Quantitative measurement method (Green Village Method) of sternum and vertebral torsion using CT images

– Intra and inter examiner reproducibility and reliability.

Abstract

We developed a quantitative measurement method (Green Village Method: GVM) for thoracic deformity using CT images, an ordinary computer software, and a free software. In this study, inter and intra examiner reliability of GVM was studied. A total of 29 CT images of sternum and vertebral transverse process obtained from 29 patients with severe motor and intellectual disabilities, who had taken CT in the past for diseases such as pneumonia, were measured. There were 5 examiners, and they measured the distances between “the midpoint of sternum” and “the midpoint of transverse process” (GVM-d1), “the midpoint of sternum” and “the tip of spinous process” (GVM-d2), and between “the point of intersection of the sternum perpendicular bisector” and “the point of intersection of the perpendicular bisector of the transverse process and the vertebral anterior margin” (GVM-d3) as well as the angle between the sternum and vertebral torsions (GVM-a). GVM-d1, 2 and 3 showed high reliability with small standard error of less than 7 mm; this could be a highly accurate evaluation method. Since reliability of GVM-a was slightly low, it was thought necessary to reconsider the standard of the vertebral perpendicular bisector and further improve the accuracy.