetheless, the study design specified an intention-to-treat analysis, which found no significant differences between the groups in the composite primary end point or any of its components. There were, however, significant reductions in hospitalization for cardiovascular problems and in time to atrial fibrillation recurrence in the ablation group, and the latter finding is consistent with results from past studies.

Because of the large number of crossovers, there was much interest in the as-treated analysis, which was prespecified as a sensitivity analysis of the primary results.

  • This analysis showed a 3.9 percent absolute risk reduction — and
  • a 27 percent relative reduction — in the primary end point with ablation versus drug therapy.
  • That was a statistically significant effect, as was the 3.1 percent absolute reduction in all-cause death with ablation versus drug therapy.

SOURCE

https://consultqd.clevelandclinic.org/ablation-vs-medical-therapy-for-atrial-fibrillation-putting-cabana-in-perspective/?utm_campaign=qd%20tweets&utm_medium=social&utm_source=twitter&utm_content=180920%20ablation%20fibrillation&cvosrc=social%20network.twitter.qd%20tweets&cvo_creative=180920%20ablation%20fibrillation

Read Full Post »


Clever experiment: GWAS of 500 time points in an EKG – The genetic makeup of the electrocardiogram

Reporter: Aviva Lev-Ari, PhD, RN

The genetic makeup of the electrocardiogram

Niek VerweijJan-Walter BenjaminsMichael P. MorleyYordi van de VegteAlexander TeumerTeresa TrenkwalderWibke ReinhardThomas P. CappolaPim van der Harst

Abstract

Since its original description in 1893 by Willem van Einthoven, the electrocardiogram (ECG) has been instrumental in the recognition of a wide array of cardiac disorders1,2. Although many electrocardiographic patterns have been well described, the underlying biology is incompletely understood. Genetic associations of particular features of the ECG have been identified by genome wide studies. This snapshot approach only provides fragmented information of the underlying genetic makeup of the ECG. Here, we follow the effects of individual genetic variants through the complete cardiac cycle the ECG represents. We found that genetic variants have unique morphological signatures not identified by previous analyses. By exploiting identified abberations of these morphological signatures, we show that novel genetic loci can be identified for cardiac disorders. Our results demonstrate how an integrated approach to analyse high-dimensional data can further our understanding of the ECG, adding to the earlier undertaken snapshot analyses of individual ECG components. We anticipate that our comprehensive resource will fuel in silico explorations of the biological mechanisms underlying cardiac traits and disorders represented on the ECG. For example, known disease causing variants can be used to identify novel morphological ECG signatures, which in turn can be utilized to prioritize genetic variants or genes for functional validation. Furthermore, the ECG plays a major role in the development of drugs, a genetic assessment of the entire ECG can drive such developments.

SOURCE

https://www.biorxiv.org/content/10.1101/648527v1

made available under a CC-BY-ND 4.0 International license.

Read Full Post »


Lesson 8 Cell Signaling and Motility: Lesson and Supplemental Information on Cell Junctions and ECM: #TUBiol3373

Curator: Stephen J. Williams, Ph.D.

Please click on the following link for the PowerPoint Presentation for Lecture 8 on Cell Junctions and the  Extracellular Matrix: (this is same lesson from 2018 so don’t worry that file says 2018)

cell signaling 8 lesson 2018

 

Some other reading on this lesson on this Open Access Journal Include:

On Cell Junctions:

Translational Research on the Mechanism of Water and Electrolyte Movements into the Cell     

(pay particular attention to article by Fischbarg on importance of tight junctions for proper water and electrolyte movement)

The Role of Tight Junction Proteins in Water and Electrolyte Transport

(pay attention to article of role of tight junction in kidney in the Loop of Henle and the collecting tubule)

EpCAM [7.4]

(a tight junction protein)

Signaling and Signaling Pathways

(for this lesson pay attention to the part that shows how Receptor Tyrosine Kinase activation (RTK) can lead to signaling to an integrin and also how the thrombin receptor leads to cellular signals both to GPCR (G-protein coupled receptors like the thrombin receptor, the ADP receptor; but also the signaling cascades that lead to integrin activation of integrins leading to adhesion to insoluble fibrin mesh of the newly formed clot and subsequent adhesion of platelets, forming the platelet plug during thrombosis.)

On the Extracellular Matrix

Three-Dimensional Fibroblast Matrix Improves Left Ventricular Function Post MI

Arteriogenesis and Cardiac Repair: Two Biomaterials – Injectable Thymosin beta4 and Myocardial Matrix Hydrogel

 

Read Full Post »


VIDEO: Editor’s Choice of the Most Innovative New Cardiac Technology at AHA 2018

Read Full Post »


Heart Murmur Detection done by AI Algorithm (Eko Core and Eko Duo) Devices Outperform most Auscultatory Skills of Cardiologists

Reporter: Aviva Lev-Ari, PhD, RN

 

AI Algorithm Outperforms Most Cardiologists in Heart Murmur Detection

Eko’s heart murmur detection algorithm outperformed four out of five cardiologists in recent clinical study

“Artificial Intelligence Detects Pediatric Heart Murmurs With Cardiologist-Level Accuracy,” the study demonstrates the power of machine learning and artificial intelligence (AI) to enhance cardiac care.

The neural network AI algorithm was trained on thousands of heart sound recordings. The algorithm was then tested on an independent dataset of pediatric heart sounds and compared to gold-standard echocardiogram imagery. Five pediatric cardiologists also listened to the heart sound recordings and independently made a determination whether a recording contained a murmur. This advancement will help narrow the clinical skill gap between the 27,000 cardiologists in the U.S. — the experts at murmur detection — and the 3.8 million other clinicians who are less experienced in the identification of heart murmurs through a stethoscope.

A study published in the Journal of the American Medical Association revealed that, on average, internal medicine and family practice physician residents misdiagnose 80 percent of common cardiac events.1 Cardiologists on the other hand, can effectively diagnose 90 percent of cardiac events using a stethoscope.2

Eko’s murmur screening algorithm, when coupled with the company’s U.S. Food and Drug Administration (FDA)-cleared Eko Core and Eko Duo devices, will enable any and all clinicians to more accurately screen for heart murmurs.

Eko is currently pursuing FDA clearance for the algorithm and will be rolling it out with its existing cardiac monitoring devices upon securing regulatory clearance.

For more information: http://www.ekohealth.com

References

1. Mangione S., Nieman L.Z. Cardiac auscultatory skills of internal medicine and family practice trainees. A comparison of diagnostic proficiency. Journal of the American Medical Association, Sept. 3, 1997. doi:10.1001/jama.1997.03550090041030

2. Thompson W.R. In defence of auscultation: a glorious future? Heart Asia, Feb. 1, 2017. doi:  [10.1136/heartasia-2016-010796]

 

SOURCE

https://www.dicardiology.com/content/ai-algorithm-outperforms-most-cardiologists-heart-murmur-detection?eid=333021707&bid=2308309

Read Full Post »

Older Posts »