Fracture detection from X-ray images using different Machine Learning Techniques

In this period, X-rays are the main instruments used to examine suspected human fractures. It takes a lot of time to manually examine X-rays, which calls for experienced radiologists or skilled orthopedic surgeons. The primary care physicians' excessive referrals contribute to the challenging diagnostic environment and the shortage of radiologists at primary/community health centre. Additionally, the lack of skilled radiologists and orthopedic surgeons in medically underdeveloped regions, such as rural India, has driven us to create an automated fracture identification model. Digital image processing is a technique of processing an image with the help of a digital computer. In computer image processing, an edge detection algorithm looks for any kind of edge modifications. In this work, edge detection is used to first identify edges, and it is discovered that canny edge detection is the best option. In second stage, HOG feature extraction method is used to extract features from X-ray images. The images are then categorise using classifiers to distinguish between bones with and without fractures. For classification, the classifiers like SVM (Support Vector Machine), KNN (K-Nearest Neighbour), and RF (Random Forest) are utilised. This study strongly suggests SVM for bone fracture detection.