Towards an Improved 3D Reconstruction by the Use of Automatic Bone Segmentation from CT Scan Images

Osteoarthritis is the most disabling joint disease manifested by the destruction of cartilage, bones, and synovial tissue. As a medical and surgical treatment, the knee prosthesis knows a great interest in scientific research. Indeed, knee prosthesis surgery is so delicate, therefore it would be interesting if the operating time and risks of surgery can be reduced and limited. In addition, clinicians use known clinical parameters to diagnose symptoms. Some of these parameters may be difficult to obtain with conventional X-rays. It is then possible to turn to other means of acquisition that allow visualization of bone structures in 3D and extract these parameters. The main objective of this work is to propose a new approach resides on automatic bone segmentation from CT scan images provided by the Cheikh Zaid International University Hospital in Rabat, Morocco. For this purpose, several computer vision techniques are used, namely morphological operations, edge detection, and clustering on slice-by-slice images to obtain a 3D reconstruction on Anatomage Table.