Enhancing patient engagement in hip prosthesis Surgery: A conversational agent using NLP-based question answering and self-reporting tools

Abstract

Recent advances in NLP and large language models (LLMs) have enabled conversation agents to deliver human-like conversations. However, in healthcare, challenges such as outdated data, lack of response control, and hallucinations limit their safe use. This project develops a German-language conversational agent to support patients undergoing hip prosthesis surgery by answering questions and collecting health status data (EQ-5D). Using a lightweight Retrieval-Augmented Generation (RAG) approach, the system ensures more accurate, reliable, and context-specific responses while remaining resource-efficient.