Performance Comparison of a Domain‐Specific Chatbot and General‐Purpose Chatbots in Dental Traumatology
Dental Traumatology · 2026
DOI: 10.1111/edt.70039 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Dental Traumatology, 2026. Verfasst von Vipin Kumar und 6 weiteren Autor:innen. Beteiligte Einrichtung: Division of Conservative Dentistry and Endodontics, Centre for Dental Education and Research, All India Institute of Medical Sciences, New Delhi, Ind…. Hauptthemen laut MEDLINE: Traumatology, Artificial Intelligence, Tooth Injuries, Tooth Fractures. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Use of artificial intelligence chatbots in dental traumatology has increased. However, concerns regarding their reliability are yet to be addressed. This study aims to evaluate the accuracy of a new AI chatbot Dental Trauma Evo in responding to queries on dental fractures and luxations. Materials and Methods A total of 45 questions, including multiple‐choice questions (MCQs), true/false, and yes/no types of questions were created and validated in accordance with the International Association of Dental Traumatology's position statement on fractures and luxations. Over the course of nine consecutive days in incognito mode, they were simultaneously exposed to four chatbots: ChatGPT‐4o, DeepSeek R1, Google Gemini 2.5, and Dental Trauma Evo. The obtained answers were verified for accuracy and consistency. The Fisher's exact test was used for statistical analysis.
Results
The best overall accuracy was shown by Dental Trauma Evo (85.43%), which was followed by Google Gemini (81.72%), DeepSeek (80.24%), and ChatGPT‐4o (79.75%). Regarding question type, ChatGPT‐4o, Google Gemini, and Dental Trauma Evo recorded the best responses to Yes/No, True/False, and MCQs respectively. There was an insignificant difference between the question type and the different chatbots ( p > 0.05).
Conclusion
The Dental Trauma Evo chatbot developed using the International Association of Dental Traumatology guidelines exhibited a favorable preliminary performance in the current study. Further research, clinical validation, and model enhancements are necessary to ensure the effective implementation of the same in practice.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Division of Conservative Dentistry and Endodontics, Centre for Dental Education and Research, All India Institute of Medical Sciences, New Delhi, India.
- Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India.
ORCID-Profile: Akshat Sachdeva, Sidhartha Sharma, Amrita Chawla, Vijay Kumar, Shivam Pandey
Zitieren
APA
Kumar, V., Sachdeva, A., Sharma, S., Chawla, A., Kumar, V., Pandey, S., & Logani, A. (2026). Performance Comparison of a Domain‐Specific Chatbot and General‐Purpose Chatbots in Dental Traumatology. Dental Traumatology. https://doi.org/10.1111/edt.70039
Vancouver
Kumar V, Sachdeva A, Sharma S, Chawla A, Kumar V, Pandey S, et al. Performance Comparison of a Domain‐Specific Chatbot and General‐Purpose Chatbots in Dental Traumatology. Dental Traumatology. 2026. doi:10.1111/edt.70039
BibTeX
@article{Kumar2026,
author = {Kumar, Vipin and Sachdeva, Akshat and Sharma, Sidhartha and Chawla, Amrita and Kumar, Vijay and Pandey, Shivam and Logani, Ajay},
title = {Performance Comparison of a Domain‐Specific Chatbot and General‐Purpose Chatbots in Dental Traumatology},
journal = {Dental Traumatology},
year = {2026},
doi = {10.1111/edt.70039}
}
EUVIMED