Feeds:
Posts
Comments

Posts Tagged ‘CT scan’

Using A.I. to Detect Lung Cancer gets an A!

Reporter: Irina Robu, PhD

3.3.19

3.3.19   Using A.I. to Detect Lung Cancer gets an A!, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 2: CRISPR for Gene Editing and DNA Repair

Google researchers hypothesized that computers are as good or better than doctors at detecting tiny lung cancers on CT scans, since CT scan combines data from several X-rays to produce a detailed image of a structure inside the body. CT scans produce 2-dimensional images of a slice of the body and the data can also be used to construct 3-D images.

However, the technology published in Nature Medicine offers input in the future of artificial intelligence in medicine. By feeding vast amounts of data from medical imaging into systems called artificial neural networks, scientists can teach computers to identify patterns linked to a specific condition, like pneumonia, cancer or a wrist fracture that would be hard for a person to see. The system trails an algorithm, or set of instructions, and learns as it goes. The more data it receives, the better it becomes at interpretation.

The process, known as deep learning enables computers to identify objects and understand speech but it also created systems to help pathologists read microscope slides to diagnose cancer, and to help ophthalmologists detect eye disease in people with diabetes. In their recent study, the scientist used artificial intelligence to CT scans used to screen people for lung cancer, which caused 160,000 deaths in the United States last year, and 1.7 million worldwide. The scans are recommended for people at high risk because of a long history of smoking.

Screening studies showed that it can reduce the risk of dying from lung cancer and can also identify spots that might later become cancer, so that radiologists can categorize patients into risk groups and decide whether they need biopsies or more frequent follow-up scans to keep track of the suspect regions.

However, the test has errors. It can miss tumors or mistake benign spots for malignancies and shove patients into invasive, risky procedures like lung biopsies or surgery.

SOURCE

https://www.nytimes.com/2019/05/20/health/cancer-artificial-intelligence-ct-scans.html

Other related articles were published in this Online Scientific Open Access Journal including the following:

https://pharmaceuticalintelligence.com/2019/07/21/multiple-barriers-identified-which-may-hamper-use-of-artificial-intelligence-in-the-clinical-setting/

https://pharmaceuticalintelligence.com/2019/06/28/ai-system-used-to-detect-lung-cancer/

Read Full Post »

Reporter: Ritu Saxena, Ph.D.

With the number of cancer cases plummeting every year, there is a dire need for finding a cure to wipe the disease out. A number of therapeutic drugs are currently in use, however, due to heterogeneity of the disease targeted therapy is required. An important criteria that needs to be addressed in this context is the –‘tumor response’ and how it could be predicted, thereby improving the selection of patients for cancer treatment. The issue of tumor response has been addressed in a recent editorial titled “Tumor response criteria: are they appropriate?” published recently in Future Oncology.

The article talks about how the early tumor treatment response methods came into practice and how we need to redefine and reassess the tumor response.

Defining ‘tumor response’ has always been a challenge

WHO defines a response to anticancer therapy as 50% or more reduction in the tumor size measured in two perpendicular diameters. It is based on the results of experiments performed by Moertel and Hanley in 1976 and later published by Miller et al in 1981. Twenty years later, in the year 2000, the US National Cancer Institute, with the European Association for Research and Treatment of Cancer, proposed ‘new response criteria’ for solid tumors; a replacement of 2D measurement with measurement of one dimen­sion was made. Tumor response was defined as a decrease in the largest tumor diameter by 30%, which would translate into a 50% decrease for a spherical lesion. However, response criteria have not been updated after that and there a structured standardization of treatment response is still required especially when several studies have revealed that the response of tumors to a therapy via imaging results from conventional approaches such as endoscopy, CT scan, is not reliable. The reason is that evaluating the size of tumor is just one part of the story and to get the complete picture inves­tigating and evaluating the tissue is essential to differentiate between treatment-related scar, fibrosis or micro­scopic residual tumor.

In clinical practice, treatment response is determined on the basis of well-established parameters obtained from diagnostic imaging, both cross-sectional and functional. In general, the response is classified as:

  • Complete remission: If a tumor disappears after a particular therapy,
  • Partial remission: there is residual tumor after therapy.

For a doctor examining the morphology of the tumor, complete remission might seem like good news, however, mission might not be complete yet! For example, in some cases, with regard to prognosis, patients with 0% residual tumor (complete tumor response) had the same prognosis com­pared with those patients with 1–10% residual tumor (subtotal response).

Another example is that in patients demonstrating complete remission of tumor response as observed with clinical, sonographic, functional (PET) and histopathological analysis experience recur­rence within the first 2 years of resection.

Adding complexity to the situation is the fact that the appropriate, clinically relevant timing of assess­ment of tumor response to treatment remains undefined. An example mentioned in the editorial is – for gastrointestinal (GI) malignancies, the assessment timing varies considerably from 3 to 6 weeks after initia­tion of neoadjuvant external beam radiation. Further, time could vary depending upon the type of radiation administered, i.e., if it is external beam, accelerated hyperfractionation, or brachytherapy.

Abovementioned examples remind us of the intricacy and enigma of tumor biol­ogy and subsequent tumor response.

Conclusion

Owing to the extraordinary het­erogeneity of cancers between patients, and pri­mary and metastatic tumors in the same patients, it is important to consider several factors while determining the response of tumors to different therapie in clinical trials. Authors exclaim, “We must change the tools we use to assess tumor response. The new modality should be based on individualized histopathology as well as tumor molecular, genetic and functional characteristics, and individual patients’ charac­teristics.”

Future perspective

Editorial points out that the oncologists, radiotherapists, and immunologists all might have a different opinion and observation as far as tumor response is considered. For example, surgical oncologists might determine a treatment to be effective if the local tumor control is much better after multimodal treatment, and that patients post-therapeutically also reveal an increase of the rate of microscopic and macroscopic R0-resection. Immunologists, on the other hand, might just declare a response if immune-competent cells have been decreased and, possibly, without clinical signs of decrease of tumor size.

What might be the answer to the complexity to reading tumor response is stated in the editorial – “an interdisciplinary initiative with all key stake­holders and disciplines represented is imperative to make predictive and prognostic individualized tumor response assessment a modern-day reality. The integrated multidisciplinary panel of international experts need to define how to leverage existing data, tissue and testing platforms in order to predict individual patient treatment response and prog­nosis.”

Sources:

Editorial : Björn LDM Brücher et al Tumor response criteria: are they appropriate? Future Oncology August 2012, Vol. 8, No. 8, 903-906.

Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer 1981, 47(1),207–214.

Related articles to this subject on this Open Access Online Scientific Journal:

See comment written for :

Knowing the tumor’s size and location, could we target treatment to THE ROI by applying

http://pharmaceuticalintelligence.com/2012/10/16/knowing-the-tumors-size-and-location-could-we-target-treatment-to-the-roi-by-applying-imaging-guided-intervention/imaging-guided intervention?

Personalized Medicine: Cancer Cell Biology and Minimally Invasive Surgery (MIS)

http://pharmaceuticalintelligence.com/2012/12/01/personalized-medicine-cancer-cell-biology-and-minimally-invasive-surgery-mis/

Read Full Post »

%d bloggers like this: