Automatic measurement of intervertebral movements using radiographic images

Abstract

Measurement of intervertebral movements is essential in the assessment and diagnosis of patient's instability. However, diagnosis of the underlying causes remains problematic despite of extensive study. Reasons for this arise from the variability of detecting vertebral body landmarks, labor and time-consuming of manual point placement, incompletely description of the vertebral body shape and also from the structural complexity of the spine. In this study, the precision and accuracy of a new automatic method for morphometry of intervertebral movements were estimated. Active Contour is a key feature of segmentation and provides rapid and accurate measurement of vertebral shape. Fourier descriptors are used to represent the vertebral shapes. Genetic Algorithm (GA) is then applied to determine the spinal kinematics. Reproducible and reliable determinations of the intervertebral movements of the lumbosacral spine, when performing Flexion-Extension motions, are addressed. This paper describes the accuracy and feasibility of an active shape model (ASM) and Genetic Algorithm (GA) to measure spine kinematics. © 2005 IEEE.