Fuzzy-based Image Enhancement and Ensemble CNN Model for Bone Fracture Detection and Classification System

Bone fracture detection and classification play a critical role in the early diagnosis and treatment of orthopedic conditions. However, medical X-ray images often occur from low contrast and noise, complicating accurate fracture identification. So, this system is proposed as the effective system that integrates fuzzy-based image enhancement techniques with an ensemble Convolutional Neural Network (CNN) model for bone fracture detection and classification. To enhance image quality and diagnostic performance, this system combines the histogram equalization (HE) enhancement method with a fuzzy logic triangular membership function. This fuzzy-based enhancement method adjusts image contrast dynamically by utilizing fuzzy logic to improve the visibility of bone structures in noisy or low-contrast medical images. In this system, the enhanced images are then fed into an ensemble CNN model, designed to leverage the strengths of ResNet-50 and VGG-16 deep learning methods. To show the high precision in both detecting and classifying different types of fractures, this system is tested by using the MURA (Musculoskeletal Radiographs) dataset. The result shows that the proposed system outperforms the traditional image processing and single-model CNN approach. This system also supports to reduce the workload of medical professionals and minimize diagnostic errors.