Classification of Knee Arthritis Diseases Using Convolutional Neural Networks

gopi krishna tiruveedula; Berihun  Molla; Teklu  Urgessa; Tsedenya  Debebe; Getaneh  Melese Mare

gopi krishna tiruveedula adama science and technology university
Berihun Molla 4Health Informatics, School of public Health, College of Medicine and Health Science, Arbaminch university , Arbaminch, Ethiopia
Teklu Urgessa Adama Science and Technology University
Tsedenya Debebe Arbaminch university , Arbaminch, Ethiopia
Getaneh Melese Mare Jimma University, Jimma, Oromia, Ethiopia

Abstract

The objective of this research turned into to create a computer-aided diagnosis (CAD) version to classify knee arthritis using deep getting to know strategies. The dataset had gathered of 667 knee X-ray images from both public and private hospitals. The models were evaluated with varying image sizes and color modes. Among the models tested, the proposed custom Knee-Arthritis-Model demonstrated the highest accuracy of 93% when using grayscale images with dimensions of 100x100 pixels. The developed CAD model exhibits significant promise result for accurate knee arthritis classification, offering potential benefits for clinical diagnosis and patient care