• Arungiri Raja U
• Harivathsa S
• Irshath Azeem M
• Dhanush J

Knee osteoarthritis (KOA) is a widespread degenerative joint condition that severely impacts quality of life, with early detection being essential for effective management. This study presents a comprehensive deep learning-based approach for the automated prediction and classification of KOA using X-ray images. Utilizing the Kellgren and Lawrence (KL) grading system, KOA severity is categorized into five levels: Normal, Doubtful, Mild, Moderate, and Severe. The project employed two phases: first, by aggregating four datasets and applying a deep convolutional generative adversarial network (DCGAN) for data augmentation, resulting in a large dataset of 110,232 enhanced X-ray images using advanced preprocessing methods like adaptive histogram equalization and fast non-local means; second, by implementing and comparing various deep learning models. A novel modified compact convolutional transformer model, KOA-CCTNet, was developed and optimized for efficient training on large-scale data. KOA-CCTNet outperformed existing models including Swin Transformer, Vision Transformer, and Involutional Neural Network, achieving a test accuracy of 94.58%, surpassing ResNet50 (80.77%), MobileNetV2 (79.98%), DenseNet201 (80.23%), InceptionV3 (76.89%), and VGG16 (79.58%). Additionally, a separate system using Flask, Python, HTML, CSS, and JavaScript was built to classify 1,650 manually annotated high-quality grayscale X-ray images collected from medical facilities. MobileNetV2 and VGG16 were employed in this system, achieving 96% and 92% test accuracy, respectively. The overall results affirm the robustness and clinical viability of deep learning for accurate, scalable, and cost-effective KOA diagnosis and grading.

Osteoarthritis Detection, Knee Osteoarthritis, KOA Classification, X-ray Image Analysis, Deep Learning, Convolutional Neural Network, Image Preprocessing, MobileNetV2, VGG16, Automated Diagnosis, Clinical Decision Support, Medical Image Classification, Artificial Intelligence in Healthcare, Diagnostic Imaging System

"AUTOMATED KNEE OSTEOARTHRITIS CLASSIFICATION FROM X-RAY IMAGES USING DEEP LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.b702-b707, May-2025, Available :http://www.jetir.org/papers/JETIR2505181.pdf

"AUTOMATED KNEE OSTEOARTHRITIS CLASSIFICATION FROM X-RAY IMAGES USING DEEP LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppb702-b707, May-2025, Available at : http://www.jetir.org/papers/JETIR2505181.pdf