Eye Disease Classification Using ResNet-18 Deep Learning Architecture
Gurjot Kaur, Neha Sharma, Rahul Chauhan, Sanjeev Kukreti, Rupesh Gupta
Abstract
The present study aims to investigate the crucial topic of automated categorization of eye diseases using medical photographs by utilizing the capabilities of the ResNet-18 model. The purpose of doing this research derives from the positive outcomes documented in recent studies that have employed ResNet-18 for comparable objectives. This study introduces a customized ResNet-18 model architecture explicitly designed for classifying eye illness images into four categories with significant medicinal properties. The methodology employed in this study utilizes a dataset consisting of 4,217 photos of eye diseases. The model was trained over 30 iterations, using a batch size 128 and default learning rates. The results demonstrate a noteworthy accomplishment, as the ResNet-18 model suggested in this study achieved a commendable accuracy rate of 94%. This highlights the model's efficacy in distinguishing between various eye illnesses, offering substantial enhancements in the precision and effectiveness of diagnostic protocols. The research findings are significant, as they lay the foundation for advancing automated systems that can efficiently and precisely identify eye disorders. This has the potential to bring about a transformative impact on the field of ophthalmic diagnostics and enhance the quality of patient treatment.