Utilizing deep learning for the diagnosis of eye disorders and localization of affected areas
College of Dentistry, University of Al-Ameed, 56001 Karbala, Iraq.
Research Article
Open Access Research Journal of Science and Technology, 2024, 11(02), 145-150.
Article DOI: 10.53022/oarjst.2024.11.2.0101
Publication history:
Received on 23 April 2024; revised on 2 June 2024; accepted on 4 June 2024
Abstract:
A deep learning-based system for identifying afflicted areas in eye images and diagnosing eye illnesses is offered. Convolutional neural networks (CNNs) are one of the used methods to routinely classify eye disorders from optical coherence tomography (OCT) pictures and retinal scans including, glaucoma, diabetic retinopathy and macular degeneration. 93.5% accuracy on the test set was achieved by optimizing using one of the optimizers called Adam optimizer with definite cross-entropy loss. Assessment indicators including F1-score, accuracy, and recall offer additional indication of the model's flexibility. Next the system is placed into use in a medical experimental, where it helps medical specialists by showing impacted areas using division heatmaps and giving analyses in real time. This method shows that deep learning can support with both precisely diagnosing eye problems and analytical exactly which is obliging for specialists making decisions
Keywords:
Convolution Neural Networks (Cnns); Optical Coherence Tomography (OCT); Artificial Intelligence (AI); Age-Related Macular Degeneration (AMD)
Full text article in PDF:
Copyright information:
Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0