Medical chatbot interface for large language models

• Main Author: Chua, Yu Hao
• Other Authors: Dusit Niyato
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science
• Online Access: https://hdl.handle.net/10356/174530
• Institution: Nanyang Technological University

In recent years, there has been a growing strain on accident and emergency departments in hospitals, often causing nurses and doctors to be overworked. Hence, there is a need to streamline and optimise healthcare services. This project addresses this challenge with a chatbot interface designed to tackle patient inquiries and provide simple medical advice. The key focus lies in the maintenance of personal data privacy while still providing personalised responses. The chatbot interface, implemented in Python, leverages open-source libraries like Chainlit and LangChain to seamlessly integrate large language models(LLMs) with a user-friendly interface. By using a LLM, the chatbot can tap into the collective knowledge of diverse healthcare datasets. Patients can ask a wide range of medical questions, receive relevant advice, and gain insights into their health concerns without adding to the load of accident and emergency departments. The implementation details including the integration of Chainlit and LangChain are presented in this report, demonstrating the feasibility and effectiveness of the proposed chatbot interface.