Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers

Seçkin Gül; İbrahim Halil Avcılar

Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers

Abstract

Aim: To compare the accuracy and comprehensibility of Turkish responses generated by ChatGPT and Gemini to patient-oriented questions regarding direct composite and ceramic laminate veneer selection. Materials and Methods: Ten patient-oriented questions covering esthetics, tooth structure preservation, longevity, color stability, repairability, parafunctional habits, and treatment selection were reviewed by a prosthodontist and a restorative dentistry specialist. On August 3, 2026, the questions were submitted in Turkish to ChatGPT (GPT-5.5 Instant; OpenAI) and Gemini (Gemini 3.5 Flash; Google) using the same standardized prompt in separate chat sessions. The first response generated by each model was anonymized, randomized, and independently rated by two blinded specialists for accuracy and comprehensibility using 4-point scales. Inter-rater agreement was assessed using ICC(2,1), and model scores were compared using the Wilcoxon signed-rank test. Results: Mean accuracy scores were 3.20 ± 0.48 for ChatGPT and 3.40 ± 0.66 for Gemini, with no significant difference between the models (p = 0.383). Mean comprehensibility scores were 3.35 ± 0.41 and 3.80 ± 0.35, respectively; the difference was also not statistically significant (p = 0.063). Inter-rater agreement was moderate for both accuracy (ICC = 0.528) and comprehensibility (ICC = 0.513). Conclusion: ChatGPT and Gemini showed comparable accuracy in responding to Turkish patient-oriented questions on direct composite and ceramic laminate veneers. Gemini achieved numerically higher comprehensibility scores, but without statistical significance. Both models may support basic patient education but should not replace clinical evaluation or professional decision-making.

Keywords

References

Gümüş S. Kompozit laminate uygulamalarına yönelik YouTubeTM videolarının değerlendirilmesi: Kalite ve yararlılık analizi. J Dent Fac Usak Univ. 2024;3(2):7–18.
Reddy Y, Dhaded S V. Direct composite veneers -restoring esthetics by procuring patient demand: a case report. Int J Dent Mater. 2019;01(03):93–6. doi:10.37983/IJDM.2019.1305
Mazzetti T, Collares K, Rodolfo B, da Rosa Rodolpho PA, van de Sande FH, Cenci MS. 10-year practice-based evaluation of ceramic and direct composite veneers. Dent Mater. 2022 May;38(5):898–906. doi:10.1016/j.dental.2022.03.007
Ferraris F. Direct or indirect veneers? Overview with a focus on ceramic restorations. Int J Esthet Dent. 2026;21:52.
Araujo E, Perdigão J. Anterior veneer restorations - An evidence-based minimal-intervention perspective. J Adhes Dent. 2021;23(2):91–110.
Najaf N, Dhaded S. A conservative and costeffective approach for enhancing aesthetics – a case report. Int J Sci Res. 2020;9(11):56–7.
Dunn J. Direct composite or bonded porcelain: a clinical choice for anterior aesthetics. J Calif Dent Assoc. 1994;22(4):73–6.
Merchant HA, Sharma S, Devanathan N, Hegde MN. Anterior Esthetic Restorations using direct composite restoration and all ceramic veneers - 2 case reports. Int J Adv Multidiscip Sci Res. 2018 May 15;1(3):9–14. doi:10.31426/ijamsr.2018.1.3.129

