With great pleasure and joy we are sharing our latest scientific publication titled „AI-Assisted Training for Teleconsultation Competencies in Undergraduate Medical Education: A Narrative Review„, co-authored by Wojciech Michał GLINKOWSKI (1,2), Barbara JACENNIK (2,3), Aldona Katarzyna JANKOWSKA (4,5), Tomasz CEDRO (2), Szymon WILK (2,6), Rafał DONIEC (2,7), from (1) Center of Excellence “TeleOrto” for Telediagnostics and Treatment of Disorders and Injuries of the Locomotor System, Department of Medical Informatics and Telemedicine, Medical University of Warsaw, Warsaw, Poland; (2) The Polish Telemedicine and eHealth Society, Warsaw, Poland; (3) Faculty of Social Sciences, Prince Mieszko I Poznań Medical University of Applied Sciences, Poznań, Poland; (4) Polish Academy of Sciences Scientific Station in Paris, Paris, France; (6) Faculty of Computing and Telecommunications, Poznan University of Technology, Poznan, Poland; (7) Faculty of Medical Sciences in Zabrze, Medical University of Silesia, Katowice, Poland. Paper was published in Applied Sciences 2026, 16(10), 4858 journal (IF2.5, CS5.5) and is available online in open-access at https://doi.org/10.3390/app16104858.
Abstract
Telemedicine has become a routine component of healthcare delivery, creating a need for dedicated undergraduate training in teleconsultation-specific competencies. Although artificial intelligence (AI)-assisted educational systems have been proposed as scalable tools to support teleconsultation training, the evidence remains fragmented, and their educational role is not yet clearly defined.
Objective: To map and critically synthesize empirical evidence on AI-assisted teleconsultation training systems used in undergraduate medical education, with attention to skill domains, system capabilities, and implementation considerations.
Methods: A structured narrative review with transparent search and study selection procedures was conducted. Literature published between January 2019 and December 2025 was identified through searches of major bibliographic databases and supplementary semantic and citation-based sources. Studies involving undergraduate medical students and evaluating AI-assisted interventions targeting teleconsultation-related skills were included.
Results: Eight empirical full-text studies met the final eligibility criteria and were included in the structured narrative synthesis. Across the included studies, AI-assisted systems tended to show favorable patterns in structured domains such as verbal communication, history-taking, and selected aspects of early clinical reasoning during virtual consultations. Evidence regarding nonverbal communication and empathic or relational skills was more limited and methodologically heterogeneous, and human-based simulation remained important in these domains. Students generally reported favorable perceptions of usability, accessibility, and psychological safety, although satisfaction and perceived realism were not uniformly superior to human-based approaches. AI-assisted systems also appeared scalable and potentially cost-efficient, particularly as preparatory or supplementary training modalities.
Conclusions: Current evidence suggests that AI-assisted teleconsultation training systems may be useful as preparatory and supportive tools in undergraduate medical education, particularly for structured and repeatable components of remote consultation practice. However, the evidence base remains limited and heterogeneous, and these systems do not replace human-led training for relational, nonverbal, and context-sensitive competencies. Their educational value appears greatest within blended training models that align platform capabilities with specific teleconsultation skills.
Keywords: artificial intelligence; teleconsultation; medical education; undergraduate medical students; simulation-based training; large language models; digital health education; clinical communication skills; narrative review.