ArticleDiagnostic pathology2026
Virtual assistants based on artificial intelligence for oral diagnosis: help for clinicians AI oral diagnosis helper.
Article in Diagnostic pathology, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
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
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
12 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundOral cancer is one of the most aggressive malignancies in the head and neck region, with high potential for and a poor prognosis. Despite advancements in diagnostic methods, early detection remains a challenge owing to the lack of specific symptoms in the initial stages and socioeconomic barriers limiting access to health care. Artificial intelligence has emerged as a promising tool for improving diagnostic accuracy in various medical fields, including oral pathology. This study aimed to evaluate the performance of the AI Oral Diagnosis Helper (AODH), developed using ChatGPT™ (versions 4 and 4o), in the diagnosis and management of oral diseases.
methodsThirty clinical cases were evaluated by two specialists for oral diagnosis and AODH. Diagnosis and treatment suggestions were compared with those of a gold standard specialist. Agreement and accuracy rates were calculated via Fleiss’ generalized kappa coefficient with a 95% confidence interval via RStudio.
resultsAODH via ChatGPT™ 4 (AODH4) correctly identified 22 out of 30 cases (73.3%), whereas AODH via version 4° (AODH4°) correctly identified 24 cases (80%). Expert 1 and Expert 2 correctly diagnosed 27 (90%) and 26 (86.7%) cases, respectively. Fleiss’s kappa indicated substantial reliability among AODHs and experts (K = 0.79). The AODH4 recommended appropriate diagnostic tests in 21 cases (70%), whereas the AODH4o did so in 26 cases (86.7%). For biopsy recommendations, Specialist 1 was accurate in 27 cases (90%), Specialist 2 was accurate in 26 cases (86.7%), AODH4 was accurate in 17 cases (56.7%), and AODH4o was accurate in 26 cases (86.7%).
conclusionAODH demonstrated high diagnostic accuracy comparable to that of human specialists and can assist in oral diagnosis, especially where access to specialists is limited. The performance of AODH4o was superior to that of AODH4. Further research is needed to define the role of AI in enhancing clinical decision-making.
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.