Minimizing Chat-Bot Risks in Telemedical Applications: A Semi-Rule-Based System Approach for Large Language Model (LLM) Interactions in an Outpatient Setting

Abstract

We present a patient-centric system integrating Large Language Models (LLMs) into medical applications, focusing on a diverse set of use cases. An initial use case for symptom reporting was explored using natural language, addressing the limitations of traditional questionnaires. This collected data can be analysed by healthcare professionals during visits. Designed within the EU's Medical Device Regulation (MDR), our system incorporates a semi-rule-based approach to guide conversations, ensuring control over the LLM's outputs. With modular architecture and open standards like FHIR, our system supports personalized medicine and future advancements in AI-driven healthcare tools.

Keywords: large language models (LLMs), telemedical applications, risk minimization, systems engineering, human-AI interaction

DOI: 10.54941/ahfe1005699