Radiographic scoliosis angle estimation: spline-based measurement reveals superior reliability compared to traditional C
- PDF / 4,205,984 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 56 Downloads / 152 Views
ORIGINAL ARTICLE
Radiographic scoliosis angle estimation: spline‑based measurement reveals superior reliability compared to traditional COBB method Peter Bernstein1 · Johannes Metzler2 · Marlene Weinzierl1 · Carl Seifert1 · Wadim Kisel1 · Markus Wacker2 Received: 15 May 2020 / Revised: 25 July 2020 / Accepted: 17 August 2020 © The Author(s) 2020
Abstract Introduction and objective Although being standard for scoliosis curve size estimation, COBB angle measurement is well known to be inaccurate, due to a high interobserver variance in end vertebra selection and end plate contour delineation. We propose a stepwise improvement by using a spline constructed from vertebra centroids to resemble spinal curve characteristics more closely. To enhance precision even further, a neural net was trained to detect the centroids automatically. Materials & Methods Vertebra centroids in AP spinal X-ray images of varying quality from 551 scoliosis patients were manually labeled by 4 investigators. With these inputs, splines were generated and the computed curve sizes were compared to the manually measured COBB angles and to the curve estimation obtained from the neural net. Results Splines achieved a higher interobserver correlation of 0.92–0.95 compared to manual COBB measurements (0.83– 0.92) and showed 1.5–2 times less variance, depending on the anatomic region. This translates into an average of 1° of interobserver measurement deviation for spline-based curve estimation compared to 3°–8° for COBB measurements. The neural net was even more precise and achieved mean deviations below 0.5°. Conclusion In conclusion, our data suggest an advantage of spline-based automated measuring systems, so further investigations are warranted to abandon manual COBB measurements. Keywords Radiographic · Scoliosis curve · Automatic measurement · COBB angle · Low image quality · Deep learning
Introduction Drawbacks of traditional COBB method COBB angle measurement on coronal whole-spine standing X-rays for the quantification of scoliosis curve severity has been implemented as gold standard in 1966 by the Scoliosis Research Society (SRS). The method itself is flawed by high inter- and intraobserver errors, ranging from 3° to 10° (95% CI), which can mainly be attributed to the manual nature of end vertebra selection and delineation [1]. In addition to this inaccuracy, the COBB method neglects important parts of the curve characteristics, e.g., apical vertebra translation. * Peter Bernstein [email protected] 1
Department for Orthopaedics and Traumatology, University Comprehensive Spine Center, University Hospital Dresden, Fetscherstrasse 74, 01307 Dresden, Germany
Faculty of Informatics/Mathematics, HTW Dresden, Friedrich‑List‑Platz 1, 01069 Dresden, Germany
2
Whole-spine X-ray images are currently produced at a varying expense in radiation exposure (ranging from conventional X-ray to modern low-dose imaging systems) with variable results. Thus, treatment decisions are being based on a high image acquisition variability and only mode
Data Loading...
