Leveraging LLMs for Structured Data Extraction from Unstructured Patient Records

📅 2025-12-03
📈 Citations: 0
Influential: 0
📄 PDF

career value

186K/year
🤖 AI Summary
In clinical research, manual extraction of structured clinical features from unstructured electronic health records (EHRs) is time-consuming, inefficient, and error-prone. To address this, we propose a privacy-preserving, modular, on-premises large language model (LLM) framework that integrates retrieval-augmented generation (RAG) with structured-output prompt engineering, enabling secure, scalable, containerized deployment in HIPAA-compliant environments. Our approach uniquely synergizes RAG with deterministic structured-response mechanisms for clinical text parsing—balancing domain adaptability and strict data privacy. Evaluated across multiple medical feature extraction tasks, the framework achieves high accuracy, substantially reducing manual annotation effort and improving data consistency. Notably, its systematic evaluation uncovered previously undetected systematic errors in prior human annotations, thereby validating both its reliability and its capacity for quality assurance and error discovery.

Technology Category

Application Category

📝 Abstract
Manual chart review remains an extremely time-consuming and resource-intensive component of clinical research, requiring experts to extract often complex information from unstructured electronic health record (EHR) narratives. We present a secure, modular framework for automated structured feature extraction from clinical notes leveraging locally deployed large language models (LLMs) on institutionally approved, Health Insurance Portability and Accountability Act (HIPPA)-compliant compute infrastructure. This system integrates retrieval augmented generation (RAG) and structured response methods of LLMs into a widely deployable and scalable container to provide feature extraction for diverse clinical domains. In evaluation, the framework achieved high accuracy across multiple medical characteristics present in large bodies of patient notes when compared against an expert-annotated dataset and identified several annotation errors missed in manual review. This framework demonstrates the potential of LLM systems to reduce the burden of manual chart review through automated extraction and increase consistency in data capture, accelerating clinical research.
Problem

Research questions and friction points this paper is trying to address.

Automates structured data extraction from unstructured patient records
Reduces manual chart review burden in clinical research
Enhances data consistency and accuracy using LLMs
Innovation

Methods, ideas, or system contributions that make the work stand out.

Uses locally deployed LLMs for secure data extraction
Integrates RAG and structured response methods in container
Achieves high accuracy across diverse clinical domains
🔎 Similar Papers
No similar papers found.
M
Mitchell A. Klusty
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY
E
Elizabeth C. Solie
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY
C
Caroline N. Leach
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY
W
W. Vaiden Logan
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY
L
Lynnet E. Richey
Spinal Cord and Brain Injury Research Center and Department of Physiology, University of Kentucky College of Medicine, Lexington, KY
J
John C. Gensel
Spinal Cord and Brain Injury Research Center and Department of Physiology, University of Kentucky College of Medicine, Lexington, KY
D
David P. Szczykutowicz
Spinal Cord and Brain Injury Research Center and Department of Physiology, University of Kentucky College of Medicine, Lexington, KY
B
Bryan C. McLellan
Spinal Cord and Brain Injury Research Center and Department of Physiology, University of Kentucky College of Medicine, Lexington, KY
E
Emily B. Collier
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY
S
Samuel E. Armstrong
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY
V
V. K. Cody Bumgardner
Center For Applied Artificial Intelligence, University of Kentucky, Lexington, KY