Guzman-Perez G, Jurado CA, Azpiazu-Flores F, Afrashtehfar KI, Tsujimoto A. Minimally Invasive Laminate Veneer Therapy for Maxillary Central Incisors. Medicina (B Aires). 2023 Mar 18;59(3):603. doi:10.3390/medicina59030603
Yıldırım Z, Gömeç Ö. Treatment approaches for traumatized anterior teeth with excessive tissue loss: three case reports. J Istanb Univ Fac Dent. 2017;51(2):54–60. doi:10.17096/jiufd.47682
Alghazzawi T. Clinical Survival Rate and Laboratory Failure of Dental Veneers: A Narrative Literature Review. J Funct Biomater. 2024 May 16;15(5):131. doi:10.3390/jfb15050131
Meade M, Dreyer C. Orthodontic treatment consent forms: A readability analysis. J Orthod. 2022 Mar 29;49(1):32–8. doi:10.1177/14653125211033301
Asa’ad F. Shared decision‐making (SDM) in dentistry: A concise narrative review. J Eval Clin Pract. 2019 Dec 28;25(6):1088–93. doi:10.1111/jep.13129
Tian S, Jin Q, Yeganova L, Lai PT, Zhu Q, Chen X, et al. Opportunities and challenges for ChatGPT and large language models in biomedicine and health. Brief Bioinform. 2023 Nov 22;25(1). doi:10.1093/bib/bbad493
Sözen E, Akpınar H, Yaman S. Comparative performance of large language models for patient-oriented support in dental trauma emergencies. BMC Oral Health. 2026 Mar 7;26(1):665. doi:10.1186/s12903-026-08037-8
Esen Ç, Gül M, Karayürek F. Clinical safety, content coverage, and patient-centered language of AI responses to periodontal complaint-based queries: a comparative study of four large language models. Odontol. 2026 Jul 7. doi:10.1007/s10266-026-01498-x
Şensoy E, Çıtırık M. Cross-linguistic evaluation of artificial intelligence chatbots: performance of ChatGPT-3.5, Copilot and Gemini in neuro-ophthalmologic evaluation in english and turkish. GTD. 2025 Aug 29;35(4):597–604. doi:10.54005/geneltip.1627508
Touati R, Sailer I, Marchand L, Ducret M, Strasding M. Communication tools and patient satisfaction: A scoping review. J Esthet Restor Dent. 2022 Jan 25;34(1):104–16. doi:10.1111/jerd.12854
Çolpak ED, Yılmaz D. Benchmarking different natural language processing models for their responses to queries on toothsupported fixed dental prostheses in terms of accuracy and consistency. ADO J Clin Sci. 2025 Sep 29;14(3):215–23. doi:10.54617/adoklinikbilimler.1698260
Hedén P, Adams WP, Maxwell P, Nava M, Scheflan M, Stan C. Aesthetic breast surgery: Consulting for the future-proposals for improving doctor-patient interactions. Aesthet Plast Surg. 2009 May 14;33(3):388–94. doi:10.1007/s00266-009-9329-9
Ekici Ö. Comparative evaluation of four large language models in turkish dentistry specialization exam. selcukdentj. 2025 Sep 19;12(4):6–10. doi:10.15311/selcukdentj.1674113
Kavasoglu N, Ertugrul OF, Kotan S, Hazar Y, Eratilla V. Artificial Intelligence-Assisted Wrist Radiography Analysis in Orthodontics: Classification of Maturation Stage. Applied Sciences. 2025 Oct 31;15(21):11681. doi:10.3390/app152111681
Othman AA, Sharqawi AJ, MohammedAziz AA, Ali WA, Alatiyyah AA, Mirah MA. Assessing the Accuracy and Completeness of AI-Generated Dental Responses: An Evaluation of the Chat-GPT Model. Healthcare. 2025 Aug 28;13(17):2144. doi:10.3390/healthcare13172144
Klein P, Spitznagel FA, Zembic A, Prott LS, Pieralli S, Bongaerts B, et al. Survival and Complication Rates of Feldspathic, Leucite‐Reinforced, Lithium Disilicate and Zirconia Ceramic Laminate Veneers: A Systematic Review and Meta‐Analysis. Journal of Esthetic and Restorative Dentistry. 2025 Mar 10;37(3):601–19. doi:10.1111/jerd.13351
Boztaş Demir G, Görgülü S. The Turkish Proficiency of ChatGPT and Gemini in Orthodontics: An Evaluation of the Accuracy, Completeness, and Readability of Responses to Patient Questions. Yeditepe Dental Journal. 2025;21(3):151–8. doi:10.5505/yeditepe.2025.15010
Alqutaibi AY, Alharbi AT, Alassaf MS, Alerwi AH, Alturki KN, Al-Sanabani FA, et al. Assessment of Quality and Readability of Online Patient-centered Information on Dental Veneers: An Infodemiological Study. J Contemp Dent Pract. 2025 Nov 25;26(11):1060–6. doi:10.5005/jp-journals-10024-3964
Esmailpour H, Rasaie V, Babaee Hemmati Y, Falahchai M. Performance of artificial intelligence chatbots in responding to the frequently asked questions of patients regarding dental prostheses. BMC Oral Health. 2025 Apr 15;25(1):574. doi:10.1186/s12903-025-05965-9
Bedel C, Zortuk Ö, Selvi F. Evaluation of readability indices of chatgpt-4 and google gemini in patient education about intracranial hemorrhages. Sci Healthc. 2025;27(4):14.
Bayram MA, Fincan AA, Gümüs AS, Diri B, Yıldırım S, and Aytas O Setting standards in turkish nlp: Tr-mmlu for large language model evaluation. arXiv preprint arXiv:2501.00593, 2024.
Dundar Sari MB, Sezer B. Comparative performance evaluation of ChatGPT-4 Omni and Gemini Advanced in the Turkish Dentistry Specialization Exam. BMC Med Educ. 2026 Jan 17;26(1):251. doi:10.1186/s12909-026-08621-0
Sismanoglu S, Capan BS. Performance of artificial intelligence on Turkish dental specialization exam: can ChatGPT-4.0 and gemini advanced achieve comparable results to humans? BMC Med Educ. 2025 Feb 10;25(1):214. doi:10.1186/s12909-024-06389-9
Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016 Jun;15(2):155–63. doi:10.1016/j.jcm.2016.02.012
Atilan AU, Cetin N. An old disease, a new linguistic challenge for large language models: patient education on psoriasis and psoriatic arthritis in an underrepresented medical language. Int J Med Inform. 2026 Apr;209:106246. doi:10.1016/j.ijmedinf.2025.106246
Prasad S, Koseoglu M, Antonopoulou S, Huber HM, Azarbal A, Kurniawan S, et al. Assessing readability and accuracy of content produced by the American College of Prosthodontists and large language models for patient education in prosthodontics. Journal of Prosthodontics. 2025 Aug 22. doi:10.1111/jopr.70022
Çetin SG, Ertuğrul ÖF, Kavasoğlu N, Eratilla V. Artificial Intelligence-Based Evaluation of Permanent First Molar Extraction Indications in Children Using Panoramic Radiographs. Children. 2026 Feb 17;13(2):277. doi:10.3390/children13020277
Sivaramakrishnan G, Almuqahwi M, Ansari S, Lubbad M, Alagamawy E, Sridharan K. Assessing the power of AI: a comparative evaluation of large language models in generating patient education materials in dentistry. BDJ Open. 2025 Jun 18;11(1):59. doi:10.1038/s41405-025-00349-1

