AI-Powered Avatars in Medical Education: Advancing Virtual Coaching for Clinical Readiness

Authors

  • Fernando Salvetti e-REAL Labs at Logosnet, Turin, Italy – Lugano, Switzerland –Houston, TX, USA – New York, NY, USA https://orcid.org/0009-0009-5197-305X
  • Barbara Bertagni e-REAL Labs at Logosnet, Turin, Italy – Lugano, Switzerland –Houston, TX, USA – New York, NY, USA; Center for Medical Simulation, Boston, MA, USA
  • Roxane Gardner Brigham and Women’s Hospital/Children’s Hospital/ Massachusetts General Hospital/Harvard Medical School, Boston, MA, USA
  • Jenny Rudolph Center for Medical Simulation/Beth Israel Deaconess Medical Center, Boston, MA, USA

DOI:

https://doi.org/10.3991/ijac.v19i2.58755

Keywords:

Medical Education, AI-Powered Avatars, Virtual Coaching, Clinical Readiness, Healthcare Simulation, Bias Mitigation.

Abstract


The increasing complexity of healthcare environments requires innovative educational approaches to better prepare students and professionals for the realities of clinical practice. Traditional lectures and limited role-play simulations often fall short in providing repeated, standardized, and unbiased opportunities to practice essential communication skills. At e-REAL Labs, we have developed Conversation Mastery™, a proprietary platform powered by AI-driven avatars—embodied conversational agents (ECAs)—specifically designed to enhance virtual coaching in medical education. These avatars recreate difficult and emotionally charged clinical conversations, offering real-time adaptive feedback and fostering critical skills such as conflict resolution, empathy, and inclusive communication. Unlike generic AI coaching tools, the system integrates generative AI, natural language processing, and behavioral analytics into a controlled and curriculum-aligned framework. Learners benefit from personalized, structured feedback that directly supports the acquisition of clinical competencies and professional readiness. In collaboration with the Harvard Center for Medical Simulation, we conducted a controlled study comparing lecture-based instruction, traditional role-play, and AI-driven avatar training among nursing students. Results demonstrate that students training with avatars achieved significantly higher success in conflict de-escalation, improved confidence, and stronger adherence to best practices. By embedding Conversation Mastery into simulation-based education, institutions can expand practice opportunities, ensure biasaware evaluation, and support compliance with GDPR, HIPAA, and emerging AI governance frameworks. This paper extends previous work presented during The Learning Ideas Conference 2025 (held at Columbia University in New York City and online) by providing a deeper exploration of the pedagogical, ethical, and regulatory implications of integrating AI-driven avatars into healthcare training. We argue that these technologies represent a new paradigm for clinical readiness, complementing faculty expertise while scaling access to highquality, adaptive, and equitable learning experiences.

Author Biography

Fernando Salvetti, e-REAL Labs at Logosnet, Turin, Italy – Lugano, Switzerland –Houston, TX, USA – New York, NY, USA

Founder and Managing Partner, e-REAL Labs at Logosnet (Turin, Italy; Lugano, Switzerland; Houston and New York, USA)

Downloads

Published

2026-06-03

How to Cite

Salvetti, F., Bertagni, B., Gardner, R., & Rudolph, J. (2026). AI-Powered Avatars in Medical Education: Advancing Virtual Coaching for Clinical Readiness. International Journal of Advanced Corporate Learning (iJAC), 19(2), pp. 80–102. https://doi.org/10.3991/ijac.v19i2.58755

Issue

Section

TLIC Papers