Hybrid Deep Transfer Learning Framework for Humerus Fracture Detection and Classification from X-ray Images

The detection and classification of humerus fractures from X-ray images are crucial for effective medical diagnosis and treatment planning. Manual assessment of such fractures is time-consuming and prone to errors, emphasizing the need for automated systems. In this study, we propose a Hybrid Deep Transfer Learning Framework for Humerus Fracture Detection and Classification from X-ray Images. Leveraging deep learning techniques, we amassed a dataset of 1266 radiographic images from the publicly available MURA dataset, encompassing both negative (non-fractured) and positive (fractured) cases. Preprocessing techniques were employed to enhance image quality, followed by data augmentation to mitigate overfitting and bolster system accuracy. Subsequently, a hybrid model comprising ResNet50 and DenseNet121 architectures was utilized for feature extraction and classification. Through experimentation with various optimizers, we achieved the highest accuracy of $93.41 \%$ using the Adam optimizer. Additionally, precision, recall, and F1-score metrics were computed to evaluate model performance comprehensively. Comparative analyses were conducted with other pre-trained models, showcasing the effectiveness of our proposed framework. Our results highlight the deep transfer learning’s effectiveness in humerus fracture detection, providing a promising path forward for the development of medical imaging technologies.