Deep CNN Based Multi-Class Retinal Disease Detection with Reduced Memory Overhead

A.Jahnavi; S.Yashwanth Kumar; E.Sophia Raj; M.Tejaswi; S.Krishnakumar

doi:10.71097/IJSAT.v16.i1.3086

Deep CNN Based Multi-Class Retinal Disease Detection with Reduced Memory Overhead

Author(s)	A.Jahnavi, S.Yashwanth Kumar, E.Sophia Raj, M.Tejaswi, S.Krishnakumar
Country	India
Abstract	Extensive research directions for important disease diagnosis and classification includes ANN, Deep learning, RNN, Alex Net, and ResNet. Retinal disease classification has been transformed by CNN and U-Net Segmentation. The intricacy of feature extraction causes U-Net to pass the entire feature map to the decoder, which causes errors related to memory and CPU. It stops pooling index reuse when combined with the unsampled decoder feature map. For problems involving multiple classes of data, this study presents a convolutional neural network (CNN) model that makes optimal use of memory. The suggested model was evaluated using an Eye Net benchmark dataset that contained 32 different forms of retinal illnesses. The proposed model enhances memory management and precision, according to the experimental results. In order to compare accuracy, precision, and recall, various epochs and step timings were utilized. The suggested approach performed well on Eye-net.
Keywords	Classification, CNN, deep learning, Eye-Net, retina, U-Net
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 16, Issue 1, January-March 2025
Published On	2025-03-31
Cite This	Deep CNN Based Multi-Class Retinal Disease Detection with Reduced Memory Overhead - A.Jahnavi, S.Yashwanth Kumar, E.Sophia Raj, M.Tejaswi, S.Krishnakumar - IJSAT Volume 16, Issue 1, January-March 2025. DOI 10.71097/IJSAT.v16.i1.3086
DOI	https://doi.org/10.71097/IJSAT.v16.i1.3086
Short DOI	https://doi.org/g9dgpv

View / Download PDF File

About IJSAT Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijsat.org

International Journal on Science and Technology

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Deep CNN Based Multi-Class Retinal Disease Detection with Reduced Memory Overhead

Share this