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{166575,
        author = {Maddela Shreshta and Dr. V. Uma Rani },
        title = {ADVANCEMENTS IN KNEE DISEASE DIAGNOSIS: CUSTOM MODELS AND IMAGE FUSION},
        journal = {International Journal of Innovative Research in Technology},
        year = {2024},
        volume = {11},
        number = {2},
        pages = {1403-1414},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=166575},
        abstract = {Utilizing X-ray knee pictures, the drive means to identify knee osteoarthritis. This method utilizes X-ray pictures to precisely distinguish osteoarthritis in knee joints, which are economical and normal. Image processing-based knee osteoarthritis detection strategies are wrong and loose. The review presents a new and fitted knee osteoarthritis detection and classification strategy to address current imperatives. A state of the art object recognizable proof design, CenterNet, is created for the recommended strategy. A pixel-wise voting methodology extricates highlights at a granular level in this CenterNet. This CentreNet customization endeavors to further develop knee osteoarthritis detection accuracy and dependability. The feature extraction model purposes DenseNet201. The firmly connected layers of DenseNet increment highlight reuse and lessen inclination concerns. The model purposes DenseNet201 to remove the most delegate knee attributes to further develop highlight extraction. The model plans to distinguish knee osteoarthritis in X-rays precisely. The Kellgren and Lawrence (KL) evaluating framework will likewise be utilized to go past location to decide osteoarthritis seriousness. This comprehensive methodology empowers a refined sickness figuring out, further developing determination and treatment. The undertaking gives a coordinated procedure consolidating strong order models (Xception, InceptionV3), proficient item discovery strategies (YOLOv5, YOLOv8), and a practical Flask front end. This strategy utilizes progressed classification and identification calculations to establish a protected and smooth testing climate.},
        keywords = {Machine learning, detection performance, HCI, classification, deep learning, multi-scale features.},
        month = {July},
        }