Microsatellite instability in gastrointestinal cancer using deep learning: A review

Abstract

Currently, the health management is a foremost concern all over the world. Particularly in India, due to lack of physical activity, overweight, stress and unhealthy eating habits, various diseases like heart disease, diabetes and cancer disease are increasing at a very rapid rate. The common problem which is affecting all age generation is a gastrointestinal disorder which includes various conditions like colitis, colon polyps, perianal infection and cancer. Microsatellite Instability (MSI) in gastrointestinal (GI) cancer is the second leading cause of death worldwide. However, this cancer can frequently be asymptomatic during the early stages and stay undetected until the later stages of tumor advancement. Deep Learning has the capacity to detect MSI in gastrointestinal cancer at an early stage, which is very helpful for the patients. There are number of research papers, focusing on different techniques used for detection of MSI in gastrointestinal caner. In this paper, we focus on the recent trends of deep learning methods in this field. The main goal of this review paper is to provide a detailed discussion from technological perspective. This article discusses various challenges, molecular subtypes, pathologist approach and computer aided approach for gastrointestinal cancer detection. Also, this article brings into light the existing literature and state of art with their contribution in different aspects of detection of microsatellite instability in gastrointestinal tract with the help of deep learning techniques.