Automatic Detection and Segmentation of the Spinal Canal in Computed Tomographic Images

R.M. Rangayyan and H.J. Deglint (Canada)

Keywords

Automatic 3D image segmentation, detection of circles, spinal canal, fuzzy connectivity, Hough transform, com puted tomography

Abstract

A method for the automatic three-dimensional segmenta tion of the spinal canal in computed tomographic (CT) im ages is presented. The method uses a priori radiologi cal and anatomical information, mathematical morphology, and region-growing methods. The skin and peripheral fat structures are determined by delineating the air and other materials external to the body. Using the fat layer as ref erence, the bone structure is segmented. The Hough trans form for the detection of circles is applied to a cropped bone edge map that includes the thoracic vertebral struc ture. The centers of the detected circles are used as the seed voxels for the fuzzy connectivity algorithm that is em ployed to segment the spinal canal. The method was suc cessfully applied to eight CT volumes of four patients.

Important Links:

Go Back