Health-LLM: Personalized Retrieval-Augmented Disease Prediction System

2024-02-01

arXiv.org

AI Summary

Existing medical AI systems struggle to efficiently process heterogeneous health data and unify patient symptom representations, limiting the accuracy and interpretability of personalized disease prediction. To address this, we propose a health-report-driven Retrieval-Augmented Generation (RAG)-enhanced prediction framework that integrates medical knowledge graphs, multimodal health feature extraction, and fine-tuned large language models (LLMs). Our RAG mechanism enables precise feature retrieval and question-answering–based clinical reasoning. We further introduce a novel semi-automated dynamic feature update strategy to ensure clinical semantic alignment. The framework delivers knowledge-guided, interpretable predictions. Evaluated on large-scale real-world health report data, our method achieves a 12.6% accuracy improvement over state-of-the-art baselines. Clinical experts validated the interpretability and clinical relevance of its outputs, significantly enhancing clinical adaptability and trustworthiness.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Feature Construction/Reformulation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Recent advancements in artificial intelligence (AI), especially large language models (LLMs), have significantly advanced healthcare applications and demonstrated potentials in intelligent medical treatment. However, there are conspicuous challenges such as vast data volumes and inconsistent symptom characterization standards, preventing full integration of healthcare AI systems with individual patients' needs. To promote professional and personalized healthcare, we propose an innovative framework, Heath-LLM, which combines large-scale feature extraction and medical knowledge trade-off scoring. Compared to traditional health management applications, our system has three main advantages: (1) It integrates health reports and medical knowledge into a large model to ask relevant questions to large language model for disease prediction; (2) It leverages a retrieval augmented generation (RAG) mechanism to enhance feature extraction; (3) It incorporates a semi-automated feature updating framework that can merge and delete features to improve accuracy of disease prediction. We experiment on a large number of health reports to assess the effectiveness of Health-LLM system. The results indicate that the proposed system surpasses the existing ones and has the potential to significantly advance disease prediction and personalized health management.

Problem

• Medical AI
• Data Handling
• Personalized Medicine

Innovation

• Health-LLM
• Information Extraction
• Semi-automatic Update Mechanism