Deep Learning Techniques for Prediction and Diagnosis of Diabetes Mellitus
Someswar Pal, Navneet Mishra, Megha Bhushan, Pankaj Singh Kholiya, Meghavi Rana, Arun Negi
Abstract
Diabetes Mellitus (DM) is a serious chronic disease that affects billions of people worldwide. 8.5% of the population suffers from diabetes between the age range of18 to 70 years. In recent years, with the advancement of early diabetes detection processes the probability of dying from diabetes has decreased by 18% between the age group from 30 to 70 years in the time span of 2000 to 2016. Diabetes predicted in early stages can prevent many health issues such as damage of kidneys, eyes, ulcer, heart, lungs, and blood vessels. There are many existing processes to detect diabetes at an early stage, and more work needs to be done to detect and cure diabetes. This paper provides a review on the existing deep learning techniques used to predict diabetes in the early stage. It can assist researchers working in this area by providing them insight on the latest techniques for the diagnosis of diabetes at an early stage.