This project aims to enhance detecting accuracy and efficiency in medical settings, offering healthcare professionals a reliable tool for bone fracture detection. It used the deep learning various Convolutional Neural Network (CNN) model architectures, such as Hyper-tuned Visual Geometry Group 16-layer network (VGG16) and custom-developed architectures (CDCNNA) for the automatic detection of bone fractures in X-ray images. A dataset of 4800 X-ray images was utilized to train and test the system. The models were trained and fine-tuned using Keras Tuner to optimize their performance. Evaluation metrics such as accuracy, precision, recall, F1 score, and AUC-ROC were used to assess the models’ effectiveness. The hyper-tuned custom CNN and VGG16 models have shown significant improvements in bone fracture detection, achieving a near-perfect validation metrics. These results highlight the key role of hyperparameter optimization in deep learning, emphasizing the high potential of artificial intelligence in enhancing medical diagnostics with high accuracy and reliability.