Related Experiment Video
Updated: Aug 16, 2026
07:14
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
Published on: December 23, 2025
Artificial Intelligence-Augmented Standardized Patient Models for AETCOM (Attitude, Ethics, and Communication)
Nayyar Iqbal1, Magi Murugan2, Sunil Subramanyam3
1General Medicine, Pondicherry Institute of Medical Sciences, Puducherry, IND.
Cureus
|August 15, 2026
Summary
Artificial intelligence (AI) chatbots offer a reliable and consistent method for assessing medical students' communication skills. This AI-driven approach provides immediate feedback, enhancing training and evaluation in medical education.
Area of Science:
- Medical Education Technology
- Artificial Intelligence in Healthcare
- Communication Skills Assessment
Background:
- Communication skills are vital in medical education, crucial for patient trust, history taking, and informed consent.
- Traditional assessment methods like OSCEs are reliable but resource-intensive and offer delayed feedback.
- The Attitude, Ethics, and Communication (AETCOM) module emphasizes these core competencies.
Purpose of the Study:
- To develop and evaluate AI chatbots for assessing communication skills in medical interns.
- To compare AI-based assessment with expert evaluation for reliability and agreement.
- To explore AI's potential in providing standardized and timely feedback for communication skills training.
Main Methods:
- Two AI chatbots were developed: a virtual patient and an evaluation system with a validated rubric.
- Thirty-three interns interacted with the virtual patient, with transcripts assessed by AI and five subject experts.
- Intraclass correlation coefficients (ICCs) and Bland-Altman analysis were used to assess AI reliability and agreement with expert scores.
Main Results:
- AI scoring demonstrated excellent internal reliability (ICC (2, 1) = 0.904; ICC (2, 5) = 0.979).
- Expert ratings improved with consensus scoring (ICC (2, 5) = 0.852).
- Bland-Altman analysis showed close alignment between AI and expert consensus (bias -0.529, limits of agreement -1.998 to 0.940).
Conclusions:
- AI offers a highly reproducible, consistent, and standardized method for communication skills assessment.
- AI virtual patients facilitate immediate feedback and repeated practice, reducing assessment variability.
- AI-based systems, combined with expert assessment, present a scalable, learner-centered approach for enhanced medical communication training.