Abstract

In medical diagnostics, bone fracture detection is critical because quick and precise identification is necessary for efficient treatment and patient care. Conventional approaches frequently have poor accuracy, complicated interfaces, and little user participation. This paper presents a bone fracture detection scheme with GUI-based application that uses Convolutional Neural Networks (CNNs) to identify bone fractures, focusing the ResNet-50 architecture. This proposed technique provides real-time prediction feedback, an easy-to-use image upload interface, and automated result storage, addressing the drawbacks of current systems such as restricted user input and complicated interfaces. The proposed research intends to improve detection efficiency, accuracy, and user experience by comparing ResNet-50 with alternative methods such as AlexNet, VGGNet, and k-means clustering. The proposed method closes the gap between conventional detection techniques and recent CNN technology by achieving fracture recognition with high reliability by utilizing CNN models, specifically ResNet50. Doctors now have a useful and precise technique for detecting bone fractures.