A non-ionizing technique for three-dimensional measurement of the lumbar spine.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 27823804.
- Also identified by DOI 10.1016/j.jbiomech.2016.10.048.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Comprehensive assessments of scoliotic deformity and spinal instability require repetitive three-dimensional (3D) measurements of motion segments at different functional postures. However, accurate 3D measurement of the spine is a challenging task. In this paper, we present a novel, non-invasive, non-ionizing technique to quantify 3D poses of lumbar motion segments in terms of clinically meaningful anatomical coordinates. The technique used ultra-short echo time (UTE) magnetic resonance (MR) images to construct subject-specific geometrical models of individual vertebrae and registered them with 3D ultrasound dataset acquired during pose measurements. A hierarchical registration approach was used to minimize the detrimental effects of speckle noise and artifacts within soft tissues on registration accuracy. The technique was validated using a human dry bone specimen as well as a fresh porcine cadaver. Registration errors were determined by comparing with a gold standard fiducial-based registration. Results showed that the technique is accurate and reliable with bias in sub-degree and sub-millimeter level (except for the flexion-extension of the porcine cadaver experiment, which was -1.74°), and average precision of 1.11° in rotation and 0.86mm in position for the human dry bone experiment, and 1.26° and 1.23mm for the porcine cadaver experiment. Given its non-ionizing nature, the UTE MR-ultrasound registration technique is particularly useful for repeated measurements and longitudinal follow-up. With further refinement and validation, it could be a powerful tool for 3D spinal assessment.
Medical subject headings
- Imaging, Three-Dimensional
- Lumbar Vertebrae