International Journal of Scientific Research and Engineering Development

The diagnosis of a bone tumor from medical images is a challenging task due to the tumor’s nature complexity and the image’s interpretation complexity. Although deep learning-based approaches can effectively diagnose a tumor automatically, the available methods have limited contribution to the overall understanding of the diagnosis. This work proposed an hybrid explainable machine learning approach for the automated diagnosis of bone cancer from medical images. The paper discussed the possibility of using the deep learning-based features extracting with handcrafted texture features for the tumor classification task. Together with transfer learning, data augmentation, and hyperparameter optimization, Grad-CAM was applied for better interpretation of the model. The work proposed to implement a web application for medical image uploading for automatic diagnosis. The study aimed to contribute to the field by providing higher accuracy and explainability for diagnosis, thus improving detection and treatment of cancer.

K S Seethalakshmi, Abhishek K C, Chethan P, Krithika S Poojari, Neha J Shet, "A Hybrid Explainable Learning System for Automated Bone Cancer Diagnosis Using Medical Imaging" International Journal of Scientific Research and Engineering Development, V9(4): Page(1063-1068) July-August 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.