Related Experiment Video
Updated: Jan 9, 2026
03:14
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
986
Leveraging large language models for patient instructions in dentistry-A systematic review and meta-analysis
Summary
Large language models (LLMs) show promise in answering dental patient questions, with an overall accuracy of 81.87%. While clinically acceptable, further research is needed for complex cases.
Area of Science:
- Artificial Intelligence in Healthcare
- Dental Informatics
- Natural Language Processing
Background:
- Patient inquiries are increasingly common in dentistry.
- Large language models (LLMs) offer potential solutions for patient communication.
- Evaluating the accuracy of LLMs in dental contexts is crucial.
Purpose of the Study:
- To systematically review the application of ChatGPT and other LLMs in answering dental patient inquiries.
- To assess the accuracy and clinical acceptability of LLM-generated responses to dental questions.
Main Methods:
- Systematic review following PRISMA guidelines (November 2022 - June 2024).
- Searched seven databases (PubMed, Scopus, Cochrane).
- Evaluated study quality using the modified QUADAS-2 framework and performed meta-analysis.
Main Results:
- Included 25 studies evaluating various LLMs.
- Pooled accuracy for all LLMs was 81.87% (95% CI: 77.24%-86.51%).
- 69.9% of responses were clinically acceptable; ChatGPT-3.5 showed higher accuracy than Bing, comparable to GPT-4 and Bard.
Conclusions:
- LLMs like ChatGPT are promising for dental patient inquiries and oral health education.
- Challenges include accuracy, response variability, and handling complex clinical scenarios.
- Further research is essential to refine LLM applications in dentistry.
