Related Experiment Video
Updated: Jul 15, 2025

Digital Home-Monitoring of Patients after Kidney Transplantation: The MACCS Platform
Published on: April 12, 2021
Accuracy and Reliability of Chatbot Responses to Physician Questions
Rachel S Goodman1, J Randall Patrinely2, Cosby A Stone3
1Vanderbilt University School of Medicine, Nashville, Tennessee.
Artificial intelligence chatbots provide mostly accurate medical information for physicians, with accuracy improving over time and across different versions. However, limitations necessitate further research and development for clinical use.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
- Natural Language Processing
Background:
- Natural language processing tools like ChatGPT offer potential to improve medical information accessibility for professionals and patients.
- Assessing AI chatbot safety and efficacy in clinical settings is crucial for decision-making and healthcare efficiency.
Purpose of the Study:
- To evaluate the accuracy and completeness of AI chatbot responses to physician-generated medical questions.
- To identify the reliability and limitations of AI-generated medical information.
Main Methods:
- 33 physicians across 17 specialties generated 284 medical questions.
- Physicians rated chatbot answers for accuracy (6-point scale) and completeness (3-point scale).
- Responses were compared between GPT-3.5 and GPT-4, and re-evaluated over time.
Main Results:
- The median accuracy score was 5.5 (out of 6), and median completeness was 3.0 (out of 3).
- Accuracy slightly decreased with question difficulty but remained high for binary and descriptive questions.
- Re-querying and using GPT-4 significantly improved response accuracy.
Conclusions:
- AI chatbots generate largely accurate medical information for physician queries, with notable improvements over time and with newer versions.
- Despite high accuracy, limitations exist, indicating a need for further research, model development, and validation before widespread clinical adoption.
More Related Videos
05:47Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
Published on: June 13, 2025
03:14Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Related Concept Videos
Errors occurring during blood pressure monitoring
Several factors...
Ethics and Bioethics
Nurses' Legal Responsibilities II
Communication between nurses and...
Guidelines and Strategies for Safe Computer Charting
Maintain Confidentiality and Security:
Barriers to Effective Communication II
Cultural barriers:
Differences in values, beliefs, religion, knowledge, and tradition can significantly impact communication. Awareness of nonverbal cues is critical, especially when conversing with a patient from a different culture. What appears appropriate in one culture may be inappropriate in another.
Semantic barriers:
As a result of their tendency to use...
Patient-centered Care