Implementation of Computer Aided System for Automated Bone Fracture Detection using Digital Geometry

Abstract

A computer-aided detection and diagnosis (CADD) system must include automated fracture identification, which can lead to improve the treatment of orthopedics. This article provides a unified approach for detecting and evaluating orthopedic fractures in digital X-ray images of different types of long bones. Initially, the test images are applied to the gaussian filtering, which performs the image pre-processing, noise removal, artifact removal and also performs the image enhancements. Then, the preprocessed images are applied to the watershed segmentation approach. Then, the digital geometry-based line extraction operation is performed on the segmented images, which draws the lines on the fracture location. Finally, back propagated artificial neural network (BP-ANN) is applied on the fractured region, which identifies the test image status as fractured or not. The simulation results shows that the proposed method resulted in better performance as compared to the conventional approaches.