Clinical oral investigations · 2025
Performance of artificial intelligence chatbots in the diagnosis and management of simulated dental trauma cases: an evaluation based on IADT guidelines.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Keleş Öznur Küçük, Arslan Zeynep Betül
Аннотация
AIM: This study aims to comparatively evaluate the performance of four different artificial intelligence-based chatbots (ChatGPT-4o (Free), ChatGPT-5 (Plus), DeepSeek, and Google Gemini) in the diagnosis and treatment processes of dental trauma cases.
MATERIAL AND METHODS: Based on the International Association of Dental Traumatology (IADT) guidelines, eleven fictional cases were developed, each representing different diagnostic possibilities of dental trauma. The cases cenarios were based on a standardized dataset including clinical examination, radiographic findings, and pulp sensitivity tests. All AI models were tested for three days for each case. Two researchers analyzed the responses using ablinding method according to five evaluation criteria (diagnostic accuracy, treatment plan appropriateness, splinting duration accuracy, antibiotic indication, and reference accuracy).
RESULTS: According to the results, while no significant difference was found in terms of diagnostic accuracy (p > 0.05), Google Gemini showed the highest performance with 100% accuracy. ChatGPT-4o (Free) stood out with a 97% accuracy rate in antibiotic indication, while in splinting duration prediction, ChatGPT-5 (Plus) was the most successful model with 75.8%. DeepSeek exhibited the highest variability (p < 0.05).
CONCLUSIONS: The findings show that AI chatbots have promising potential as complementary tools in dental trauma diagnosis, but evidence-based validation, expert supervision, and methodological standards are needed for safe use in clinical practice.
MATERIAL AND METHODS: Based on the International Association of Dental Traumatology (IADT) guidelines, eleven fictional cases were developed, each representing different diagnostic possibilities of dental trauma. The cases cenarios were based on a standardized dataset including clinical examination, radiographic findings, and pulp sensitivity tests. All AI models were tested for three days for each case. Two researchers analyzed the responses using ablinding method according to five evaluation criteria (diagnostic accuracy, treatment plan appropriateness, splinting duration accuracy, antibiotic indication, and reference accuracy).
RESULTS: According to the results, while no significant difference was found in terms of diagnostic accuracy (p > 0.05), Google Gemini showed the highest performance with 100% accuracy. ChatGPT-4o (Free) stood out with a 97% accuracy rate in antibiotic indication, while in splinting duration prediction, ChatGPT-5 (Plus) was the most successful model with 75.8%. DeepSeek exhibited the highest variability (p < 0.05).
CONCLUSIONS: The findings show that AI chatbots have promising potential as complementary tools in dental trauma diagnosis, but evidence-based validation, expert supervision, and methodological standards are needed for safe use in clinical practice.