Redefining precision: the current and future roles of artificial intelligence in spine surgery

Abstract

The integration of artificial intelligence (AI) into spine surgery presents a transformative approach to preoperative and postoperative care paradigms. This paper explores the application of AI within spine surgery, focusing on diagnostic and predictive applications. AI-driven analysis of radiographic images, facilitated by machine learning (ML) algorithms such as convolutional neural networks (CNNs), potentially promises enhanced accuracy in identifying spinal pathologies and deformities; by combining these techniques with patient-specific data, predictive modeling can guide and inform diagnosis, prognosis, surgery selection, and treatment. Postoperatively, AI can leverage data from digital wearable technology to enhance the quantity and quality of outcome measures surgeons use to define and understand treatment success or failure. Still, challenges such as internal and external validation of AI models remain pertinent. Future directions include incorporating continuous variables from digital biomarkers and standardizing reporting metrics for AI studies in spine surgery. As AI continues to evolve, transparent validation frameworks and adherence to reporting guidelines will be crucial for its successful integration into clinical practice.

Keywords

Artificial intelligence / adult spinal deformity / radiographic imaging / machine learning / predictive modeling / objective outcomes

References

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

[2] Ruiz-España S,Moratal D.Semiautomatic computer-aided classification of degenerative lumbar spine disease in magnetic resonance imaging.Comput Biol Med2015;62:196-205

[3] 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.

[4] Pfirrmann CW,Zanetti M,Boos N.Magnetic resonance classification of lumbar intervertebral disc degeneration.Spine2001;26:1873-8

[5] Johnson PM.Conditional generative adversarial network for 3D rigid-body motion correction in MRI.Magn Reson Med2019;82:901-10

[6] Galbusera F,Casaroli G.Generative models: an upcoming innovation in musculoskeletal radiology? A preliminary test in spine imaging.Eur Radiol Exp2018;2:29 PMCID:PMC6207611

[7] Charles YP,Ntilikina Y.Artificial intelligence and treatment algorithms in spine surgery.Orthop Traumatol Surg Res2023;109:103456

[8] 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.

[9] Zhang J,Lv L.Computer-aided cobb measurement based on automatic detection of vertebral slopes using deep neural network.Int J Biomed Imaging2017;2017:9083916 PMCID:PMC5651147

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

[11] 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

[12] Pellisé F, Serra-Burriel M, Smith JS, et al; International Spine Study Group, European Spine Study Group. Development and validation of risk stratification models for adult spinal deformity surgery. J Neurosurg Spine 2019;31:587-99.

[13] Ames CP, Smith JS, Pellisé F, et al; European Spine Study Group, International Spine Study Group. Artificial intelligence based hierarchical clustering of patient types and intervention categories in adult spinal deformity surgery: towards a new classification scheme that predicts quality and value. Spine 2019;44:915-26.

[14] Lee MJ,Hamilton D.Predicting surgical site infection after spine surgery: a validated model using a prospective surgical registry.Spine J2014;14:2112-7

[15] Scheer JK, Smith JS, Schwab F, et al; International Spine Study Group. Development of a preoperative predictive model for major complications following adult spinal deformity surgery. J Neurosurg Spine 2017;26:736-43.

[16] Durand WM,Daniels AH.Predictive modeling for blood transfusion after adult spinal deformity surgery: a tree-based machine learning approach.Spine2018;43:1058-66

[17] Wang SK,Li ZE.Development and external validation of a predictive model for prolonged length of hospital stay in elderly patients undergoing lumbar fusion surgery: comparison of three predictive models.Eur Spine J2024;33:1044-54

[18] Sebastian A,Alvi MA.Assessing the performance of national surgical quality improvement program surgical risk calculator in elective spine surgery: insights from patients undergoing single-level posterior lumbar fusion.World Neurosurg2019;126:e323-9

[19] Goyal A,Kerezoudis P,Storlie C.Can machine learning algorithms accurately predict discharge to nonhome facility and early unplanned readmissions following spinal fusion? Analysis of a national surgical registry.J Neurosurg Spine2019;31:568-78

[20] Broda A,Turcotte J.Development of a risk prediction model with improved clinical utility in elective cervical and lumbar spine surgery.Spine2020;45:E542-51

[21] Churruca K,Ellis LA.Patient-reported outcome measures (PROMs): a review of generic and condition-specific measures and a discussion of trends and issues.Health Expect2021;24:1015-24 PMCID:PMC8369118

[22] Dansie EJ.Assessment of patients with chronic pain.Br J Anaesth2013;111:19-25 PMCID:PMC3841375

[23] Mobbs RJ.From the subjective to the objective era of outcomes analysis: how the tools we use to measure outcomes must change to be reflective of the pathologies we treat in spinal surgery.J Spine Surg2021;7:456-7 PMCID:PMC8511557

[24] Ahmad HS,Basil GW.Developing a prediction model for identification of distinct perioperative clinical stages in spine surgery with smartphone-based mobility data.Neurosurgery2022;90:588-96

[25] Chauhan D,Subtirelu R.Defining the minimal clinically important difference in smartphone-based mobility after spine surgery: correlation of survey questionnaire to mobility data.J Neurosurg Spine2023;39:427-37

[26] Boaro A,Reeder HT.Smartphone GPS signatures of patients undergoing spine surgery correlate with mobility and current gold standard outcome measures.J Neurosurg Spine2021;35:796-806 PMCID:PMC9012532

[27] Lewandrowski KU,Pereira MG.Polytomous rasch analyses of surgeons’ decision-making on choice of procedure in endoscopic lumbar spinal stenosis decompression surgeries.Int J Spine Surg2024;18:164-77 PMCID:PMC11287823

[28] Lorio M,Ferrara L.Paired comparison survey analyses utilizing rasch methodology of the relative difficulty and estimated work relative value units of CPT^® code 27279.Int J Spine Surg2016;10:40 PMCID:PMC5374999

[29] Ramspek CL,Dekker FW,van Diepen M.External validation of prognostic models: what, why, how, when and where?.Clin Kidney J2021;14:49-58 PMCID:PMC7857818

[30] 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

[31] 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

[32] Collins GS,Dhiman P.TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods.BMJ2024;385:e078378 PMCID:PMC11019967

[33] Collins GS, Reitsma JB, Altman DG, Moons KG; TRIPOD Group. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. The TRIPOD Group. Circulation 2015;131:211-9. PMCID:PMC4297220

[34] Zamanipoor Najafabadi AH,Dekker FW.TRIPOD statement: a preliminary pre-post analysis of reporting and methods of prediction models.BMJ Open2020;10:e041537 PMCID:PMC7511612