Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{173971,
        author = {Sanskar Masirkar and Prof . R. B. Khule and Shrikrishna Bawankule},
        title = {KNEE OSTEOARTHRITIS SEVRITY GRADING USING DEEP LEARNING MODELS},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {10},
        pages = {2333-2341},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=173971},
        abstract = {Osteoarthritis (OA) of the knee is a prevalent joint condition that impacts movement & quality of life, especially in older people. Early detection & accurate severity grading help in better treatment & improved patient outcomes. This study compares deep learning models to automate knee OA severity grading using X-ray images. The dataset included labeled knee OA X-ray images, which were preprocessed through resizing, normalization, & data augmentation to improve model accuracy & generalization. Seven advanced deep learning models—CNN, DenseNet121, ResNet50, Xception, VGG16, VGG19, MobileNet, and AlexNet—were trained & evaluated. Every model was adjusted for optimal gaining knowledge rates, batch sizes, and dropout rates. Their performance was measured using F1 score, recall, accuracy, precision, and confusion matrices. Among these models, DenseNet121 performed the best, Reaching 98.55% accuracy due to its efficient feature reuse. ResNet50 followed with 97.92% accuracy, benefiting from residual connections for deep feature learning. Xception (97.45%) & MobileNet (96.87%) were efficient for use in low-resource environments. VGG16 (95.32%) & VGG19 (95.78%) showed stable performance, while AlexNet (93.47%) was the simplest yet effective model. This study highlights DenseNet121 as the most promising model for automated knee OA severity grading. Deep learning can transform medical diagnostics with accurate & scalable solutions. Future research could explore ensemble models & explainable AI for better reliability & clinical acceptance.},
        keywords = {X-ray images, Deep learning Models, DenseNet121, ResNet50, VGG 19, VGG 16, Xception, MobileNet, Alexnet, Accuracy, Feature Extraction.},
        month = {March},
        }