How AI Improves Cancer Diagnosis

For many years, biopsy—which requires tissue extraction for analysis—was the only effective way to diagnose malignant illnesses. This does not, however, give a complete image of the organ tissue. Digital scans of a particular region that may be impacted by cell mutations are a key component of contemporary histopathology techniques. Pathologists can look at considerably bigger portions of human beings at once using entire slide pictures or WSI.

For the purpose of determining prognostic and therapeutic targets as well as for cancer diagnosis, histopathology is regarded as the gold standard. The likelihood of survival is greatly increased when cancer is discovered early. Unfortunately, pathological analysis is a challenging, time-consuming process that necessitates in-depth understanding. Pathologists differed on a diagnosis on average 24.7% of the time, according to a study that looked at breast biopsy concordance among pathologists. The need for developing computer-aided tools to assist pathologists in histology is highlighted by the high prevalence of misdiagnosis. 

The enormous resolution of the image makes using WSI seem difficult. Even though WSI scans are quite illuminating, it takes hours of meticulous zooming in and out and scrolling from region to area before scrutiny yields the answer. As a result, AI programs that process WSI using convolutional neural networks and computer vision have begun to appear. By drawing attention to the area of concern where probable cancer cells may be found, this method aids medical experts by intensifying the diagnostic process.

Members:

Reference:

Advancements in Healthcare Technology – Benefits for Today’s Healthcare Students. (2019, May 1). Retrieved November 1, 2022, from https://americancareercollege.edu/pulse/health-e-news/advancements-in-healthcare-technology-benefits-for-todays-healthcare-students.html

Khened, M., Kori, A., Rajkumar, H., Krishnamurthi, G., & Srinivasan, B. (2021, June 2). A generalized deep learning framework for whole-slide image segmentation and analysis – Scientific Reports. Retrieved November 1, 2022, from https://www.nature.com/articles/s41598-021-90444-8