Clinical deployment of machine learning models in craniofacial surgery: considerations for adoption and implementation

Abstract

The volume and complexity of clinical data are growing rapidly. The potential for artificial intelligence (AI) and machine learning (ML) to significantly impact plastic and craniofacial surgery is immense. This manuscript reviews the overall landscape of AI in craniofacial surgery, highlighting the scarcity of prospective and clinically translated models. It examines the numerous clinical promises and challenges associated with AI, such as the lack of robust legislation and structured frameworks for its integration into medicine. Clinical translation considerations are discussed, including the importance of ensuring clinical utility for real-world use. Finally, this commentary brings forward how clinicians can build trust and sustainability toward model-driven clinical care.

Keywords

Artificial intelligence / machine learning / craniofacial surgery / clinical translation

References

[1] Topol EJ.High-performance medicine: the convergence of human and artificial intelligence.Nat Med2019;25:44-56

[2] Hernandez-Boussard T,Ioannidis JPA.MINIMAR (MINimum Information for Medical AI Reporting): developing reporting standards for artificial intelligence in health care.J Am Med Inform Assoc2020;27:2011-5 PMCID:PMC7727333

[3] Schiff GD.Can electronic clinical documentation help prevent diagnostic errors?.N Engl J Med2010;362:1066-9

[4] Mak ML,Phillips J.Prevalence of machine learning in craniofacial surgery.J Craniofac Surg2020;31:898-903

[5] Jarvis T,Rebecca AM.Artificial intelligence in plastic surgery: current applications, future directions, and ethical implications.Plast Reconstr Surg Glob Open2020;8:e3200 PMCID:PMC7647513

[6] Kanevsky J,Gaster R,Lin S.Big data and machine learning in plastic surgery: a new frontier in surgical innovation.Plast Reconstr Surg2016;137:890e-7e

[7] Zhu VZ,Au AF.Promise and limitations of big data research in plastic surgery.Ann Plast Surg2016;76:453-8

[8] Kim YJ,Nasser JS.Implementing precision medicine and artificial intelligence in plastic surgery: concepts and future prospects.Plast Reconstr Surg Glob Open2019;7:e2113 PMCID:PMC6467615

[9] Murphy DC.Artificial intelligence in plastic surgery: what is it? Where are we now? What is on the horizon?.Ann R Coll Surg Engl2020;102:577-80 PMCID:PMC7538735

[10] Dhillon H,Dhingra K.Current applications of artificial intelligence in cleft care: a scoping review.Front Med2021;8:676490 PMCID:PMC8355556

[11] Wu J,Shapiro LG.Learning to rank the severity of unrepaired cleft lip nasal deformity on 3D mesh data.Proc IAPR Int Conf Pattern Recogn2014;2014:460-4 PMCID:PMC4280842

[12] Maier A,Bocklet T.Automatic detection of articulation disorders in children with cleft lip and palate.J Acoust Soc Am2009;126:2589-602

[13] Bouletreau P,Ibrahim B,Sigaux N.Artificial intelligence: applications in orthognathic surgery.J Stomatol Oral Maxillofac Surg2019;120:347-54

[14] Patcas R,Volokitin A,Rothe R.Applying artificial intelligence to assess the impact of orthognathic treatment on facial attractiveness and estimated age.Int J Oral Maxillofac Surg2019;48:77-83

[15] Du W,Liu Y,Tai Y.Machine learning-based decision support system for orthognathic diagnosis and treatment planning.BMC Oral Health2024;24:286 PMCID:PMC10902963

[16] Qamar A,Barve R.Artificial intelligence applications in diagnosing and managing non-syndromic craniosynostosis: a comprehensive review.Cureus2023;15:e45318 PMCID:PMC10577020

[17] Mashouri P,Phillips J. 3D photography based neural network craniosynostosis triaging system. In: Proceedings of the Machine Learning for Health NeurIPS Workshop PMLR; 2020. pp.226-37. Available from: https://proceedings.mlr.press/v136/mashouri20a.html. [Last accessed on 5 Dec 2024]

[18] Pham TD,Coulthard P.A review on artificial intelligence for the diagnosis of fractures in facial trauma imaging.Front Artif Intell2023;6:1278529 PMCID:PMC10797131

[19] Wang HC,Yan JL.Artificial intelligence model trained with sparse data to detect facial and cranial bone fractures from head CT.J Digit Imaging2023;36:1408-18 PMCID:PMC10407005

[20] Moon G,Kim WJ,Sung KY.Very fast, high-resolution aggregation 3D detection CAM to quickly and accurately find facial fracture areas.Comput Methods Programs Biomed2024;256:108379

