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

nanoASM: Long-Read Allele-Specific DNA Methylation Profiling Enables Functional Annotation of Regulatory Noncoding Variants in Human Prostate Tissues.

bioRxiv : the preprint server for biology·2026
Same author

Proteomics risk scores and mortality in heart failure: Generalizability across populations.

PloS one·2026
Same author

Personal Care Products and Incident Hypertension: Prospective Cohort Study of U.S. Women.

medRxiv : the preprint server for health sciences·2026
Same author

Population Estimates and Hypertension and Diabetes Prevalence: Cross-Sectional Quantitative Study Comparing Electronic Health Record-Derived Counts, Census, and Centers for Disease Control and Prevention Population Level Analysis and Community Estimates.

JMIR public health and surveillance·2026
Same author

MRI Deep Learning for Differentiating Glioblastoma, IDH Wild-type from Central Nervous System Diffuse Large B-cell Lymphoma.

Cancer research communications·2026
Same author

Reframing AI for Rare Disease Recognition.

Research square·2026

Related Experiment Video

Updated: Nov 14, 2025

Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide
09:52

Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide

Published on: January 15, 2017

17.5K

Natural Language Processing and Machine Learning for Identifying Incident Stroke From Electronic Health Records:

Yiqing Zhao1, Sunyang Fu1, Suzette J Bielinski1

  • 1Department of Health Sciences Research, Mayo Clinic, Rochester, MN, United States.

Journal of Medical Internet Research
|March 8, 2021
PubMed
Summary

We developed a machine learning algorithm to accurately identify incident stroke cases from electronic health records, improving cardiovascular research efficiency. This tool also distinguishes stroke subtypes, offering a significant advancement in clinical data analysis.

Keywords:
electronic health recordsmachine learningnatural language processingstroke

More Related Videos

A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation
11:06

A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation

Published on: April 12, 2016

10.7K
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
06:37

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke

Published on: July 14, 2023

1.1K

Related Experiment Videos

Last Updated: Nov 14, 2025

Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide
09:52

Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide

Published on: January 15, 2017

17.5K
A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation
11:06

A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation

Published on: April 12, 2016

10.7K
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
06:37

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke

Published on: July 14, 2023

1.1K

Area of Science:

  • Cardiovascular research
  • Biomedical informatics
  • Machine learning applications in healthcare

Background:

  • Stroke is a critical cardiovascular outcome, but manual chart abstraction for incident cases is time-consuming.
  • Current electronic health record phenotyping for stroke often misses incident disease due to temporal sequence requirements.

Purpose of the Study:

  • To develop a machine learning algorithm for incident stroke ascertainment.
  • To utilize diagnosis codes, procedure codes, and natural language processing (NLP)-extracted clinical concepts.
  • To enable accurate temporal sequencing for incident disease identification.

Main Methods:

  • Trained and validated a machine learning algorithm on an epidemiology cohort (n=4914) with curated incident stroke events.
  • Compared various feature sets and machine learning classifiers, including random forest.
  • Developed a heuristic rule for stroke subtype detection (ischemic/hemorrhagic).
  • Validated the algorithm on a general population sample (n=150) from Olmsted County, Minnesota.

Main Results:

  • The best algorithm combined clinical concepts, diagnosis codes, and procedure codes using a random forest classifier.
  • Achieved 86% positive predictive value and 96% negative predictive value in a general population sample.
  • Demonstrated high accuracy for stroke subtype identification (83% in AF cohort, 80% in general population).

Conclusions:

  • A validated machine learning algorithm effectively identifies incident stroke and its subtype from electronic health records.
  • The algorithm's strong performance in a general population sample indicates its generalizability.
  • This tool has the potential for widespread adoption by healthcare institutions to streamline stroke research.