Details

Primary Language

English

Subjects

Prosthodontics, Restorative Dentistry

Journal Section

Research Article

Authors

Seçkin Gül ^*
0000-0003-0846-6342
Türkiye

İbrahim Halil Avcılar
0009-0002-7167-9367
Türkiye

Publication Date

August 21, 2026

Submission Date

August 8, 2026

Acceptance Date

August 18, 2026

Published in Issue

Year 2026 Volume: 1 Number: 2

IZ

https://izlik.org/JA84NS24PR

Cite

RIS / Bibtex

APA

Gül, S., & Avcılar, İ. H. (2026). Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers. Journal of Odonto Science, 1(2), 14-25. https://izlik.org/JA84NS24PR

AMA

1.Gül S, Avcılar İH. Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers. JODS. 2026;1(2):14-25. https://izlik.org/JA84NS24PR

Chicago

Gül, Seçkin, and İbrahim Halil Avcılar. 2026. “Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers”. Journal of Odonto Science 1 (2): 14-25. https://izlik.org/JA84NS24PR.

EndNote

Gül S, Avcılar İH (August 1, 2026) Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers. Journal of Odonto Science 1 2 14–25.

IEEE

[1]S. Gül and İ. H. Avcılar, “Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers”, JODS, vol. 1, no. 2, pp. 14–25, Aug. 2026, [Online]. Available: https://izlik.org/JA84NS24PR

ISNAD

Gül, Seçkin - Avcılar, İbrahim Halil. “Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers”. Journal of Odonto Science 1/2 (August 1, 2026): 14-25. https://izlik.org/JA84NS24PR.

JAMA

1.Gül S, Avcılar İH. Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers. JODS. 2026;1:14–25.

MLA

Gül, Seçkin, and İbrahim Halil Avcılar. “Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers”. Journal of Odonto Science, vol. 1, no. 2, Aug. 2026, pp. 14-25, https://izlik.org/JA84NS24PR.

Vancouver

1.Seçkin Gül, İbrahim Halil Avcılar. Evaluation of the Accuracy and Comprehensibility of Turkish Responses Generated by Large Language Models Regarding the Selection of Direct Composite and Ceramic Laminate Veneers. JODS [Internet]. 2026 Aug. 1;1(2):14-25. Available from: https://izlik.org/JA84NS24PR