Implementation of Support Vector Machine Method in Digital Image Classification for Eye Disease Detection

Abstract

Eye diseases such as diabetic retinopathy, glaucoma, cataracts, and age-related macular degeneration (AMD) are among the leading causes of vision impairment and blindness worldwide. Early detection plays a crucial role in preventing severe vision loss, yet traditional diagnostic methods often rely on manual assessments by ophthalmologists, which can be time-consuming, subjective, and inaccessible in resource-limited areas. To address this challenge, this study explores the implementation of the Support Vector Machine (SVM) method in digital image classification for eye disease detection. The research focuses on optimizing SVM by integrating advanced feature extraction techniques, including Histogram of Oriented Gradients (HOG), Principal Component Analysis (PCA), and Wavelet Transforms, to enhance classification accuracy. A dataset of retinal images will be processed, and different SVM kernel functions (linear, polynomial, and radial basis function) will be evaluated to determine the most effective model configuration. The study also compares SVM’s performance with other machine learning algorithms such as Random Forest, K-Nearest Neighbors (KNN), and Convolutional Neural Networks (CNNs) to assess its efficiency and reliability. Experimental results will be analyzed using performance metrics such as accuracy, precision, recall, and F1-score to determine the model’s effectiveness. The findings will provide insights into SVM’s potential as a computationally efficient alternative to deep learning approaches for early eye disease detection. This research contributes to the advancement of AI-driven medical diagnostics by demonstrating the viability of SVM for eye disease classification. The results could help develop cost-effective and scalable diagnostic solutions, improving early detection rates, reducing preventable blindness, and enhancing healthcare accessibility in underserved communities.