ArticleBMC oral health2026
Comparative diagnostic accuracy of multiple large language models in oral and maxillofacial radiology specialty examinations: a 13-year analysis of performance and topic trends.
Article in BMC oral health, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 1 paper.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
1 citing paper in PubMed.
- Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
2 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundTo the best of our knowledge, this is the first study in oral and maxillofacial radiology to comprehensively include all examination questions, to systematically analyse their topic distribution across years and examination periods, and to concurrently compare multiple contemporary large language models within a unified methodological framework. This study aimed to compare the accuracy performance of six different artificial intelligence (AI) systems based on large language models (LLMs) of questions asked in the field of oral and maxillofacial radiology in the Dental Specialization Examination (DSE) over the past 13 years, and to analyze the subject matter in detail.
methodsA total of 200 oral and maxillofacial radiology questions from the DSE held between 2012 and 2025 were included in the analysis. The questions were grouped according to their topics and divided into early-late periods (2012-2018 and 2019-2025) to observe changes over time. ChatGPT-5.2, ChatGPT-4.0, Gemini-3, Claude 4.5, Microsoft Copilot, and Perplexity AI were tested using the original question formats. The models' answers were evaluated against the official answer key. In addition, the questions were analyzed according to exam years and periods.
resultsIn the early and late periods, ChatGPT-5.2 showed accuracy rates of 91.9% and 95.7%, respectively. This was followed by ChatGPT-4.0 (79.8% - 82.8%). Differences between the models were statistically significant across periods (p < 0.001). Oral diseases and oral pathology retained their importance in both early and late stages of oral health. Furthermore, while oral diseases and oral pathology were more frequently inquired about in the spring, advanced imaging techniques, radiation physics, and temporomandibular joint disorders were also included in the autumn surveys.
conclusionChatGPT-5.2 demonstrated the highest and most consistent accuracy among the evaluated models in DSE oral and maxillofacial radiology questions. In the field of oral and maxillofacial radiology, these model prototypes have the potential to generate new knowledge. Interest in oral diseases and pathology has remained important, while attention to jaw lesions and advanced imaging techniques has increased in recent years.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.