An automated CT-based method to quantify the 3D anatomy of the posterior elements of the human thoracic and lumbar spine

Abstract

This study presents the development and implementation of a highly automated computed tomography-based method to quantify 3D posterior element anatomy. The use of the posterior elements as the site of the bone–implant interface in spinal posterior fusion plating requires quantitative morphologic human data on these structures. Human thoracic and lumbar vertebrae (n = 307) were automatically segmented to isolate the spinous processes and laminae using atlas-based deformable registration. Analysis was conducted to determine lateral facing surface area, transverse and coronal plane angulation, and interspinous process distance between adjacent levels. The considered metrics demonstrated division based on spinal level (T12–L5 vs. T1–T11), with larger measured outcomes for males. This anatomic study provides an automated method for acquiring 3D morphological information that can be used to guide posterior element device design.