Objective: To develop and refine an AI-based patient actor tool to train neurology residents to deliver a diagnosis of functional movement disorder (FMD) with clarity and structured communication.

Background: FMD presents with involuntary movements despite no structural neurological pathology, creating challenges in both diagnosis and patient communication [1]. Delivering a FMD diagnosis requires empathy and structured conversation to minimize distress and foster patient engagement. Traditional training methods such as role-play and standardized patient programs are limited by costs and actor variability [2].

Method: An AI-based patient actor training platform was developed to simulate clinical encounters for neurology residents. The system integrates structured prompts, clinical vignettes, and rubric-based feedback to evaluate patient interactions. The chatbot was developed using a large language model API and deployed through a web-based application framework. The rubric was designed to center around two main domains: 1) patient-centered communication and 2) effective communication of a FMD diagnosis. For #1, rubric elements were derived from the NURSE framework (Name, Understand, Respect, Support, Explore) and the Master Interview Rating Scale (MIRS) [3,4]. For #2, expert recommendations were incorporated, such as clearly stating the diagnosis and the rationale for the diagnosis [1,5]. The AI-based patient actor was then piloted among team members.

Results: Iterative refinements to the AI patient actor were made in response to feedback during the pilot phase, including improvements in response realism, more natural conversational pacing, and reduction of verbal fillers. Additional modifications included the incorporation of an AI evaluator agent and refinement of system prompts for the AI patient actor and evaluator.

Conclusion: Generative AI-based patient actors may provide a scalable and structured approach to communication training in neurology. By enabling realistic practice and feedback, this educational innovation has the potential to improve patient experiences in FMD care. Future research aims to assess neurology resident communication skills before and after implementation of the AI patient actor and satisfaction with this educational prototype.

Previous Presentations: This abstract was presented at the 2025 UCLA DGSOM Medical Education Day on November 15, 2025.

References: [1] Hallett M, Aybek S, Dworetzky BA, McWhirter L, Staab JP, Stone J. Functional neurological disorder: new subtypes and shared mechanisms. Lancet Neurol. 2022;21(6):537–550. doi:10.1016/S1474-4422(21)00422-1.
[2] Zidoun Y, Mardi AE. Artificial intelligence (AI)-based simulators versus simulated patients in undergraduate programs: a protocol for a randomized controlled trial. BMC Med Educ. 2024;24:1260. doi:10.1186/s12909-024-06236-x.
[3] Childers JW, Arnold RM, Back AL, Tulsky JA. Beyond the NURSE acronym: the functions of empathy in serious illness conversations. J Pain Symptom Manage. 2023;65(4):e375–e379. doi:10.1016/j.jpainsymman.2022.11.029.
[4] Eastern Virginia Medical School, Professional Skills Teaching and Assessment Center. Master interview rating scale (MIRS) [Internet]. Norfolk (VA): Eastern Virginia Medical School; 2003 [cited 2026 Mar 12]. Available from: https://www.dmu.edu/wp-content/uploads/Master-Interview-Rating-Scale.pdf
[5] Stone J, Carson A, Hallett M. Chapter 44: Explanation as treatment for functional neurologic disorders. In: Aminoff MJ, Boller F, Swaab DF, editors. Handbook of Clinical Neurology. Vol. 139. Amsterdam: Elsevier; 2016. p. 1–22.

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/enhancing-neurology-resident-communication-of-functional-movement-disorder-using-an-artificial-intelligence-ai-based-patient-actor/