Photograph-Based Visual Classification of Gingival Recession Using Multimodal Large Language Models: A Comparative Study

Abstract

Purpose: This study aimed to evaluate the photograph-based visual classification performance of contemporary multimodal large language models for gingival recession defects according to Miller’s classification. Material and Methods: A dataset of 100 standardized intraoral photographs showing gingival recession was used. Five periodontists independently classified each photograph according to Miller’s classification using only the visual information available in the image. In cases of disagreement, a panel discussion was performed to establish a final periodontist visual consensus classification, which served as the reference comparator. Then photographs were submitted once to ChatGPT 5, Gemini 2.5 Pro, Grok 3, and Copilot GPT-5. Model outputs were compared with the periodontist panel consensus classification using overall accuracy, class-wise precision, sensitivity/recall, F1-score, Cohen’s kappa, and weighted kappa. Results: The periodontist panel showed almost perfect inter-rater agreement (Fleiss’ κ = 0.885; bootstrap 95% CI: 0.827–0.934). Overall classification accuracy differed significantly among the evaluated models (p < 0.001). Gemini showed the highest accuracy, correctly classifying 49% of photographs, followed by ChatGPT 5 with 32%, Grok 3 with 26%, and Copilot GPT-5 with 24%. Conclusion: Under standardized single-prompt and photograph-only conditions, the evaluated multimodal large language models showed limited and inconsistent agreement with the periodontist panel visual consensus classification.

Keywords

• Artificial Intelligence
• Gingival Recession
• Large Language Models
• Periodontics.

Supporting Institution

None

Ethical Statement

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Izmir Tinaztepe University (Protocol No: MUKCE2025/78, August 28, 2025). Informed consent was obtained from all participants who agreed to the use of their intraoral photographs for the study.

Thanks

None

Photograph-Based Visual Classification of Gingival Recession Using Multimodal Large Language Models: A Comparative Study

Abstract

References

1. Pini Prato G, Baccetti T, Giorgetti R, Agudio G, Cortellini P. Mucogingival interceptive surgery of buccally-erupted premolars in patients scheduled for orthodontic treatment. II. Surgically treated versus nonsurgically treated cases. J Periodontol. 2000;71(2):182–7.
2. Wagner TP, Costa RSA, Rios FS, Moura MS, Maltz M, Jardim JJ, et al. Gingival recession and oral health‐related quality of life: a population‐based cross‐sectional study in Brazil. Community Dent Oral Epidemiol. 2016;44(4):390–9.
3. Teixeira DNR, Zeola LF, Machado AC, Gomes RR, Souza PG, Mendes DC, et al. Relationship between noncarious cervical lesions, cervical dentin hypersensitivity, gingival recession, and associated risk factors: A cross-sectional study. J Dent. 2018;76:93–7.
4. Chambrone L, Pannuti CM, Tu Y, Chambrone LA. Evidence‐Based Periodontal Plastic Surgery. II. An Individual Data Meta‐Analysis for Evaluating Factors in Achieving Complete Root Coverage. J Periodontol. 2012 Apr;83(4):477–90.
5. Cairo F, Nieri M, Cincinelli S, Mervelt J, Pagliaro U. The interproximal clinical attachment level to classify gingival recessions and predict root coverage outcomes: an explorative and reliability study. J Clin Periodontol. 2011;38(7):661–6.
6. Miller PD. A classification of marginal tissue recession. Int J Periodontics Restorative Dent. 1985;5(2):8–13.
7. Pini-Prato G. The Miller classification of gingival recession: limits and drawbacks. J Clin Periodontol. 2011;38(3):243–5.
8. Mao K, Thu KM, Hung KF, Yu OY, Hsung RTC, Lam WYH. Artificial Intelligence in Detecting Periodontal Disease From Intraoral Photographs: A Systematic Review. Int Dent J. 2025;75(5):100883.

