Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Methods of Documentation VII: EMR01:30

Methods of Documentation VII: EMR

824
Electronic Medical Records (EMRs) primarily center around electronically documenting patients' health information within a single healthcare organization or practice. They contain essential clinical data related to a patient's medical history, diagnoses, medications, treatment plans, lab results, and other pertinent information relevant to the specific encounter or episode of care. EMRs are designed to streamline documentation and workflow processes within individual healthcare...
824
Issues And Trends In Healthcare Delivery System01:29

Issues And Trends In Healthcare Delivery System

5.6K
The issues and trends in healthcare delivery are constantly changing. The COVID-19 pandemic is one recent issue that wreaked havoc on healthcare systems, causing a shortage of healthcare workers, high demand for medicines and supplies, and increased medical expenditure due to a lack of insurance. Other issues include rising healthcare costs and care fragmentation.
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
5.6K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Comparison of Cannulation Strategies for Minimally Invasive Mitral Valve Surgery.

Interdisciplinary cardiovascular and thoracic surgery·2026
Same author

Open-source large language model-based on-premises pipeline for automated data extraction from unstructured electronic health records: a pilot study.

BMJ health & care informatics·2026
Same author

Mind the Gap: Inadequate Performance of Embolic Risk Scores in Infective Endocarditis.

Open forum infectious diseases·2026
Same author

Influence of Human Factors on Coagulation Management in Cardiac Surgery: A Retrospective Analysis.

Journal of cardiothoracic and vascular anesthesia·2026
Same author

High-Sensitivity Cardiac Troponin T-gen6 Assay in Suspected Myocardial Infarction: Diagnostic Accuracy, Cutoffs, and Clinical Implications.

Journal of the American College of Cardiology·2026
Same author

Rupture of a Nonaneurysmal Left Sinus of Valsalva Into the Pericardium.

JACC. Case reports·2026

Related Experiment Video

Updated: Jun 1, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

494

Large language models for data extraction from unstructured and semi-structured electronic health records: a multiple

Vasileios Ntinopoulos1,2, Hector Rodriguez Cetina Biefer1,2, Igor Tudorache1,2

  • 1Department of Cardiac Surgery, University Hospital of Zurich, Zurich, Switzerland.

BMJ Health & Care Informatics
|January 20, 2025
PubMed
Summary

Top large language models (LLMs) demonstrate excellent performance in extracting data from electronic health records. These advanced LLMs show high accuracy and consistency, paving the way for improved healthcare research and efficiency.

Keywords:
Artificial intelligenceElectronic Data ProcessingElectronic Health RecordsHealth Information Management

More Related Videos

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts
07:50

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

15.8K
TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients
09:00

TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients

Published on: April 13, 2021

4.4K

Related Experiment Videos

Last Updated: Jun 1, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

494
A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts
07:50

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

15.8K
TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients
09:00

TBase - an Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients

Published on: April 13, 2021

4.4K

Area of Science:

  • Artificial Intelligence in Medicine
  • Natural Language Processing for Healthcare

Background:

  • Electronic health records (EHRs) contain vast amounts of unstructured and semi-structured data.
  • Efficient data extraction from EHRs is crucial for research and clinical decision-making.
  • Current methods for EHR data extraction can be time-consuming and labor-intensive.

Purpose of the Study:

  • To evaluate the performance of various large language models (LLMs) in extracting data from unstructured and semi-structured EHRs.
  • To compare LLM performance against a baseline transformer model.
  • To assess the consistency of LLM responses in data extraction tasks.

Main Methods:

  • 50 synthetic English medical notes with structured and unstructured components were created and expert-validated.
  • 18 LLMs were prompted using these notes for data extraction and classification tasks.
  • Performance was measured by accuracy in four entity extraction and five binary classification tasks.
  • Response consistency was evaluated over three identical prompt iterations.

Main Results:

  • Several LLMs, including Claude 3.0 Opus, GPT 4, and Gemini Advanced, achieved excellent overall accuracy exceeding 0.98.
  • These top-performing LLMs significantly outperformed the baseline RoBERTa model (0.742 accuracy).
  • High response consistency (Krippendorff's alpha values near 1) was observed for the leading LLMs across multiple runs.

Conclusions:

  • Claude 3.0 Opus, Claude 3.0 Sonnet, Claude 2.0, GPT 4, Claude 2.1, Gemini Advanced, PaLM 2 chat-bison, and Llama 3-70b demonstrated superior performance in EHR data extraction.
  • These LLMs offer reliable data extraction capabilities for both entity extraction and classification tasks.
  • Further validation with real-world EHR data is recommended to confirm these promising results.