Treffer: Classification of Adolescent Idiopathic Scoliosis Curvature Using Contrastive Clustering.
Title:
Classification of Adolescent Idiopathic Scoliosis Curvature Using Contrastive Clustering.
Authors:
Kim DH; School of IT Convergence, University of Ulsan., Park S; Department of Orthopedic Surgery, Asan Medical Center, University of Ulsan College of Medicine, Ulsan., Kwon DW; School of IT Convergence, University of Ulsan., Lee CS; Department of Spine Surgery, Gangnam Saint Peter's Hospital, Seoul, Republic of Korea., Lee DH; Department of Orthopedic Surgery, Asan Medical Center, University of Ulsan College of Medicine, Ulsan., Cho JH; Department of Orthopedic Surgery, Asan Medical Center, University of Ulsan College of Medicine, Ulsan., Hwang CJ; Department of Orthopedic Surgery, Asan Medical Center, University of Ulsan College of Medicine, Ulsan.
Source:
Spine [Spine (Phila Pa 1976)] 2025 Dec 15; Vol. 50 (24), pp. 1692-1701. Date of Electronic Publication: 2025 May 23.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Lippincott Williams & Wilkins Country of Publication: United States NLM ID: 7610646 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1528-1159 (Electronic) Linking ISSN: 03622436 NLM ISO Abbreviation: Spine (Phila Pa 1976) Subsets: MEDLINE
Imprint Name(s):
Publication: Hagerstown, MD : Lippincott Williams & Wilkins
Original Publication: Hagerstown, Md., Medical Dept., Harper & Row.
Original Publication: Hagerstown, Md., Medical Dept., Harper & Row.
MeSH Terms:
Scoliosis*/classification , Scoliosis*/diagnostic imaging , Scoliosis*/surgery , Unsupervised Machine Learning*, Humans ; Adolescent ; Female ; Male ; Retrospective Studies ; Cluster Analysis ; Child ; Thoracic Vertebrae/diagnostic imaging ; Thoracic Vertebrae/surgery ; Reproducibility of Results ; Lumbar Vertebrae/diagnostic imaging ; Lumbar Vertebrae/surgery ; Machine Learning
References:
Kim H, Chang B-S, Chang SY. Current issues in the treatment of adolescent idiopathic scoliosis: a comprehensive narrative review. Asian Spine J. 2024;18:731–742.
Goldman SN, Hui AT, Choi S, et al. Applications of artificial intelligence for adolescent idiopathic scoliosis: mapping the evidence. Spine Deform. 2024;12:1545–1570.
Ovadia D. Classification of adolescent idiopathic scoliosis (AIS). J Child Orthop. 2013;7:25–28.
Hosseinpour-Feizi H, Soleimanpour J, Sales JG, et al. Lenke and King classification systems for adolescent idiopathic scoliosis: interobserver agreement and postoperative results. Int J Gen Med. 2011;4:821–825.
Parisini P, Di Silvestre M, Lolli F, et al. Selective thoracic surgery in the Lenke type 1A: King III and King IV type curves. Eur Spine J. 2009;18(suppl 1):82–88.
Lenke LG, Betz RR, Harms J, et al. Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am. 2001;83:1169–1181.
Lenke LG, Betz RR, Bridwell KH, et al. Intraobserver and interobserver reliability of the classification of thoracic adolescent idiopathic scoliosis. J Bone Joint Surg Am. 1998;80:1097–1106.
Dong Y, Li J, Huang S, et al. Artificial intelligence measurement of preoperative radiographs in adolescent idiopathic scoliosis based on multiple-view semantic segmentation. Global Spine J. 2024;15:21925682241270036.
Zhang B, Chen K, Yuan H, et al. Automatic Lenke classification of adolescent idiopathic scoliosis with deep learning. JOR Spine. 2024;7:e1327.
Lafage R, Fourman MS, Smith JS, et al. Can unsupervised cluster analysis identify patterns of complex adult spinal deformity with distinct perioperative outcomes? J Neurosurg Spine. 2023;38:547–557.
Kim HJ, Virk S, Elysee J, et al. The morphology of cervical deformities: a two-step cluster analysis to identify cervical deformity patterns. J Neurosurg Spine. 2020;32:353–359.
Kato S, Ohtomo N, Matsubayashi Y, et al. Post-operative shift in pain profile following fusion surgery for adult spinal deformity: a cluster analysis. Eur Spine J. 2024;33:2804–2812.
