TrialMatchAI: An End-to-End AI-powered Clinical Trial Recommendation System to Streamline Patient-to-Trial Matching

2025-05-13

AI Summary

Clinical trial patient recruitment is persistently hindered by low-efficiency manual screening and challenges in integrating heterogeneous data—particularly structured electronic health records and unstructured clinician notes. To address this, we propose an end-to-end, interpretable AI system that introduces medical Chain-of-Thought–guided eligibility reasoning—novel in clinical trial matching—for transparent, stepwise qualification assessment. Our method integrates biomedical entity standardization, hybrid lexical-semantic retrieval, and a RAG-enhanced fine-tuned open-source large language model, while supporting Phenopackets interoperability and on-premises privacy-preserving deployment. A modular architecture ensures model replaceability and decision transparency. Evaluated in real-world oncology settings, the system achieves 92% recall within the top-20 trial recommendations per patient; domain experts validated >90% accuracy in criterion-level matching, with exceptional performance in biomarker-based eligibility assessment.

Technology Category

• Humans and AI: AI for Accessibility
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Patient recruitment remains a major bottleneck in clinical trials, calling for scalable and automated solutions. We present TrialMatchAI, an AI-powered recommendation system that automates patient-to-trial matching by processing heterogeneous clinical data, including structured records and unstructured physician notes. Built on fine-tuned, open-source large language models (LLMs) within a retrieval-augmented generation framework, TrialMatchAI ensures transparency and reproducibility and maintains a lightweight deployment footprint suitable for clinical environments. The system normalizes biomedical entities, retrieves relevant trials using a hybrid search strategy combining lexical and semantic similarity, re-ranks results, and performs criterion-level eligibility assessments using medical Chain-of-Thought reasoning. This pipeline delivers explainable outputs with traceable decision rationales. In real-world validation, 92 percent of oncology patients had at least one relevant trial retrieved within the top 20 recommendations. Evaluation across synthetic and real clinical datasets confirmed state-of-the-art performance, with expert assessment validating over 90 percent accuracy in criterion-level eligibility classification, particularly excelling in biomarker-driven matches. Designed for modularity and privacy, TrialMatchAI supports Phenopackets-standardized data, enables secure local deployment, and allows seamless replacement of LLM components as more advanced models emerge. By enhancing efficiency and interpretability and offering lightweight, open-source deployment, TrialMatchAI provides a scalable solution for AI-driven clinical trial matching in precision medicine.

Problem

• Automates patient-to-trial matching using AI and clinical data
• Ensures transparency and reproducibility with lightweight deployment
• Improves efficiency and accuracy in clinical trial recruitment

Innovation

• Uses fine-tuned LLMs for clinical data processing
• Hybrid search strategy combining lexical and semantic similarity
• Modular design with privacy-focused local deployment