PriHA: A RAG-Enhanced LLM Framework for Primary Healthcare Assistant in Hong Kong

TL;DR

PriHA is a specialized LLM framework enhanced with retrieval techniques to improve primary healthcare information access and accuracy in Hong Kong.

Contribution

The paper introduces a novel DRAG architecture and a tri-stage pipeline for localized, accurate healthcare information retrieval using LLMs.

Findings

Outperforms baseline models in accuracy and clarity

Effective in handling mixed-source retrieval tasks

Provides a reliable framework for high-risk localized applications

Abstract

To address the unsustainable rise in public health expenditures, the Hong Kong SAR Government is shifting its strategic focus to primary healthcare and encouraging citizens to use community resources to self-manage their health. However, official clinical guidelines are fragmented across disparate departments and formats, creating significant access barriers. While general-purpose Large Language Models (LLMs) such as ChatGPT and DeepSeek offer potential solutions for information accessibility, they are prone to generating factually inaccurate content due to a lack of localized and domain-specific knowledge. To this end, we propose a Retrieval-Augmented Generation-Enhanced LLM system as Primary Healthcare Assistant (PriHA) in Hong Kong. Specifically, a tri-stage pipeline is proposed that leverages a query optimizer to generalize user intent-oriented sub-queries, followed by a novel Dual Retrieval Augmented Generation (DRAG) architecture for mixed-source retrieval and context-reorganized generation. Comprehensive experiments and a detailed case study demonstrate that our proposed method can outperform both ablations and baseline in terms of accuracy and clarity. Our research provides a reliable and traceable dialogue retrieval framework for exploring other high-risk, localized application scenarios.