Mohanty S, Hassan FM, Lenke LG, et al. Machine learning clustering of adult spinal deformity patients identifies four prognostic phenotypes: a multicenter prospective cohort analysis with single surgeon external validation. Spine J. 2024;24:1095–1108.
Tingsheng L, Chunshan L, Shudan Y, et al. Validation of artificial intelligence in the classification of adolescent idiopathic scoliosis and the compairment to clinical manual handling. Orthop Surg. 2024;16:2040–2051.
Shandilya SK, Srivastav A, Yemets K, et al. YOLO-based segmented dataset for drone vs. bird detection for deep and machine learning algorithms. Data Brief. 2023;50:109355.
Edelmers E, Nikulins A, Sprudza KL, et al. AI-assisted detection and localization of spinal metastatic lesions. Diagnostics (Basel). 2024;14:2458.
Shen X, Tao L, Chen X, et al. Contrastive learning of shared spatiotemporal EEG representations across individuals for naturalistic neuroscience. Neuroimage. 2024;301:120890.
Zhang A, Yu Y, Li S, et al. Contrastive learning-based personalized tag recommendation. Sensors (Basel). 2024;24:6061.
Song J, McNeany J, Wang Y, et al. Riemannian manifold-based geometric clustering of continuous glucose monitoring to improve personalized diabetes management. Comput Biol Med. 2024;183:109255.
Yang Y, Yang M, Shi D, et al. Single-cell RNA Seq reveals cellular landscape-specific characteristics and potential etiologies for adolescent idiopathic scoliosis. JOR Spine. 2021;4:e1184.
Vergari C, Skalli W, Gajny L. A convolutional neural network to detect scoliosis treatment in radiographs. Int J Comput Assist Radiol Surg. 2020;15:1069–1074.
Kim HS, Jeong JY, Cho YJ, et al. Validation of the visual body image classification in adolescent idiopathic scoliosis: a retrospective study. Asian Spine J. 2024;18:829–835.
Kalanjiyam GP, Chandramohan T, Raman M, et al. Artificial intelligence: a new cutting-edge tool in spine surgery. Asian Spine J. 2024;18:458–471.
Passias PG, Onafowokan OO, Tretiakov P, et al. Current concepts in adult cervical spine deformity surgery. J Neurosurg Spine. 2024;40:439–452.
Buell TJ, Smith JS. Adult spinal deformity and novel classifications: is coronal malalignment making a comeback?: Commentary on “Obeid-coronal malalignment classification is age related and independently associated to personal reported outcome measurement scores in the nonfused spine”. Neurospine. 2021;18:481–483.
Baroncini A, Larrieu D, Bourghli A, et al. Machine learning can predict surgical indication: new clustering model from a large adult spine deformity database. Eur Spine J. 2025. doi: 10.1007/s00586-025-08653-y. (PMID: 10.1007/s00586-025-08653-y)
Tavana P, Akraminia M, Koochari A, et al. Classification of spinal curvature types using radiography images: deep learning versus classical methods. Artif Intell Rev. 2023:1–33 doi: 10.1007/s10462-023-10480-w. (PMID: 10.1007/s10462-023-10480-w.)
Erdem MN, Karaca S, Korkmaz MF, et al. Criteria for ending the distal fusion at the L3 vertebra vs. L4 in surgical treatment of adolescent idiopathic scoliosis patients with Lenke type 3C, 5C, and 6C curves: results after ten years of follow-up. Cureus. 2018;10:e2564.
Gu Q, Bao H, Shu S, et al. Hyper-selective posterior fusion is recommended when the modified S-line is positive in Lenke 5C adolescent idiopathic scoliosis. Orthop Surg. 2024;16:1390–1398.
Oba H, Takahashi J, Kobayashi S, et al. Upper instrumented vertebra to the right of the lowest instrumented vertebra as a predictor of an increase in the main thoracic curve after selective posterior fusion for the thoracolumbar/lumbar curve in Lenke type 5C adolescent idiopathic scoliosis: multicenter study on the relationship between fusion area and surgical outcome. J Neurosurg Spine. 2019;31:857–864.
Park SJ, Park JS, Kang DH, et al. The optimal lowest instrumented vertebra to prevent the distal adding-on phenomenon in patients undergoing selective thoracic fusion for adolescent idiopathic scoliosis with Lenke type 1A and 1B curves: comparison of nine selection criteria. J Clin Med. 2024;13:3859.
Lee CS, Hwang CJ, Jung HS, et al. Association between vertebral rotation pattern and curve morphology in adolescent idiopathic scoliosis. World Neurosurg. 2020;143:e243–e252.
Goldman SN, Hui AT, Choi S, et al. Applications of artificial intelligence for adolescent idiopathic scoliosis: mapping the evidence. Spine Deform. 2024;12:1545–1570.
Ovadia D. Classification of adolescent idiopathic scoliosis (AIS). J Child Orthop. 2013;7:25–28.
Hosseinpour-Feizi H, Soleimanpour J, Sales JG, et al. Lenke and King classification systems for adolescent idiopathic scoliosis: interobserver agreement and postoperative results. Int J Gen Med. 2011;4:821–825.
Parisini P, Di Silvestre M, Lolli F, et al. Selective thoracic surgery in the Lenke type 1A: King III and King IV type curves. Eur Spine J. 2009;18(suppl 1):82–88.
Lenke LG, Betz RR, Harms J, et al. Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am. 2001;83:1169–1181.
Lenke LG, Betz RR, Bridwell KH, et al. Intraobserver and interobserver reliability of the classification of thoracic adolescent idiopathic scoliosis. J Bone Joint Surg Am. 1998;80:1097–1106.
Dong Y, Li J, Huang S, et al. Artificial intelligence measurement of preoperative radiographs in adolescent idiopathic scoliosis based on multiple-view semantic segmentation. Global Spine J. 2024;15:21925682241270036.
Zhang B, Chen K, Yuan H, et al. Automatic Lenke classification of adolescent idiopathic scoliosis with deep learning. JOR Spine. 2024;7:e1327.
Lafage R, Fourman MS, Smith JS, et al. Can unsupervised cluster analysis identify patterns of complex adult spinal deformity with distinct perioperative outcomes? J Neurosurg Spine. 2023;38:547–557.
Kim HJ, Virk S, Elysee J, et al. The morphology of cervical deformities: a two-step cluster analysis to identify cervical deformity patterns. J Neurosurg Spine. 2020;32:353–359.
Kato S, Ohtomo N, Matsubayashi Y, et al. Post-operative shift in pain profile following fusion surgery for adult spinal deformity: a cluster analysis. Eur Spine J. 2024;33:2804–2812.
Mohanty S, Hassan FM, Lenke LG, et al. Machine learning clustering of adult spinal deformity patients identifies four prognostic phenotypes: a multicenter prospective cohort analysis with single surgeon external validation. Spine J. 2024;24:1095–1108.
Tingsheng L, Chunshan L, Shudan Y, et al. Validation of artificial intelligence in the classification of adolescent idiopathic scoliosis and the compairment to clinical manual handling. Orthop Surg. 2024;16:2040–2051.
Shandilya SK, Srivastav A, Yemets K, et al. YOLO-based segmented dataset for drone vs. bird detection for deep and machine learning algorithms. Data Brief. 2023;50:109355.
Edelmers E, Nikulins A, Sprudza KL, et al. AI-assisted detection and localization of spinal metastatic lesions. Diagnostics (Basel). 2024;14:2458.
Shen X, Tao L, Chen X, et al. Contrastive learning of shared spatiotemporal EEG representations across individuals for naturalistic neuroscience. Neuroimage. 2024;301:120890.
Zhang A, Yu Y, Li S, et al. Contrastive learning-based personalized tag recommendation. Sensors (Basel). 2024;24:6061.
Song J, McNeany J, Wang Y, et al. Riemannian manifold-based geometric clustering of continuous glucose monitoring to improve personalized diabetes management. Comput Biol Med. 2024;183:109255.
Yang Y, Yang M, Shi D, et al. Single-cell RNA Seq reveals cellular landscape-specific characteristics and potential etiologies for adolescent idiopathic scoliosis. JOR Spine. 2021;4:e1184.
Vergari C, Skalli W, Gajny L. A convolutional neural network to detect scoliosis treatment in radiographs. Int J Comput Assist Radiol Surg. 2020;15:1069–1074.
Kim HS, Jeong JY, Cho YJ, et al. Validation of the visual body image classification in adolescent idiopathic scoliosis: a retrospective study. Asian Spine J. 2024;18:829–835.
Kalanjiyam GP, Chandramohan T, Raman M, et al. Artificial intelligence: a new cutting-edge tool in spine surgery. Asian Spine J. 2024;18:458–471.
Passias PG, Onafowokan OO, Tretiakov P, et al. Current concepts in adult cervical spine deformity surgery. J Neurosurg Spine. 2024;40:439–452.
Buell TJ, Smith JS. Adult spinal deformity and novel classifications: is coronal malalignment making a comeback?: Commentary on “Obeid-coronal malalignment classification is age related and independently associated to personal reported outcome measurement scores in the nonfused spine”. Neurospine. 2021;18:481–483.
Baroncini A, Larrieu D, Bourghli A, et al. Machine learning can predict surgical indication: new clustering model from a large adult spine deformity database. Eur Spine J. 2025. doi: 10.1007/s00586-025-08653-y. (PMID: 10.1007/s00586-025-08653-y)
Tavana P, Akraminia M, Koochari A, et al. Classification of spinal curvature types using radiography images: deep learning versus classical methods. Artif Intell Rev. 2023:1–33 doi: 10.1007/s10462-023-10480-w. (PMID: 10.1007/s10462-023-10480-w.)
Erdem MN, Karaca S, Korkmaz MF, et al. Criteria for ending the distal fusion at the L3 vertebra vs. L4 in surgical treatment of adolescent idiopathic scoliosis patients with Lenke type 3C, 5C, and 6C curves: results after ten years of follow-up. Cureus. 2018;10:e2564.
Gu Q, Bao H, Shu S, et al. Hyper-selective posterior fusion is recommended when the modified S-line is positive in Lenke 5C adolescent idiopathic scoliosis. Orthop Surg. 2024;16:1390–1398.
Oba H, Takahashi J, Kobayashi S, et al. Upper instrumented vertebra to the right of the lowest instrumented vertebra as a predictor of an increase in the main thoracic curve after selective posterior fusion for the thoracolumbar/lumbar curve in Lenke type 5C adolescent idiopathic scoliosis: multicenter study on the relationship between fusion area and surgical outcome. J Neurosurg Spine. 2019;31:857–864.
Park SJ, Park JS, Kang DH, et al. The optimal lowest instrumented vertebra to prevent the distal adding-on phenomenon in patients undergoing selective thoracic fusion for adolescent idiopathic scoliosis with Lenke type 1A and 1B curves: comparison of nine selection criteria. J Clin Med. 2024;13:3859.
Lee CS, Hwang CJ, Jung HS, et al. Association between vertebral rotation pattern and curve morphology in adolescent idiopathic scoliosis. World Neurosurg. 2020;143:e243–e252.
Contributed Indexing:
Keywords: adolescent idiopathic scoliosis; classification; clustering; deep learning; machine learning; unsupervised learning
Entry Date(s):
Date Created: 20250523 Date Completed: 20251124 Latest Revision: 20251124
Update Code:
20251125
DOI:
10.1097/BRS.0000000000005381
PMID:
40407029
Database:
MEDLINE
Weitere Informationen
Study Design: Retrospective image analysis study.
Objective: To propose a novel classification system for adolescent idiopathic scoliosis (AIS) curvature using unsupervised machine learning and evaluate its reliability and clinical implications.
Summary of Background Data: Existing AIS classification systems, such as King and Lenke, have limitations in accurately describing curve variations, particularly long C-shaped curves or curves with distinct characteristics. Unsupervised machine learning offers an opportunity to refine classification and enhance clinical decision-making.
Materials and Methods: A total of 1156 AIS patients who underwent deformity correction surgery were analyzed. Standard posteroanterior radiographs were segmented using U-net algorithms. Contrastive clustering was employed for automatic grouping, with the number of clusters ranging from three to 10. Cluster quality was assessed using t-SNE and Silhouette scores. Clusters were defined based on consensus among spine surgeons. Interobserver reliability was evaluated using kappa coefficients.
Results: Six clusters were identified, reflecting variations in structural curve location, single (C-shaped) versus double (S-shaped) curves, and thoracolumbar curve characteristics. Cluster reliability was moderate (kappa = 0.701-0.731). The silhouette score was 0.308, with t-SNE demonstrating distinct clustering patterns. The classification highlighted differences not captured by the Lenke classification, such as thoracic curves confined to the thoracic spine versus those extending to the lumbar spine.
Conclusion: Unsupervised machine learning successfully categorized AIS curvatures into six distinct clusters, revealing meaningful patterns such as unique variations in thoracic and lumbar curves. These findings could potentially inform surgical planning and prognostic assessments. However, further studies are needed to validate clinical applicability and improve clustering quality.
Level of Evidence: Level III.
(Copyright © 2025 Wolters Kluwer Health, Inc. All rights reserved.)
The authors report no conflicts of interest.