Abstract
We present a method for three dimensional (3D) reconstruction of in vivo human lumbar spine from biplanar radiographs with comparable results to Computerised Tomography (CT) scans or Magnetic Resonance Imaging (MRI) models. In this work, we used uncalibrated radiographs to reconstruct the 3D vertebrae and a priori information stored in an Active Shape Model (ASM) that is constructed using the Spherical Demons Algorithm. The method is semi-automatic as bounding boxes are required to delimit the positions of the vertebrae on biplanar radiographs of a patient. Optimisation is based on comparisons between simulated and actual radiographs. Finally, we compare the results to the models generated from MRI and CT scans. The results show the feasibility of generating personalised models of patients from biplanar radiographs.
The problem is the 3D reconstruction of lumbar spine from two radiographs (one lateral view and one postero-anterior view). Firstly, we generated a statistical shape model of each group of lumbar vertebrae from the data set (group of L1, group of L2 etc). Then, we fitted the statistical shape model to in vivo radiographs of patients with different abnormalities. The reconstruction has been validated against ground truth data generated from CT scans or MRI. [Display omitted]
•Three-dimensional reconstruction and visualisation of vertebrae from uncalibrated biplanar radiographs.•A landmark-free method of creating active shape models of lumbar vertebrae.•Mean errors within 1.8 mm compared to CT or MRI.