Work

Comparison of ResNet-50, EfficientNet-B1, and VGG-16 Algorithms for Cataract Eye Image Classification

Year: 2025

Type: article

Abstract: Cataract is a leading cause of blindness worldwide, emphasizing the need for an effective early detection approach. This study evaluates the capabilities of three widely-used deep learning models—ResNet-50, EfficientNet-B1, and VGG-16—in classifying visual data. The analysis was conducted on a dataset of 2,112 images, comprising 1,074 normal cases and 1,038 cataract cases. The findings reveal that ResNet-50 achieved the best accuracy at 98.61%, followed by EfficientNet-B1 at 96.64% and VGG-16 at 93.82%. In comparison, previous research using Convolutional Neural Network (CNN) techniques reported an accuracy of 92.93%. These results highlight ResNet-50's superior potential for image classification tasks in this domain. This study contributes significantly to the selection of robust models for building an automated cataract detection framework.

Source: Journal of Applied Informatics and Computing

DOI: 10.30871/jaic.v9i2.8968

Authors: Ilham Santoso, Ayub Michaelangelo Manurung, Egia Rosi Subhiyakto

Institution: Universitas Dian Nuswantoro

Language: English

Topic: Retinal Imaging and Analysis

Subfield: Radiology, Nuclear Medicine and Imaging

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $35