Deep learning for automated spinopelvic parameter measurement from radiographs: a meta-analysis

Abstract

Aim: Quantitative measurement of spinopelvic parameters from radiographs is important for assessing spinal disorders but is limited by the subjectivity and inefficiency of manual techniques. Deep learning may enable automated measurement with accuracy rivaling human readers.

Methods: PubMed, Embase, Scopus, and Cochrane databases were searched for relevant studies. Eligible studies were published in English, used deep learning for automated spinopelvic measurement from radiographs, and reported performance against human raters. Mean absolute errors and correlation coefficients were pooled in a meta-analysis.

Results: Fifteen studies analyzing over 10,000 radiographs met the inclusion criteria, employing convolutional neural networks (CNNs) and other deep learning architectures. Pooled mean absolute errors were 4.3° [95% confidence interval (CI) 3.2-5.4] for Cobb angle, 3.9° (95%CI 2.7-5.1) for thoracic kyphosis, 3.6° (95%CI 2.8-4.4) for lumbar lordosis, 1.9° (95%CI 1.3-2.5) for pelvic tilt (PT), 4.1° (95%CI 2.7-5.5) for pelvic incidence (PI), and 1.3 cm (95%CI 0.9-1.7) for sagittal vertical axis (SVA). Intraclass correlation coefficients exceeded 0.81, indicating strong agreement between automated and manual measurements.

Conclusion: Deep learning demonstrates promising accuracy for automated spinopelvic measurement, potentially rivaling experienced human readers. However, further optimization and rigorous multicenter validation are required before clinical implementation. These technologies may eventually improve the efficiency and reliability of quantitative spine image analysis.

Keywords

Deep learning / spine parameters / pelvic parameters

References

[1] Le Huec JC, Thompson W, Mohsinaly Y, Barrey C, Faundez A. Sagittal balance of the spine.Eur Spine J2019;28:1889-905

[2] Vrtovec T,Likar B,Viergever MA.A review of methods for evaluating the quantitative parameters of sagittal pelvic alignment.Spine J2012;12:433-46

[3] Legaye J,Hecquet J.Pelvic incidence: a fundamental pelvic parameter for three-dimensional regulation of spinal sagittal curves.Eur Spine J1998;7:99-103 PMCID:PMC3611230

[4] Glassman SD,Dimar JR,Berven S.The impact of positive sagittal balance in adult spinal deformity.Spine2005;30:2024-9

[5] Maillot C,Fort D,Le Huec JC.Reproducibility and repeatability of a new computerized software for sagittal spinopelvic and scoliosis curvature radiologic measurements: Keops®.Eur Spine J2015;24:1574-81

[6] Lafage R,Henry JK.Validation of a new computer-assisted tool to measure spino-pelvic parameters.Spine J2015;15:2493-502

[7] Litjens G,Bejnordi BE.A survey on deep learning in medical image analysis.Med Image Anal2017;42:60-88

[8] Shen D,Suk HI.Deep learning in medical image analysis.Annu Rev Biomed Eng2017;19:221-48 PMCID:PMC5479722

[9] Galbusera F,Bassani T.Artificial intelligence and machine learning in spine research.JOR Spine2019;2:e1044 PMCID:PMC6686793

[10] LeCun Y,Hinton G.Deep learning.Nature2015;521:436-44

[11] Lundervold AS.An overview of deep learning in medical imaging focusing on MRI.Z Med Phys2019;29:102-27

[12] Liu X,Kale AU.A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis.Lancet Digit Health2019;1:e271-97

[13] Chea P.Current applications and future directions of deep learning in musculoskeletal radiology.Skeletal Radiol2020;49:183-97

[14] Jamaludin A, Lootus M, Kadir T, et al; Genodisc Consortium. ISSLS PRIZE IN BIOENGINEERING SCIENCE 2017: automation of reading of radiological features from magnetic resonance images (MRIs) of the lumbar spine without human intervention is comparable with an expert radiologist. Eur Spine J. 2017;26:1374-83.

[15] Lopez CD,Lombardi JM.Artificial learning and machine learning applications in spine surgery: a systematic review.Global Spine J2022;12:1561-72 PMCID:PMC9393994

[16] Page MJ,Bossuyt PM.The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.Int J Surg2021;88:105906

[17] Cabitza F.The IJMEDI checklist for assessment of medical AI.Int J Med Inform.2021;153:

[18] Chae DS,Park SJ,Won C.Decentralized convolutional neural network for evaluating spinal deformity with spinopelvic parameters.Comput Methods Programs Biomed2020;197:105699

[19] Wu H,Rasoulinejad P.Automated comprehensive adolescent idiopathic scoliosis assessment using MVC-Net.Med Image Anal2018;48:1-11

[20] Wang L,Leung S,Chen B.Accurate automated Cobb angles estimation using multi-view extrapolation net.Med Image Anal2019;58:101542

[21] Zhang K,Guo C.MPF-net: an effective framework for automated cobb angle estimation.Med Image Anal2022;75:102277

[22] Zerouali M,Benbakoura M,Champsaur P.Automatic deep learning-based assessment of spinopelvic coronal and sagittal alignment.Diagn Interv Imaging2023;104:343-50

[23] Korez R,Vrtovec T.A deep learning tool for fully automated measurements of sagittal spinopelvic balance from X-ray images: performance evaluation.Eur Spine J2020;29:2295-305

[24] Kim YT,Kim YJ,Kim KG.Automatic spine segmentation and parameter measurement for radiological analysis of whole-spine lateral radiographs using deep learning and computer vision.J Digit Imaging2023;36:1447-59 PMCID:PMC10406753

[25] Yeh YC,Huang YJ,Tsai TT.Deep learning approach for automatic landmark detection and alignment analysis in whole-spine lateral radiographs.Sci Rep2021;11:7618 PMCID:PMC8027006

[26] Orosz LD,Jazini E.Novel artificial intelligence algorithm: an accurate and independent measure of spinopelvic parameters.J Neurosurg Spine2022;37:893-901

[27] Gami P,Kannappan S.Semiautomated intraoperative measurement of Cobb angle and coronal C7 plumb line using deep learning and computer vision for scoliosis correction: a feasibility study.J Neurosurg Spine2022;37:713-21

[28] Schwartz JT,Tang P.Deep learning automates measurement of spinopelvic parameters on lateral lumbar radiographs.Spine2021;46:E671-8

[29] Aubert B,Cresson T,de Guise JA.Toward automated 3D spine reconstruction from biplanar radiographs using CNN for statistical spine model fitting.IEEE Trans Med Imaging2019;38:2796-806

[30] Nguyen TP,Yoo YJ,Yoon J.Intelligent evaluation of global spinal alignment by a decentralized convolutional neural network.J Digit Imaging2022;35:213-25 PMCID:PMC8921409

[31] Galbusera F,Wilke HJ.Fully automated radiological analysis of spinal disorders and deformities: a deep learning approach.Eur Spine J2019;28:951-60

[32] Zhang T,Lu Q,Diwan A.A novel tool to provide predictable alignment data irrespective of source and image quality acquired on mobile phones: what engineers can offer clinicians.Eur Spine J2020;29:387-95

[33] Horng MH,Fu MJ,Sun YN.Cobb angle measurement of spine from X-ray images using convolutional neural network.Comput Math Methods Med2019;2019:6357171 PMCID:PMC6399566

[34] Sun H,Bailey C,Yin Y.Direct estimation of spinal Cobb angles by structured multi-output regression. In: Niethammer M, Styner M, Aylward S, Zhu H, Oguz I, Yap P, Shen D, editors. Information processing in medical imaging. Cham: Springer International Publishing; 2017. pp. 529-40.

[35] Weng CH,Huang YJ.Artificial intelligence for automatic measurement of sagittal vertical axis using ResUNet framework.J Clin Med2019;8:1826 PMCID:PMC6912675

[36] H A.Automatic quantification of spinal curvature in scoliotic radiograph using image processing.J Med Syst2012;36:1943-51

[37] Zhang J,Le LH,Raso JV.Automatic Cobb measurement of scoliosis based on fuzzy Hough Transform with vertebral shape prior.J Digit Imaging2009;22:463-72 PMCID:PMC3043716

[38] Sounderajah V, Ashrafian H, Golub RM, et al; STARD-AI Steering Committee. Developing a reporting guideline for artificial intelligence-centred diagnostic test accuracy studies: the STARD-AI protocol. BMJ Open 2021;11:e047709. PMCID:PMC8240576

[39] Liu X, Cruz Rivera S, Moher D, Calvert MJ, Denniston AK; SPIRIT-AI and CONSORT-AI Working Group. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Lancet Digit Health. 2020;2:e537-48. PMCID:PMC8183333

[40] Mongan J,Kahn CE Jr.Checklist for artificial intelligence in medical imaging (CLAIM): a guide for authors and reviewers.Radiol Artif Intell2020;2:e200029 PMCID:PMC8017414

[41] Willemink MJ,Hardell C.Preparing medical imaging data for machine learning.Radiology2020;295:4-15 PMCID:PMC7104701

[42] Ghaednia H,Sauder N.Deep learning in spine surgery.Semin Spine Surg2021;33:100876

[43] Kelly CJ,Suleyman M,King D.Key challenges for delivering clinical impact with artificial intelligence.BMC Med2019;17:195 PMCID:PMC6821018

[44] Cho BH,Cheung ZB.Automated measurement of lumbar lordosis on radiographs using machine learning and computer vision.Global Spine J2020;10:611-8 PMCID:PMC7359685

[45] Burns JE,Chalhoub D,Summers RM.A machine learning algorithm to estimate sarcopenia on abdominal CT.Acad Radiol2020;27:311-20

[46] Yang J,Fan H.Development and validation of deep learning algorithms for scoliosis screening using back images.Commun Biol2019;2:390 PMCID:PMC6814825

[47] Hu B,Ning X.Using a deep learning network to recognise low back pain in static standing.Ergonomics2018;61:1374-81

[48] Staartjes VE,Vandertop WP.Deep learning-based preoperative predictive analytics for patient-reported outcomes following lumbar discectomy: feasibility of center-specific modeling.Spine J2019;19:853-61

[49] Hines AL,Jiang HJ.Conditions with the largest number of adult hospital readmissions by payer, 2011. In: Healthcare Cost and Utilization Project (HCUP) Statistical Briefs.

[50] Valenzuela JG, Cirillo Totera JI, Turkieltaub DH, Echaurren CV, Álvarez Lemos FL, Arriagada Ramos FI. Spino-pelvic radiological parameters: comparison of measurements obtained by radiologists using the traditional method versus spine surgeons using a semi-automated software (Surgimap).Acta Radiol Open2023;12:20584601231177404 PMCID:PMC10201147

[51] Khalifé M, Lafage R, Daniels AH, et al; International Spine Study Group. Assessing abnormal proximal junctional angles in adult spinal deformity: a normative data approach to define proximal junctional kyphosis. Spine. 2025;50:103-9.