An Explainable Deep Learning Framework for Automated Bone Fracture Detection and Classification

The diagnosis of fractures and their classification are the most important factors in clinical assessment and planning of treatment. Manual interpretation of X-ray images is often time-consuming and error-prone. This work presents an AI-powered deep learning framework for automated fracture detection and classification using CNN. The system incorporates explainable AI techniques like Gradient-weighted Class Activation Mapping (Grad-CAM) and Canny edge detection to enhance interpretability and visual transparency. Furthermore, this framework has been validated in three stages: binary fracture detection, multi-class fracture classification, and explainability-based analysis. The experimental results obtained reveal that the proposed model using ResNet50 achieved a classification accuracy of 91.33%, outperforming the state-of-the-art baselines. This includes Grad-CAM and Canny overlays that improve visual understanding, thus supporting clinical decision-making and enhancing the trustworthiness of AI-driven diagnosis.