C-PATH: Conversational Patient Assistance and Triage in Healthcare System

2025-06-07

AI Summary

Healthcare system complexity impedes timely and appropriate patient access to services. To address this, we propose a patient-facing, multi-turn conversational AI triage system. Our method introduces three key innovations: (1) a GPT-driven clinical knowledge dialogization framework that automatically transforms structured diagnostic knowledge into high-quality, layperson-accessible dialogue training data; (2) a long-horizon coherent dialogue history management mechanism to enhance contextual understanding; and (3) a multi-stage fine-tuning paradigm built upon LLaMA3, integrating DDXPlus medical knowledge injection, joint training on clinical summaries and dialogue data, and automated multidimensional evaluation via GPTScore. Evaluated on a GPT-rewritten dialogue dataset, our system achieves state-of-the-art performance in clarity, informativeness, and department recommendation accuracy—demonstrating significant improvements in triage accessibility and response efficiency.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Conversational AI/Dialog Systems
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Navigating healthcare systems can be complex and overwhelming, creating barriers for patients seeking timely and appropriate medical attention. In this paper, we introduce C-PATH (Conversational Patient Assistance and Triage in Healthcare), a novel conversational AI system powered by large language models (LLMs) designed to assist patients in recognizing symptoms and recommending appropriate medical departments through natural, multi-turn dialogues. C-PATH is fine-tuned on medical knowledge, dialogue data, and clinical summaries using a multi-stage pipeline built on the LLaMA3 architecture. A core contribution of this work is a GPT-based data augmentation framework that transforms structured clinical knowledge from DDXPlus into lay-person-friendly conversations, allowing alignment with patient communication norms. We also implement a scalable conversation history management strategy to ensure long-range coherence. Evaluation with GPTScore demonstrates strong performance across dimensions such as clarity, informativeness, and recommendation accuracy. Quantitative benchmarks show that C-PATH achieves superior performance in GPT-rewritten conversational datasets, significantly outperforming domain-specific baselines. C-PATH represents a step forward in the development of user-centric, accessible, and accurate AI tools for digital health assistance and triage.

Problem

• Assists patients in recognizing symptoms via conversational AI
• Recommends appropriate medical departments through natural dialogues
• Improves healthcare accessibility with AI-powered triage system

Innovation

• LLM-powered conversational AI for healthcare triage
• GPT-based data augmentation for medical dialogues
• Scalable history management for coherent conversations