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

You might also read

Related Articles

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

Sort by
Same author

Bayesian Thinking in Rehabilitation Research.

American journal of physical medicine & rehabilitation·2026
Same author

Synthesized annotation guidelines are knowledge-lite boosters for clinical information extraction.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Impact of Prescribed and Self-Selected Music Interventions on Stress, Sleep, Heart Rate Variability, and Brain Connectivity in Surgeons Using 7-Tesla Functional Magnetic Resonance Imaging and Wearable Actigraphy: Multimodal Feasibility Randomized Controlled Trial.

JMIR formative research·2026
Same author

Delirium Leads to Poor In-Hospital and 90-Day Outcomes Among Patients With Acute Ischemic Stroke With and Without Intravenous Thrombolysis or Intraarterial Therapy.

Stroke (Hoboken, N.J.)·2026
Same author

Stroke Severity Mediates the Association Between Socioeconomic Disadvantage and Poor Outcomes Among Patients With Acute Ischemic Stroke.

Stroke (Hoboken, N.J.)·2026
Same author

Association of Corticospinal Tract Asymmetry With Ambulatory Ability After Intracerebral Hemorrhage.

Annals of clinical and translational neurology·2025

Related Experiment Video

Updated: Jun 23, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.0K

Extraction of Radiological Characteristics From Free-Text Imaging Reports Using Natural Language Processing Among

Enshuo Hsu1,2, Abdulaziz T Bako1, Thomas Potter1

  • 1Center for Health Data Science and Analytics, Houston Methodist Research Institute, Houston, TX, United States.

JMIR AI
|June 14, 2024
PubMed
Summary

A new natural language processing (NLP) pipeline accurately extracts stroke features from head CT reports. This tool helps analyze stroke severity and survival outcomes, improving medical research and patient safety.

Keywords:
cerebral hemorrhagecomputed tomographydeep learningelectronic health recordsischemic strokenatural language processingneuroimagingradiologystroke

More Related Videos

A Magnetic Resonance Imaging Protocol for Stroke Onset Time Estimation in Permanent Cerebral Ischemia
09:59

A Magnetic Resonance Imaging Protocol for Stroke Onset Time Estimation in Permanent Cerebral Ischemia

Published on: September 16, 2017

14.1K
Author Spotlight: Integrated Photoacoustic, Ultrasound, and Angiographic Tomography (PAUSAT) for NonInvasive Whole-Brain Imaging of Ischemic Stroke
06:45

Author Spotlight: Integrated Photoacoustic, Ultrasound, and Angiographic Tomography (PAUSAT) for NonInvasive Whole-Brain Imaging of Ischemic Stroke

Published on: June 2, 2023

1.3K

Related Experiment Videos

Last Updated: Jun 23, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.0K
A Magnetic Resonance Imaging Protocol for Stroke Onset Time Estimation in Permanent Cerebral Ischemia
09:59

A Magnetic Resonance Imaging Protocol for Stroke Onset Time Estimation in Permanent Cerebral Ischemia

Published on: September 16, 2017

14.1K
Author Spotlight: Integrated Photoacoustic, Ultrasound, and Angiographic Tomography (PAUSAT) for NonInvasive Whole-Brain Imaging of Ischemic Stroke
06:45

Author Spotlight: Integrated Photoacoustic, Ultrasound, and Angiographic Tomography (PAUSAT) for NonInvasive Whole-Brain Imaging of Ischemic Stroke

Published on: June 2, 2023

1.3K

Area of Science:

  • Medical informatics
  • Natural Language Processing (NLP)
  • Stroke imaging analysis

Background:

  • Neuroimaging is crucial for stroke diagnosis, but unstructured reports hinder data extraction from electronic health records.
  • Extracting specific stroke imaging features using NLP requires systematic evaluation.

Purpose of the Study:

  • To develop and evaluate an NLP pipeline for extracting 13 stroke features from head CT imaging notes.
  • To utilize a ClinicalBERT model with domain-specific pretraining and fine-tuning for stroke feature extraction.

Main Methods:

  • A HeadCT_BERT model was pretrained on 82,073 head CT notes and fine-tuned on 200 annotated notes.
  • The model extracted stroke features for 24,924 stroke patients, creating structured datasets.
  • Kaplan-Meier curves and log-rank tests compared survival between patients with and without severe stroke features.

Main Results:

  • The HeadCT_BERT model achieved high performance: AUC of 0.9831, F1-score of 0.8683, and 97% accuracy.
  • Severe stroke features (midline shift, edema, mass effect) in initial notes were associated with significantly lower survival probability in acute ischemic stroke patients (P<.001).

Conclusions:

  • The developed NLP pipeline demonstrates high performance in extracting stroke imaging features.
  • This approach has the potential to significantly advance medical research and enhance patient safety in stroke care.