Improving diabetic retinopathy classification using transfer learning and optimized deep learning models

Dai, Cheng Xiao (2024) Improving diabetic retinopathy classification using transfer learning and optimized deep learning models. Final Year Project, UTAR.

Preview

PDF Download (3136Kb) | Preview

Abstract

Diabetic retinopathy (DR) is an eye disease closely related to diabetes that may lead to severe vision loss or even blindness if not diagnosed and treated in time. With the increasing number of diabetic patients, DR has gradually become a global public health problem. Against this background, it is essential to develop a method that can accurately and efficiently diagnose DR. With the continuous advancement of Artificial Intelligence (AI) technology, deep learning and transfer learning have become powerful tools for modern medical research, especially in medical image analysis. This research project aims to combine transfer learning and deep learning techniques to solve this challenge. The main focus is to address the problem of insufficient labelled image datasets in the medical field through transfer learning, exploring how to use limited labelled data to train high-performance models effectively. It will also delve into the use of optimized deep learning models to improve the classification accuracy of DR. It is expected to provide a more accurate and efficient tool for the early diagnosis and treatment of DR, thus helping to decrease vision loss and related complications caused by DR.

Actions (login required)