3D Reconstruction of Scoliotic Spines from Stereoradiography and Depth Imaging

Abstract

Spine shape can be reconstructed from stereoradiography, but often requires specialized infrastructure or fails to account for subject posture. In this paper a protocol is presented for stereo reconstructions that integrates surface recordings with radiography and naturally accounts for variations in patient posture. Low cost depth cameras are added to an existing radiographic system to capture patient pose. A statistical model of human body shape is learned from public datasets and registered to depth scans, providing 3D correspondence across images for stereo reconstruction of radiographic landmarks. A radiographic phantom was used to validate these methods in vitro with RMS 3D landmark reconstruction error of 2.0 mm. Surfaces were automatically and reliably registered, with SD 12 mm translation disparity and SD .5° rotation. The proposed method is suitable for 3D radiographic reconstructions and may be beneficial in compensating for involuntary patient motion.

Multi-Modal Data Correspondence for the 4D Analysis of the Spine with Adolescent Idiopathic Scoliosis

“…One major challenge is to recover the external shape of the patient during the X-ray acquisitions, as the patient's pose is necessarily different from the surface scan acquired separately. A solution can be found in the introduction of 3D sensors during the X-ray acquisition [20]. We proposed a method that leverages the radio-opaque markers, visible in both the X-ray images and surface scan, to correct the pose of the model's back.…”