9. Chatzopoulos GS, Koidou VP, Tsalikis L, Kaklamanos EG. Large language models in periodontology: Assessing their performance in clinically relevant questions. J Prosthet Dent. 2025;134(6):2328–36.
10. Tastan Eroglu Z, Babayigit O, Ozkan Sen D, Ucan Yarkac F. Performance of ChatGPT in classifying periodontitis according to the 2018 classification of periodontal diseases. Clin Oral Investig. 2024;28(7):407.
11. Giannakopoulos K, Kavadella A, Aaqel Salim A, Stamatopoulos V, Kaklamanos EG. Evaluation of the Performance of Generative AI Large Language Models ChatGPT, Google Bard, and Microsoft Bing Chat in Supporting Evidence-Based Dentistry: Comparative Mixed Methods Study. J Med Internet Res. 2023;25:e51580.
12. Islam R, Moushi OM. GPT-4o: The cutting-edge advancement in multimodal LLM. In: Arai K, editor. Intelligent Computing: Proceedings of the Computing Conference. Cham: Springer Nature Switzerland; 2025. p. 47–60.
13. Rewthamrongsris P, Burapacheep J, Phattarataratip E, Kulthanaamondhita P, Tichy A, Schwendicke F, et al. Image-Based Diagnostic Performance of LLMs vs CNNs for Oral Lichen Planus: Example-Guided and Differential Diagnosis. Int Dent J. 2025;75(4):100848.
14. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–74.
15. Mahajan A. Mahajan’s Modification of the Miller’s Classification for Gingival Recession. Dent Hypotheses. 2010;1(2):45–50.
16. Bertl K, Ruckenbauer D, Müller-Kern M, Durstberger G, Lettner S, Bruckmann C, et al. Inter- and intra-observer agreement on Miller’s classification of gingival tissue recessions. Odontology. 2015;103(3):292–300.
17. Fageeh HI, Fageeh HN, Bhati AK, Thubab AY, Sharrahi HMH, Aljabri YS, et al. Assessing the Reliability of Miller’s Classification and Cairo’s Classification in Classifying Gingival Recession Defects: A Comparison Study. Medicina (Kaunas). 2024;60(2):205.
18. Sarlati F, Moghaddas O, Shabahangfar R, Safari S, Valaei N. Inter- and intra-examiner agreement of three classification systems of gingival recession. J Adv Periodontol Implant Dent. 2019;11(1):1–6.
19. Suárez A, Díaz‐Flores García V, Algar J, Gómez Sánchez M, Llorente de Pedro M, Freire Y. Unveiling the ChatGPT phenomenon: Evaluating the consistency and accuracy of endodontic question answers. Int Endod J. 2024;57(1):108–13.
20. Yılmaz M, Oduncuoğlu BF, Nişancı Yılmaz MN. Evaluation of patients’ perception of gingival recession, its impact on oral health-related quality of life, and acceptance of treatment plan. Acta Odontol Scand. 2020;78(6):454–62.
21. Yadav VS, Gumber B, Makker K, Gupta V, Tewari N, Khanduja P, et al. Global prevalence of gingival recession: A systematic review and meta‐analysis. Oral Dis. 2023;29(8):2993–3002.
22. Watanabe H, Uehara O, Morikawa T, Kojima T, Suga T, Toyofuku A, et al. Performance of large language models on image-based oral pathology questions from the Japanese National Dental Examination. J Dent Sci. 2026;21(2):879–87.
23. Fukuda H, Morishita M, Muraoka K, Yamaguchi S, Nakamura T, Yoshioka I, et al. Evaluating the image recognition capabilities of GPT-4V and Gemini Pro in the Japanese national dental examination. J Dent Sci. 2025;20(1):368–72.
24. Camlet A, Kusiak A, Ossowska A, Świetlik D. Advances in Periodontal Diagnostics: Application of MultiModal Language Models in Visual Interpretation of Panoramic Radiographs. Diagnostics. 2025;15(15):1851.
25. Mine Y, Taji T, Okazaki S, Takeda S, Peng TY, Shimoe S, et al. Analyzing the performance of multimodal large language models on visually-based questions in the Japanese National Examination for Dental Technicians. J Dent Sci. 2025;20(4):2460–6.
26. Miller PD. Miller Classification of Marginal Tissue Recession Revisited After 35 Years. Compend Contin Educ Dent. 2018;39(8):514–20.