Deep learning algorithm enables automated Cobb angle measurements with high accuracy.

Autor: Hayashi D; Department of Radiology, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA, USA. daichi.alex.hayashi@gmail.com.; Department of Radiology, Tufts Medical Center, Tufts University School of Medicine, 800 Washington Street, #299, Boston, MA, 02111, USA. daichi.alex.hayashi@gmail.com., Regnard NE; Réseau Imagerie Sud Francilien, Lieusaint, France.; Gleamer, Paris, France., Ventre J; Gleamer, Paris, France., Marty V; Gleamer, Paris, France., Clovis L; Gleamer, Paris, France., Lim L; Gleamer, Paris, France., Nitche N; Gleamer, Paris, France., Zhang Z; Gleamer, Paris, France., Tournier A; Gleamer, Paris, France., Ducarouge A; Gleamer, Paris, France., Kompel AJ; Department of Radiology, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA, USA., Tannoury C; Department of Orthopedic Surgery, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA, USA., Guermazi A; Department of Radiology, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA, USA.; Department of Radiology, Veterans Affairs Boston Healthcare System, West Roxbury, Boston, MA, USA.
Jazyk: angličtina
Zdroj: Skeletal radiology [Skeletal Radiol] 2024 Dec 17. Date of Electronic Publication: 2024 Dec 17.
DOI: 10.1007/s00256-024-04853-7
Abstrakt: Objective: To determine the accuracy of automatic Cobb angle measurements by deep learning (DL) on full spine radiographs.
Materials and Methods: Full spine radiographs of patients aged > 2 years were screened using the radiology reports to identify radiographs for performing Cobb angle measurements. Two senior musculoskeletal radiologists and one senior orthopedic surgeon independently annotated Cobb angles exceeding 7° indicating the angle location as either proximal thoracic (apices between T3 and T5), main thoracic (apices between T6 and T11), or thoraco-lumbar (apices between T12 and L4). If at least two readers agreed on the number of angles, location of the angles, and difference between comparable angles was < 8°, then the ground truth was defined as the mean of their measurements. Otherwise, the radiographs were reviewed by the three annotators in consensus. The DL software (BoneMetrics, Gleamer) was evaluated against the manual annotation in terms of mean absolute error (MAE).
Results: A total of 345 patients were included in the study (age 33 ± 24 years, 221 women): 179 pediatric patients (< 22 years old) and 166 adult patients (22 to 85 years old). Fifty-three cases were reviewed in consensus. The MAE of the DL algorithm for the main curvature was 2.6° (95% CI [2.0; 3.3]). For the subgroup of pediatric patients, the MAE was 1.9° (95% CI [1.6; 2.2]) versus 3.3° (95% CI [2.2; 4.8]) for adults.
Conclusion: The DL algorithm predicted the Cobb angle of scoliotic patients with high accuracy.
Competing Interests: Declarations. Ethics approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Approval from the Institutional Review Board was obtained. HIPAA requirements were followed. Consent to participate: In keeping with the policies for a retrospective review, informed consent was not required. Consent to publish: In keeping with the policies for a retrospective review, informed consent was not required. Competing interests: DH has received publication royalties from Wolters-Kluwer and is a Section Editor for UpToDate. AG is a shareholder of Boston Imaging Core Lab, LLC. He has received consultancies, speaking fees, and/or honoraria from Novartis, Pfizer, Coval, Medipost, TrialSpark, ICM, and TissueGene. CT has received consultancies and royalties from DePuy Synthes and publication royalties from Wolters-Kluwer. AK has received consultancy fees from Caerus Medical. JV, VM, LC, LL, NN, ZZ, and AT are employees of Gleamer; AD and NER are co-founders of Gleamer.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz: