Detection and analysis of bone fractures in radiology images

Abstract

Computer detection of fractures can assist the doctors by flagging suspicious cases for closer examinations. In medical applications, accuracy in diagnosis medical problems is one of the important issues. Computer detection of fractures through medical images is also one of the challenging problems. Bone Fractures are the breakages in bones that occur due to various reasons like fall, road accident, fight etc. Detecting hairline fractures and complex fractures are crucial. Therefore, the automated fracture detection is necessary to identify complex fractures. The computerised fracture detection can help physicians to examine CT/X-ray images and to detect the severity of the injury within a short period. The features like presence of fracture, location of fracture, length, width and depth between the fractured bones are extracted by an automated fashion is vital to determine the severity of the injuries. This research work is carried out in X-ray and CT images. The minute cracks are detected in X-ray images based on the minimization of Fuzzy Index Measure. Determination of infestation sites of a crack in the X-ray image is done using the image processing procedure such as the grid formation, local thresholding, threshold value interpolation, segmentation using fuzzy index measure, background removal, and morphological filtering. The results obtained gives better sensitivity and specificity, yields an accuracy rate of 89%. The accuracy of diagnosis is vital, any misdiagnosis cause major problems to the patients. newline