Related Experiment Video
Updated: May 5, 2026
11:15
In Situ Detection of Bacteria within Paraffin-embedded Tissues Using a Digoxin-labeled DNA Probe Targeting 16S rRNA
Published on: May 21, 2015
10.9K
Artificial Intelligence Diagnosing of Oral Lichen Planus: A Comparative Study
Sensen Yu1, Wansu Sun2, Dawei Mi1,3
1Key Laboratory of Oral Diseases Research of Anhui Province, College & Hospital of Stomatology, Anhui Medical University, Hefei 230032, China.
Bioengineering (Basel, Switzerland)
|November 27, 2024
Summary
Artificial intelligence (AI) shows promise for diagnosing oral lichen planus (OLP), improving accuracy after feature training. However, AI faces limitations with complex cases and less common OLP lesion sites.
Area of Science:
- Oral Medicine
- Artificial Intelligence
- Medical Diagnostics
Background:
- Early diagnosis of oral lichen planus (OLP) is challenging, relying on subjective clinical interpretation.
- Artificial intelligence (AI) offers potential for objective and rapid diagnostic solutions.
Purpose of the Study:
- To investigate the diagnostic potential of AI for OLP.
- To evaluate AI's effectiveness in improving diagnostic accuracy and decision-making speed for OLP.
Main Methods:
- 128 confirmed OLP patients' lesion images were collected from various anatomical sites.
- AI platforms (ChatGPT-4O, ChatGPT with Diagram-Date extension, Claude Opus) were used for direct and pre-training identification.
- AI models were trained on OLP features to assess diagnostic performance.
Main Results:
- Overall OLP recognition rates improved post-training: ChatGPT-4O (59% to 77%), ChatGPT (Diagram-Date) (68% to 80%), Claude Opus (15% to 50%).
- Pre-training recognition for buccal mucosa OLP was high (94%, 93%, 56% respectively).
- AI performance decreased for less common sites like gums (60%, 60%, 20%) and complex cases.
Conclusions:
- AI demonstrates significant potential for enhancing OLP diagnosis accuracy and efficiency.
- AI platforms exhibit varying performance, with limitations in recognizing OLP in complex scenarios and specific anatomical locations.
- Further development is needed to overcome AI limitations for comprehensive OLP diagnostics in oral medicine.
