Lightweight Retrieval-Augmented Generation and Large Language Model-Based Modeling for Scalable Patient-Trial Matching

2026-04-23

AI Summary

This work addresses the challenges of scalability, generalization, and computational efficiency in patient–clinical trial matching arising from lengthy, heterogeneous electronic health records and complex eligibility criteria. The authors propose a lightweight framework that first employs retrieval-augmented generation to extract clinically relevant text snippets, then leverages large language models—used both in frozen and fine-tuned settings—to encode structured and unstructured information. These representations are subsequently compressed via dimensionality reduction and fed into a lightweight classifier for efficient matching. By decoupling information retrieval from representation modeling, the approach achieves performance on par with end-to-end large models across multiple public benchmarks and real-world data from the Mayo Clinic, while substantially reducing computational overhead.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Patient-trial matching requires reasoning over long, heterogeneous electronic health records (EHRs) and complex eligibility criteria, posing significant challenges for scalability, generalization, and computational efficiency. Existing approaches either rely on full-document processing with large language models (LLMs), which is computationally expensive, or use traditional machine learning methods that struggle to capture unstructured clinical narratives. In this work, we propose a lightweight framework that combines retrieval-augmented generation and large language model-based modeling for scalable patient-trial matching. The framework explicitly separates two key components: retrieval-augmented generation is used to identify clinically relevant segments from long EHRs, reducing input complexity, while large language models are used to encode these selected segments into informative representations. These representations are further refined through dimensionality reduction and modeled using lightweight predictors, enabling efficient and scalable downstream classification. We evaluate the proposed approach on multiple public benchmarks (n2c2, SIGIR, TREC 2021/2022) and a real-world multimodal dataset from Mayo Clinic (MCPMD). Results show that retrieval-based information selection significantly reduces computational burden while preserving clinically meaningful signals. We further demonstrate that frozen LLMs provide strong representations for structured clinical data, whereas fine-tuning is essential for modeling unstructured clinical narratives. Importantly, the proposed lightweight pipeline achieves performance comparable to end-to-end LLM approaches with substantially lower computational cost.

Problem

• patient-trial matching
• electronic health records
• eligibility criteria
• scalability
• computational efficiency

Innovation

• Retrieval-Augmented Generation
• Large Language Models
• Patient-Trial Matching
• Lightweight Modeling
• Electronic Health Records