Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Jul 15;13(14):4122.
doi: 10.3390/jcm13144122.

Fully Automated Measurement of Cobb Angles in Coronal Plane Spine Radiographs

Kenneth Chen  1   2 , Christoph Stotter  1   3 , Thomas Klestil  1   3 , Jennyfer A Mitterer  4   5 , Christopher Lepenik  1 , Stefan Nehrer  1
Affiliations

Fully Automated Measurement of Cobb Angles in Coronal Plane Spine Radiographs

Kenneth Chen et al. J Clin Med. .

Abstract

Background/Objectives: scoliosis is a three-dimensional structural deformity characterized by lateral and rotational curvature of the spine. The current gold-standard method to assess scoliosis is the measurement of lateral curvature of the spine using the Cobb angle in coronal plane radiographs. The interrater variability for Cobb angle measurements reaches up to 10°. The purpose of this study was to describe and assess the performance of a fully automated method for measuring Cobb angles using a commercially available artificial intelligence (AI) model trained on over 17,000 images, and investigate its interrater/intrarater agreement with a reference standard. Methods: in total, 196 AP/PA full-spine radiographs were included in this study. A reference standard was established by four radiologists, defined as the median of their Cobb angle measurements. Independently, an AI-based software, IB Lab SQUIRREL (version 1.0), also performed Cobb angle measurements on the same radiographs. Results: after comparing the readers' Cobb angle end vertebrae selection to the AI's outputs, 194 curvatures were considered valid for performance assessment, displaying an accuracy of 88.58% in end vertebrae selection. The AI's performance showed very low absolute bias, with a mean difference and standard deviation of differences from the reference standard of 0.16° ± 0.35° in the Cobb angle measurements. The ICC comparing the reference standard and the AI's measurements was 0.97. Conclusions: the AI model demonstrated good results in the determination of end vertebrae and excellent results in automated Cobb angle measurements compared to radiologists and could serve as a reliable tool in clinical practice and research.

Keywords: artificial intelligence; deep learning; machine learning; scoliosis; spinal asymmetry; spinal deformity.

PubMed Disclaimer

Conflict of interest statement

C.L. is a researcher in a funded project (grant number LS20-020) and an employee of ImageBiopsy Lab. S.N. is a member of the medical advisory board for ImageBiopsy Lab. All other authors declare no conflicts of interest. The funders had no role in the design of the study, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.

Figures

Figure 1
Figure 1
Cobb angle. The angle is measured between a parallel line of the superior endplate of the uppermost vertebrae and the inferior endplate of the lowermost vertebrae (vertebral bodies that are most tilted towards each other) of the scoliotic curves.
Figure 2
Figure 2
Flowchart depicting exclusions based on the Reference Standard Quality Assurance (QA) process (left) and mismatches between the reference standard and IB Lab SQUIRREL (right).
Figure 3
Figure 3
Schematic representation of IB Lab SQUIRREL’s three-step process: (1) Prediction of vertebral bodies, (2) vertebra labeling and (3) determination of Cobb angles and coronal balance.
Figure 4
Figure 4
Graphical report, as provided by IB Lab SQUIRREL [14,42].
Figure 5
Figure 5
(a) Density plot visualizing the distribution of Cobb angle measurements for the four human expert readers and IB Lab SQUIRREL. (b) Density plot visualizing the distribution of Cobb angle difference to the median reader.
Figure 6
Figure 6
(a) Bland–Altman plot with 95% Limits of Agreement (LoA). The red bar indicates the 95% confidence interval of the mean difference between IB Lab SQUIRREL and the reference standard. The gray bars show the 95% confidence interval of the Bland–Altman Limits of Agreement. (b) Scatter plot visualizing orthogonal linear regression (OLR, solid line) of IB Lab SQUIRREL and median expert reader Cobb angle (CA) outputs.
Figure 7
Figure 7
Cobb angle measurement of outlier 1. IB Lab SQUIRREL failed to set the inferior vertebral endplate correctly.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Shakil H., Iqbal Z.A., Al-Ghadir A.H. Scoliosis: Review of types of curves, etiological theories and conservative treatment. J. Back Musculoskelet. Rehabil. 2014;27:111–115. doi: 10.3233/BMR-130438. - DOI - PubMed
    1. Janicki J.A., Alman B. Scoliosis: Review of diagnosis and treatment. Paediatr. Child Health. 2007;12:771–776. doi: 10.1093/pch/12.9.771. - DOI - PMC - PubMed
    1. Shaw M., Adam C.J., Izatt M.T., Licina P., Askin G.N. Use of the iPhone for Cobb angle measurement in scoliosis. Eur. Spine. 2012;21:1062–1068. doi: 10.1007/s00586-011-2059-0. - DOI - PMC - PubMed
    1. Roggio F., Petrigna L., Filetti V., Vitale E., Rapisarda V., Musumeci G. Infrared thermography for the evaluation of adolescent and juvenile idiopathic scoliosis: A systematic review. J. Therm. Biol. 2023;113:103524. doi: 10.1016/j.jtherbio.2023.103524. - DOI - PubMed
    1. Trac S., Zheng R., Hill D.L., Lou E. Intra- and Interrater Reliability of Cobb Angle Measurements on the Plane of Maximum Curvature Using Ultrasound Imaging Method. Spine Deform. 2019;7:18–26. doi: 10.1016/j.jspd.2018.06.015. - DOI - PubMed

LinkOut - more resources