Bone Fracture Detection using deep learning (Resnet50) - Final project in the fourth year of the degree

Bone fracture or Implantation can be classified. The classification uses VGG16 architecture.

This project uses PyTorch to classify bone fractures. As well as fine-tuning some famous CNN architectures (like VGG 19, MobileNetV3, RegNet,...), we designed our own architecture. Additionally, we used Transformer architectures (such as Vision Transformer and Swin Transformer). This dataset is Bone Fracture Multi-Region X-ray, available on Kaggle.

This project in a X-ray bone fracture detection App protoype to assist medical professionals to fast-forward the diagnosis process, it runs on the user's device.