Design and performance analysis of Cobb angle measurement from X-ray images

Scoliosis seems to be the most frequently used type of spinal abnormality. Cobb angle measurement essentially relates to the quantification of the spinal stenosis in degrees, also it is a globally standardized approach for assessing scoliosis. Based on a recent survey, there really is no accurate and comprehensive technique for calculating the Cobb angle programmatically. This problem is crucial in the medical field pertaining to Cobb angle measurement to identify the exact position of Cobb in X-ray images. However, multiple investigations have demonstrated that there is inter and intra-observer variance when assessing the Cobb angle physically. The goal is to create a computer-assisted solution to reduce user-based Cobb angle measuring errors. The preprocessing filters and semi-automatic methods determine the overall architectural curved spine. Using the Cobb technique, results are inaccuracies and improper in the estimation of a scoliotic curvature's peak or bottom vertebra. The curve-fitting approach was used in this investigation to reduce uncertainty. Every patient had a digitally recorded poster anterior radiography image. The polynomial estimation is fitted using user-defined curvature midpoints that correspond to vertebral intersection points. The Cobb angle is computed by taking the first characteristic of the fitted polynomial function and dividing it by the number of vertebrae.