A statistical lumbar spine geometry model accounting for variations by Age, Sex, Stature, and body mass index.

Abstract

The objective of this study was to develop a statistical lumbar spine geometry model accounting for morphological variations among the adult population. Five lumber vertebrae and lumber spine curvature were collected from CT scans of 82 adult subjects through CT segmentation, landmark identification, and template mesh mapping. Generalized Procrustes Analysis (GPA), Principal Component Analysis (PCA), and multivariate regression analysis were conducted to develop the statistical lumbar spine model. Two statistical models were established to predict the vertebrae geometry and lumbar curvature respectively. Using the statistical models, a lumbar spine finite element (FE) model could be rapidly generated with a given set of age, sex, stature, and body mass index (BMI). The results showed that the lumbar spine vertebral size was significantly affected by stature, sex and age, and the lumbar spine curvature was significantly affected by stature and age. This statistical lumbar spine model could serve as the geometric basis for quantifying effects of human characteristics on lumbar spine injury risks in impact conditions.