| Literature DB >> 33452611 |
David Xinzheyang Li1, Maha S Ead2, Kajsa K Duke2, Jacob L Jaremko3, Lindsey Westover2.
Abstract
Understanding bilateral pelvic symmetry can be useful for analyzing complex pelvis anatomy and simplifying difficult procedures for pelvic fractures. This paper aims to quantify the degree of regional pelvic symmetry using computer-based methods. CT scans of 30 intact pelvises were digitized into 3D models and regions were defined: the ilium, acetabulum, pubis, and ischium. The right hemipelvis was aligned with the left, and deviations between the two models were quantified using method 1 (global registration) and method 2 (local registration). Symmetry was evaluated using the root mean square (RMS) of the deviations and the percentage of points within preset thresholds of ± 2 mm and ± 1 mm. The results showed that > 86% of points are within the ± 2 mm deviation threshold and average RMS are < 1.33 mm. For all regions, method 2 showed lower deviations than method 1. The pubis and ischium regions showed a large difference in symmetry between the two methods indicating high local symmetry, but a degree of global asymmetry. Conversely, the acetabular and iliac regions showed similar levels of symmetry with the two methods. When evaluated locally, the pelvic regions can be considered highly symmetric; the acetabulum is highly symmetric globally as well. These findings can be used in future studies to assess the feasibility of patient-specific implants using the mirrored contralateral hemipelvis as a template for unilateral pelvic fracture fixation. The left image shows the "cut planes" used to define four pelvic regions: the ilium, acetabulum, pubis, and ischium. The right image shows a deviation color map (DCM) used to quantify bilateral pelvic symmetry. The scale and color illustrate the degree of deviation of the left hemipelvis with the right hemipelvis with the units in millimeters (mm).Entities:
Keywords: 3D models; Coordinate system; Defined regions; Deviation analysis; Pelvis; Symmetry
Year: 2021 PMID: 33452611 DOI: 10.1007/s11517-020-02296-5
Source DB: PubMed Journal: Med Biol Eng Comput ISSN: 0140-0118 Impact factor: 2.602