Table 1.
Accuracy of large language models for text-based, image-based questions and the total sample.
Fig 1.
Gemini’s automated response to image-based questions attaching radiographs.
AI, artificial intelligence. Note: This figure is a recreation for illustrative purposes only and is not an exact replica of the Gemini interface.
Fig 2.
Grouped stacked bar chart presenting the accuracy of large language models for text-based, image-based questions and the total sample.
Percentages with the same lowercase letter were not statistically different, as determined by the post-hoc analysis with Bonferroni correction.
Fig 3.
Grouped stacked bar chart presenting the accuracy of large language models for individual dental disciplines.
(A) Anatomic Sciences (AS), Biochemistry & Physiology (BP), Microbiology & Pathology (MP), Dental Anatomy & Occlusion (DAO), Pharmacology (PH), Operative Dentistry & Prosthodontics (ODP), Oral & Maxillofacial Surgery (OMF). (B) Orthodontics (OR), Pediatric Dentistry (PD), Endodontics (EN), Periodontics (PE), Radiology (RA), Pathology (PA), Patient Management & Public Health (PMPH).
Table 2.
Accuracy of large language models for individual dental disciplines.