Structuring of oral mucosa images with annotation of pathological lesions for machine vision algorithm training

   Oral mucosa pathologies (OMP) represent one of the most challenging groups of diseases to diagnose, characterized by polymorphic manifestations. The development and implementation of a digital machine vision algorithm for diagnosing oral mucosa diseases is a relevant direction for improving the quality and efficiency of diagnostics in modern dentistry.

   Aim: to conduct a prospective collection and structuring of images with annotation of oral mucosa pathological lesions, integrate the dataset into a digital machine vision model, and train a clinical decision support system.

   Material and methods: the study used: the United Medical Knowledge Base (UMKB), a Clinical Decision Support System (CDSS), patients with and without oral mucosa diseases, photography equipment, prospective images of pathological conditions and normal states, annotation of pathological lesions.

   Results: the conducted work demonstrated the system’s effectiveness in identifying pathological lesions from oral mucosa images at 84 %, and the accuracy of nosology identification based on the list of pathological signs ranged from 90 to 95 %. This confirms the potential to improve the diagnostic efficiency of oral mucosa diseases using an intelligent CDSS in clinical practice.

  Conclusions: a prospective collection and structuring of images was carried out, with annotation of files across 15 categories of oral mucosa lesions into a digital machine vision model. This was followed by the additional training of a CDSS, achieving an efficiency of 84 % and a testing accuracy for determining nosology of 90–95 %.

1. Berdnikova E., Osipova V. Innovative diagnostic methods in the complex assessment of the state of the oral mucosa // Innovation science. 2021. № 5.

2. Iskakova M.K., Zarkumova A.E., Nurmukhambetova G.K. Structure of morbidity of the multiple structure of the mouth of the mut // Vestnik KazNMU. 2017. № 3.

3. Shkarin V.V., Makedonova Yu.A., Poroshin A.V. Algorithm for providing medical care to patients with diseases of the oral mucosa. // Medical & pharmaceutical journal «Pulse». – 2023;25(1):49–53.

4. Margaryan E.G., Bledzhyants G.A., Kadzhoian A.G., Kurenkova Yu.S., Abdelrakhim M.T., Dlyaverovna D.K., Tan H., Shuan P., Turner A. The effectiveness of diagnosing oral mucosal diseases using a clinical decision support system. Medical alphabet. 2025;(20):94–99. (In Russ.)

5. Margaryan E.G., Kurenkova Yu.S., Lalayan K.V., Abdelrahim M.T., Kadzhoian A.G., Selifanova E.I., Budaichieva Z.S., Stepanov M.A., Bledzhyants G.A., Makeeva M.K. The necessity and demand of machine learning in treatment and diagnostics in dentistry. Medical alphabet. 2024;(18):121–126. (In Russ.)

6. Arsiwala-Scheppach LT, Chaurasia A, Müller A, Krois J, Schwendicke F. Machine Learning in Dentistry : A Scoping Review. J Clin Med. 2023 Jan 25;12(3):937. doi: 10.3390/jcm12030937. PMID: 36769585; PMCID: PMC9918184.

7. Kamath P, Kamath P, Saldanha SJR, Shetty TB, Rodrigues SJ, M M, Pai UY, Hegde PK, Bajantri P, Mukherjee S. A brief exploration of artificial intelligence in dental healthcare : a narrative review. F1000Res. 2024 Apr 8;13:37. doi: 10.12688/f1000research.140481.2. PMID: 40837849; PMCID: PMC12361878.

8. Mallineni SK, Sethi M, Punugoti D, Kotha SB, Alkhayal Z, Mubaraki S, Almotawah FN, Kotha SL, Sajja R, Nettam V, Thakare AA, Sakhamuri S. Artificial Intelligence in Dentistry : A Descriptive Review. Bioengineering (Basel). 2024 Dec 13;11(12):1267. doi: 10.3390/bioengineering11121267. PMID: 39768085; PMCID: PMC11673909.

9. Radwan-Oczko M, Bandosz K, Rojek Z, Owczarek-Drabińska JE. Clinical Study of Oral Mucosal Lesions in the Elderly-Prevalence and Distribution. Int J Environ Res Public Health. 2022 Mar 1;19(5):2853. doi: 10.3390/ijerph19052853. PMID: 35270543; PMCID: PMC8910280.

10. Surdu A, Budala DG, Luchian I, Foia LG, Botnariu GE, Scutariu MM. Using AI in Optimizing Oral and Dental Diagnoses - A Narrative Review. Diagnostics (Basel). 2024 Dec 13;14(24):2804. doi: 10.3390/diagnostics14242804. PMID: 39767164; PMCID: PMC11674583.

11. Tatano R, Berkels B, Ehrlich EE, Deserno TM, Fritz UB. Spatial agreement of demineralized areas in quantitative light-induced fluorescence images and digital photographs. Dentomaxillofac Radiol. 2018 Dec;47(8):20180099. doi: 10.1259/dmfr.20180099. Epub 2018 Jun 15. PMID: 29851354; PMCID: PMC6326398.

12. Yang EC, Vohra IS, Badaoui H, Schwarz RA, Cherry KD, Quang T, Jacob J, Lang A, Bass N, Rodriguez J, Williams MD, Vigneswaran N, Gillenwater AM, Richards-Kortum RR. Development of an integrated multimodal optical imaging system with real-time image analysis for the evaluation of oral premalignant lesions. J Biomed Opt. 2019 Feb;24(2):1–10. doi: 10.1117/1.JBO.24.2.025003. PMID: 30793567; PMCID: PMC6383051.