Advanced computerized Classification of X-ray Images using CNN

Abstract

Healthcare technologies are growing rapidly in day-to-day life. Medical image modalities like X-Ray, MRI (Magnetic resonance imaging), CT (Computed tomography) scans are playing major role in diagnosing many diseases in humans. Among these, X-rays are mostly suggested by orthopaedics domain for bone related issues like fractures. Now-'s accidents become very common, which can lead to the fracture of bones. Fractures in bones are also very common in aged population. An automated X-ray based bone fracture detection is needed to reduce the burden on doctors. Here the proposed computer based classification of x-ray images is the first step for automatic bone fracture detection. BPNN (Back propagation Neural Network) based computerized classification of x-ray images is proposed along with M3 filtering which gives an accuracy of 91%, which is far better when compared any other methodology proposed till date.