Bones detection in the pelvic area on the basis of YOLO neural network

The article focuses on application of You Only Look Once (YOLO) neural network in the process of individual bones recognition, in the set of CT slices of human pelvic area. The network was trained on custom data and then it was employed to detect bone structures in similar yet different CT data. The accuracy of the detection was investigated. In the final step, the bounding boxes of detected regions, created by the YOLO algorithm were employed to position the idealized models of bones with respect to CT data.