Deep Learning-Based Glaucoma Detection Using Cropped Optic Disc Regions

Author(s)	Dr. Chandrakala V Patil, Ms. Amruta P Vardhamane, Ms. Shivalingamma Katti, Ms. Shweta More
Country	India
Abstract	Glaucoma, a leading cause of irreversible blindness, often goes undetected due to its asymptomatic early stages. Traditional diagnosis methods are costly and require specialists. This study proposes an automated deep learning-based glaucoma detection framework using retinal fundus images. A private dataset of 634 annotated images was used to train models like EfficientNet-B3, MobileNet, DenseNet, and GoogLeNet on cropped optic disc and cup regions. EfficientNet-B3 achieved the highest performance with 96.52% accuracy. Additionally, blood vessel segmentation using a U-Net model enabled MobileNetV3 to reach 83.48% accuracy. These results highlight the potential of AI for accessible and efficient glaucoma screening, especially in low-resource settings.
Keywords	EfficientNet, MobileNet, Glaucoma Detection, Fundus Imaging, Deep Learning, Image Classification, U-Net, Image Segmentation
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 7, Issue 2, March-April 2025
Published On	2025-04-30
DOI	https://doi.org/10.36948/ijfmr.2025.v07i02.43042
Short DOI	https://doi.org/g9g727

About IJFMR 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		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research