A Mask Based Segmentation Algorithm for Automatic Measurement of Cobb Angle from Scoliosis X-Ray Image

Abstract

Scoliosis is a medical condition in which a person's spine is curved from side to side. Current methodology of diagnosis of scoliosis: The doctors analyze an X-ray image and determine the cobb angle and vertebral twist. These two parameters are critical in the treatment of scoliosis. Bottlenecks associated with current methodology are inherent errors associated with manual measurement of cobb angle and vertebral twist from X-rays by the concerned doctors and the treatment that is meted out to a particular case of 'cobb angle' and vertebral twist by different doctors may differ with varying results. Hence it becomes imperative to select the best treatment procedure for attaining the best results. Highlights of the new methodology proposed: An X-ray image is accepted as input, Cobb angle is measured by the computer which is programmed to do so, thus eliminating the errors associated with the doctors interpretation.