Automatic Detection and Classification of Diabetic Retinopathy from Optical Coherence Tomography Angiography Images using Deep Learning - A Review
Main Article Content
Abstract
Diabetes-related retinopathy (DR) is a microvascular complication of diabetes that affects many people (DR). Both the prevalence of DR cases and the number of diabetics worldwide have risen considerably. DR weakens vision and can lead to full or partial blindness if ignored. Ocular visual serves as a crucial instrument in detecting, assessing, handling, and recording DR. Over the last few years, advances in technological imaging have enabled the creation of images with improved readability and comparison while requiring significantly less time, effort, and disruption. The choriocapillaris and retinal microvasculature can be studied using optical coherence imaging the procedure is based on differences in floating in blood cells (OCTA). Additionally, OCTA can find preclinical microvascular anomalies that appear before clinically evident DR symptoms. To identify and categorize diabetic retinopathy in OCTAimages, deep learning algorithms may be developed. By using DL techniques, such as CNNs, the images can be analyzed and features extracted for accurate diagnosis.In this paper, we want to offer readers a review of Diabetic retinopathy classification techniques utilizing machine learning and deep learning from OCTAimages, as well as a summary of the various OCTA datasets currently in use for this purpose. Also, we confirmed that the proposed model could work by comparing its performance to that of anML-based categorization tool that uses manually created features extracted from OCTA images. This review gave an overview of recent studies on DL-based image analysis models for OCTA images. It also addressed possible problems with clinical deployment and future directions for research.
Article Details
How to Cite
[1]
A. M.A and S Sridevi Sathya Priya, “Automatic Detection and Classification of Diabetic Retinopathy from Optical Coherence Tomography Angiography Images using Deep Learning - A Review”, AJSE, vol. 23, no. 3, pp. 277 - 297, Jan. 2025.
Section
Articles
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
AJSE contents are under the terms of the Creative Commons Attribution License. This permits anyone to copy, distribute, transmit and adapt the worknon-commercially provided the original work and source is appropriately cited.