DR. HIMANSHU KASHYAP

DR. RAJIV RAMAN, DR. RAMYA R NADIG

Abstract

The diabetes mellitus (DM) epidemic is a significant public health concern. With the increasing diabetes trend worldwide, morbidity, mortality, and associated costs of diabetes-related complications are a global public health concern. The prevalence of DR in India is 12% to 22.4%. Screening of DR involves grading the retinal images using trained graders/ophthalmologists or the use of AI algorithms which can detect the presence of DR or sight-threatening DR.

The purpose of the study was to develop a predictive risk score model based on deep learning to diagnose DR in the Indian population. We intend to build a model independent of fundus photography, totally reliant on systemic data through targeted screening from population-based data from South India.