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
Local Anesthetics: Clinical Application as Epidural Anesthesia01:29

Local Anesthetics: Clinical Application as Epidural Anesthesia

411
Epidural anesthetics are administered in the fat-filled epidural space, the outermost part of the spinal canal. This technique is commonly employed for pain management and anesthesia during lower abdomen and pelvis surgeries or labor and delivery.
Since epidural anesthetics can be infused through an epidural catheter, all types of drugs, including short-acting ones, can be administered. Chloroprocaine and lidocaine are examples of short and long-duration anesthetics, respectively. Bupivacaine...
411
Nursing Clinical Information System01:27

Nursing Clinical Information System

753
Nursing Clinical Information System (NCIS)
A Nursing Clinical Information System (NCIS) is a specialized type of healthcare information system tailored to meet the unique needs of nursing practice. It incorporates the principles of nursing informatics to streamline information management and improve the quality of care delivery.
Critical attributes of NCIS include:
753
Local Anesthetics: Clinical Application as Spinal Anesthesia01:11

Local Anesthetics: Clinical Application as Spinal Anesthesia

582
Spinal anesthetics are given during lower abdomen and limb surgeries to block sensory and motor neurons. They are administered in the mid to low lumbar regions, primarily acting on the cauda equina's nerve roots. The blockade level depends on the local anesthetic (LA) concentration. Usually, low LA concentrations are sufficient to block sensory fibers, while only high LA concentrations block motor fibers. Other factors like injection volume and speed, the patient's posture, and the drug...
582

You might also read

Related Articles

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

Sort by
Same author

Effect of Safety Management Program on Critical Incidents During Patient Transport: Experience From Single Critical Care Transport System.

Journal of patient safety·2026
Same author

A simulation study on optimizing stationary camera locations for physician-directed telemedical supervision of prehospital advanced life support: Single-versus dual-camera setups.

Medicine·2026
Same author

Impact of the COVID-19 Pandemic on Emergency Department Practices for Cardiopulmonary Symptoms.

Journal of clinical medicine·2026
Same author

Impact of the COVID-19 Pandemic on the Mortality of Traumatic Brain Injury Patients Transported by Emergency Medical Services.

Asia-Pacific journal of public health·2026
Same author

Artificial Intelligence Analysis of Chest Radiographs for Predicting Major Adverse Events in Patients Visiting the Emergency Department With Acute Cardiopulmonary Symptoms.

Korean journal of radiology·2025
Same author

Enhanced Aerosol Containment Performance of a Negative Pressure Hood with an Aerodynamic Cap Design: Multi-Method Validation Using CFD, PAO Particles, and Microbial Testing.

Bioengineering (Basel, Switzerland)·2025

Related Experiment Video

Updated: Jun 6, 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

502

Using Large Language Models to Extract Core Injury Information From Emergency Department Notes.

Dong Hyun Choi1,2, Yoonjic Kim2,3,4, Sae Won Choi5

  • 1Department of Biomedical Engineering, Seoul National University College of Medicine, Seoul, Korea.

Journal of Korean Medical Science
|December 3, 2024
PubMed
Summary

Large language models (LLMs) can effectively extract injury data from emergency department (ED) notes, improving surveillance. This study shows LLMs offer a versatile solution for injury information extraction tasks.

Keywords:
Clinical NoteEmergency DepartmentInformation ExtractionInjuriesLarge Language Model

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
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
07:35

A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

Published on: October 13, 2023

1.6K

Related Experiment Videos

Last Updated: Jun 6, 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

502
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
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
07:35

A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

Published on: October 13, 2023

1.6K

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Healthcare
  • Public Health Surveillance

Background:

  • Injuries represent a major global health burden with high incidence and mortality.
  • Effective injury surveillance is crucial for prevention but is resource-intensive.
  • Developing automated methods for injury data extraction from clinical notes is needed.

Purpose of the Study:

  • To develop and validate locally deployable large language models (LLMs) for extracting core injury information from Emergency Department (ED) clinical notes.
  • To assess the performance of a generalizable Llama-2 model and Bidirectional Encoder Representations from Transformers (BERT) models in injury data extraction.
  • To compare the accuracy and efficiency of LLMs against traditional methods for injury surveillance.

Main Methods:

  • A diagnostic study utilized retrospective data from January 2014 to December 2020 from two urban academic tertiary hospitals.
  • A generalizable Llama-2 model and five BERT models were fine-tuned for information extraction tasks (mechanism, place, activity, intent, severity) from ED notes.
  • Model performance was evaluated using accuracy and macro-average F1 scores, with injury registry data serving as the gold standard.

Main Results:

  • The Llama-2 model demonstrated high accuracies across tasks, including injury mechanism (0.899), intent (0.972), and severity (0.935).
  • Llama-2 consistently outperformed BERT models in both accuracy and macro-average F1 scores across all extraction tasks in both derivation and test cohorts.
  • Constraining the Llama-2 model to avoid uncertain predictions further enhanced its accuracy.

Conclusions:

  • Locally deployable LLMs show strong performance in extracting key injury-related information from unstructured ED clinical notes.
  • Generative LLMs offer a versatile and efficient solution for diverse injury information extraction needs.
  • These findings support the integration of LLMs into public health surveillance systems for improved injury prevention efforts.