[21] Moon G,Kim W,Jeong Y.Computer aided facial bone fracture diagnosis (CA-FBFD) system based on object detection model.IEEE Access2022;10:79061-70

[22] Wang X,Tong Y.Detection and classification of mandibular fracture on CT scan using deep convolutional neural network.Clin Oral Investig2022;26:4593-601

[23] Roy M,Shah PS.Effectiveness and safety of the use of gracilis muscle for dynamic smile restoration in facial paralysis: a systematic review and meta-analysis.J Plast Reconstr Aesthet Surg2019;72:1254-64

[24] Boonipat T,Lin J,Mardini S.Using artificial intelligence to measure facial expression following facial reanimation surgery.Plast Reconstr Surg2020;146:1147-50

[25] Dusseldorp JR,van Veen MM,Jowett N.Automated spontaneity assessment after smile reanimation: a machine learning approach.Plast Reconstr Surg2022;149:1393-402

[26] Sendak MP,Brajer N.Presenting machine learning model information to clinical end users with model facts labels.NPJ Digit Med2020;3:41 PMCID:PMC7090057

[27] Castelvecchi D.Can we open the black box of AI?.Nature2016;538:20-3

[28] Knight W. The dark secret at the heart of AI. Available from: https://www.technologyreview.com/2017/04/11/5113/the-dark-secret-at-the-heart-of-ai/. [Last accessed on 5 Dec 2024]

[29] Buolamwini J. Gender shades: intersectional accuracy disparities in commercial gender classification. In: Proceedings of the 1st Conference on Fairness, Accountability and Transparency. PMLR; 2018. pp. 77-91. Available from: https://proceedings.mlr.press/v81/buolamwini18a.html?mod=article_inline&ref=akusion-ci-shi-dai-bizinesumedeia. [Last accessed on 5 Dec 2024]

[30] Kish LJ.Unpatients-why patients should own their medical data.Nat Biotechnol2015;33:921-4

[31] Murdoch B.Privacy and artificial intelligence: challenges for protecting health information in a new era.BMC Med Ethics2021;22:122 PMCID:PMC8442400

[32] Brundage M,Clark J. The malicious use of artificial intelligence: forecasting, prevention, and mitigation. ArXiv. [Preprint.] Dec 1, 2024 [accessed 2024 Dec 5]. Available from: https://doi.org/10.48550/arXiv.1802.07228.

[33] Watson J,Clancy SM.Overcoming barriers to the adoption and implementation of predictive modeling and machine learning in clinical care: what can we learn from US academic medical centers?.JAMIA Open2020;3:167-72 PMCID:PMC7382631

[34] Salehinejad H,Ditkofsky N.A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography.Sci Rep2021;11:17051 PMCID:PMC8382750

[35] Mechelli A.From models to tools: clinical translation of machine learning studies in psychosis.NPJ Schizophr2020;6:4 PMCID:PMC7021680

[36] Patel R,Callard F.What proportion of patients with psychosis is willing to take part in research? A mental health electronic case register analysis.BMJ Open2017;7:e013113 PMCID:PMC5353309

[37] Collaborative learning without sharing data.Nat Mach Intell2021;3:459

[38] Antoniadi AM,Guendouz Y.Current challenges and future opportunities for XAI in machine learning-based clinical decision support systems: a systematic review.Appl Sci2021;11:5088

[39] Bussone A,O’Sullivan D.The role of explanations on trust and reliance in clinical decision support systems. In: 2015 International Conference on Healthcare Informatics; 2015 Oct 21-23; Dallas, USA. IEEE; 2015. pp. 160-9.

[40] Amann J, Blasimme A, Vayena E, Frey D, Madai VI; Precise4Q consortium. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med Inform Decis Mak 2020;20:310. PMCID:PMC7706019

[41] Barredo Arrieta A,Del Ser J.Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI.Inform Fusion2020;58:82-115

[42] Linardatos P,Kotsiantis S.Explainable AI: a review of machine learning interpretability methods.Entropy2020;23:18 PMCID:PMC7824368

[43] Accuracy. Artificial Intelligence Episode 1 - overview of leading artificial intelligence clusters around the globe. Available from: https://www.accuracy.com/perspectives/overview-leading-artificial-intelligence-clusters-around-globe. [Last accessed on 5 Dec 2024]

[44] McKinsey and Company. Transforming healthcare with AI: the impact on the workforce and organizations. Available from: https://www.mckinsey.com/industries/healthcare-systems-and-services/our-insights/transforming-healthcare-with-ai. [Last accessed on 5 Dec 2024]