3-D Reconstruction of the Spine From Biplanar Radiographs Based on Contour Matching Using the Hough Transform
Authors:
- 1 July 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Engineering
- Vol. 60 (7) , 1954-1964
Abstract
The purpose of this study was to develop and evaluate a method for three-dimensional (3-D) reconstruction of the spine from biplanar radiographs. The approach was based on vertebral contour matching for estimating vertebral orientations and locations. Vertebral primitives were initially positioned under constraint of the 3-D spine midline, which was estimated from manually identified control points. Vertebral orientations and locations were automatically adjusted by matching projections of 3-D primitives with vertebral edges on biplanar radiographs based on the generalized Hough transform technique with a deformation tolerant matching strategy. We used graphics processing unit to accelerate reconstruction. Accuracy and precision were evaluated using radiographs from 15 scoliotic patients and a spine model in 24 poses. On in vivo radiographs, accuracy was within 2.8° for orientation and 2.4 mm for location; precision was within 2.3° for orientation and 2.1 mm for location. results were slightly better on model radiographs than on in vivo radiographs but without significance (p > 0.05). The duration for user intervention was less than 2 min, and the computation time was within 3 min. Results indicated the method's reliability. It is a promising tool to determine 3-D spinal geometry with acceptable user interaction.Keywords
This publication has 43 references indexed in Scilit:
- Intervertebral disc segmentation in MR images using anisotropic oriented fluxMedical Image Analysis, 2013
- Reproducibility of Standing Posture for X-Ray Radiography: A Feasibility Study of the BalancAid with Healthy Young SubjectsAnnals of Biomedical Engineering, 2010
- 2D-3D Registration of CT Vertebra Volume to Fluoroscopy Projection: A Calibration Model AssessmentEURASIP Journal on Advances in Signal Processing, 2009
- Automatic Cobb Measurement of Scoliosis Based on Fuzzy Hough Transform with Vertebral Shape PriorJournal of Digital Imaging, 2008
- Spine Localization in X-ray Images Using Interest Point DetectionJournal of Digital Imaging, 2008
- Deformation tolerant generalized Hough transform for sketch-based image retrieval in complex scenesImage and Vision Computing, 2007
- Computerized detection of vertebral compression fractures on lateral chest radiographs: Preliminary results with a tool for early detection of osteoporosisMedical Physics, 2006
- Parameterizing Arbitrary Shapes via Fourier Descriptors for Evidence-Gathering ExtractionComputer Vision and Image Understanding, 1998
- Optimized vertical stereo base radiographic setup for the clinical three-dimensional reconstruction of the human spineJournal of Biomechanics, 1994
- Estimation of Vertebral Rotation and the Spinal and Rib Cage Deformity in Scoliosis by Computer TomographySpine, 1981