Background & Aims
Studies available in literature on the use of AI for the diagnosis of OFP disorders have used different patient selection criteria, disease classifications, input data, and outcome measures for evaluation. Consequently, the performance of the various AI models vary across different studies. To the best of our knowledge, there has been no systematic review till date that summarizes such findings. Therefore, this study aimed to systematically review the current literature on the use of sematic-based algorithms in AI for the diagnosis of various OFP disorders, evaluate the quality of these studies and assess the performance of existing AI models.
Methods
The systematic review and meta-analysis was conducted and reported according to the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) 2020 guidelines. PubMed, Embase, Web of Science and Science Direct databases were searched for relevant articles from database inception to January 2024. Studies that used semantic-based AI algorithms to diagnose at least one type of chronic OFP disorder and those that assessed the performance of these algorithms were included. Studies on OFP disorders associated with odontogenic causes of pain, as well as studies on radiomic-based AI algorithms, abstracts without full texts, case reports, case series, book chapters and review papers were excluded. The risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) was used to evaluate the quality of studies for which meta-analysis was performed.
Results
11 eligible articles that used semantic-based AI algorithms for the diagnosis of OFP disorders were included; they were retrospective studies and observational studies. The types of chronic OFP disorders were classified according to the International Classification of Orofacial pain (ICOP).
6 studies were subjected to a meta-analysis for diagnostic performance. According to GRADE, the certainty of evidence was very low. The accuracy of the AI models for the diagnosis of myofascial orofacial pain was 0.91-0.97, and the pooled accuracy was 0.94 (95% CI 0.92-0.96), I2=0%. For temporomandibular joint pain, the diagnostic accuracy of the AI models was 0.71-0.99 (95% CI 0.87-0.98), I2=88.75%. The performance of the AI models for the diagnosis of migraines was 0.86-1.00, and the pooled accuracy was 0.97 (95% CI 0.75-1.00), I2=92%.
Conclusions
Numerous AI models have been developed for diagnosis of chronic OFP disorders and may serve as an additional tool to aid in increasing diagnostic accuracy. The results of this study suggest that semantic-based AI algorithms developed for the diagnosis of chronic OFP disorders can be used as a decision support tool, especially in the outpatient clinic setting. However, a high risk of bias in patient selection was noted due to the inclusion of observational studies. In addition, a significant heterogeneity was observed among the studies included for meta-analysis of diagnostic accuracy. Therefore, the certainty of evidence was deemed as very low. Future research of higher quality is strongly recommended.
References
1.Raja SN, Carr DB, Cohen M, Finnerup NB, Flor H, Gibson S, Keefe FJ, Mogil JS, Ringkamp M, Sluka KA, Song XJ, Stevens B, Sullivan MD, Tutelman PR, Ushida T, Vader K. The revised International Association for the Study of Pain definition of pain: concepts, challenges, and compromises. Pain. 2020 Sep 1;161(9):1976-1982. doi: 10.1097/j.pain.0000000000001939. PMID: 32694387; PMCID: PMC7680716.
2.International Classification of Orofacial Pain, 1st edition (ICOP). Cephalalgia. 2020 Feb;40(2):129-221. doi: 10.1177/0333102419893823. PMID: 32103673.
3.Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, Leeflang MM, Sterne JA, Bossuyt PM; QUADAS-2 Group. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011 Oct 18;155(8):529-36. doi: 10.7326/0003-4819-155-8-201110180-00009. PMID: 22007046.
4.Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, Schünemann HJ; GRADE Working Group. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008 Apr 26;336(7650):924-6. doi: 10.1136/bmj.39489.470347.AD. PMID: 18436948; PMCID: PMC2335261.
5.Bas B, Ozgonenel O, Ozden B, Bekcioglu B, Bulut E, Kurt M. Use of artificial neural network in differentiation of subgroups of temporomandibular internal derangements: a preliminary study. J Oral Maxillofac Surg. 2012 Jan;70(1):51-9. doi: 10.1016/j.joms.2011.03.069. Epub 2011 Jul 29. PMID: 21802818.
6.Diniz de Lima E, Souza Paulino JA, Lira de Farias Freitas AP, Viana Ferreira JE, Barbosa JDS, Bezerra Silva DF, Bento PM, Araújo Maia Amorim AM, Melo DP. Artificial intelligence and infrared thermography as auxiliary tools in the diagnosis of temporomandibular disorder. Dentomaxillofac Radiol. 2022 Feb 1;51(2):20210318. doi: 10.1259/dmfr.20210318. Epub 2021 Oct 6. PMID: 34613829; PMCID: PMC8802706.
7.W. Zou, B. Mao, Z. M. Fadlullah and K. Qi, “Assisting in Diagnosis of Temporomandibular Disorders: A Deep Learning Approach,” in IEEE Access, vol. 10, pp. 124076-124082, 2022, doi: 10.1109/ACCESS.2022.3224055.
8.M. Ghodsi, S. Sanei, Y. Hicks, T. Lee and S. Dunne, “Detection of Temporomandibular Disorder from Facial Pattern,” 2007 15th International Conference on Digital Signal Processing, Cardiff, UK, 2007, pp. 151-154, doi: 10.1109/ICDSP.2007.4288541.
9.Paulina Vistoso Monreal A, Veas N, Clark G. An artificially intelligent (or algorithm-enhanced) electronic medical record in orofacial pain. Jpn Dent Sci Rev. 2021 Nov;57:242-249. doi: 10.1016/j.jdsr.2021.11.001. Epub 2021 Nov 20. PMID: 34849180; PMCID: PMC8608603.
10.Limonadi FM, McCartney S, Burchiel KJ. Design of an artificial neural network for diagnosis of facial pain syndromes. Stereotact Funct Neurosurg. 2006;84(5-6):212-20. doi: 10.1159/000095167. Epub 2006 Aug 18. PMID: 16921257.
11.Cowan RP, Rapoport AM, Blythe J, Rothrock J, Knievel K, Peretz AM, Ekpo E, Sanjanwala BM, Woldeamanuel YW. Diagnostic accuracy of an artificial intelligence online engine in migraine: A multi-center study. Headache. 2022 Jul;62(7):870-882. doi: 10.1111/head.14324. Epub 2022 Jun 3. PMID: 35657603; PMCID: PMC9378575.
12.Khayamnia M, Yazdchi M, Heidari A, Foroughipour M. Diagnosis of Common Headaches Using Hybrid Expert-Based Systems. J Med Signals Sens. 2019 Aug 29;9(3):174-180. doi: 10.4103/jmss.JMSS_47_18. PMID: 31544057; PMCID: PMC6743243.
13.Kwon J, Lee H, Cho S, Chung CS, Lee MJ, Park H. Machine learning-based automated classification of headache disorders using patient-reported questionnaires. Sci Rep. 2020 Aug 20;10(1):14062. doi: 10.1038/s41598-020-70992-1. PMID: 32820214; PMCID: PMC7441379.
14.McCartney S, Weltin M, Burchiel KJ. Use of an artificial neural network for diagnosis of facial pain syndromes: an update. Stereotact Funct Neurosurg. 2014;92(1):44-52. doi: 10.1159/000353188. Epub 2013 Nov 8. PMID: 24217113.
15.Nocera L, Vistoso A, Yoshida Y, Abe Y, Nwoji C, Clark GT. Building an Automated Orofacial Pain, Headache and Temporomandibular Disorder Diagnosis System. AMIA Annu Symp Proc. 2021 Jan 25;2020:943-952. PMID: 33936470; PMCID: PMC8075456.
Presenting Author
Eunice Lua
Poster Authors
Topics
- Assessment and Diagnosis