Importance of the shape and orientation of the spine and pelvis for the vertebral column pathologies diagnosis with using machine learning methods.

Abstract

Disc-hernia and spondylolisthesis are the major disorders related to vertebral column. Expert medical doctors examine the pelvis and lumbar spine shape and orientation to diagnose the disorder. However some shape and orientation parameters may be misleading and are unclear. Therefore recently the automatic diagnosis methods (classification methods) have been proposed to help medical doctors. However the proposed results may be misleading, since are triple-classification results. Hence the automatic diagnosis of each disorder should be tested separately also. Moreover, the medical doctor need is determination of more useful parameters rather than the automatic diagnosis proposals. In this study, which parameters should be used to diagnose which disorder, was determined for more informative automatic diagnosis proposal. According to the findings, success rate of the classification of healthy-spondylolisthesis and herniated-spondylolisthesis are high (about 97% for healthy-spondylolisthesis and about 98% for hernia-spondylolisthesis classification) but the same result cannot be determined for hernia-healthy classification. Also the most important parameter for the Spondylolisthesis classification is grade of spondylolisthesis (GS).