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 Experiment Videos

EHR-based phenotyping: Bulk learning and evaluation.

Po-Hsiang Chiu1, George Hripcsak1

  • 1Department of Biomedical Informatics, Columbia University, 622 W. 168th Street, New York, NY, USA.

Journal of Biomedical Informatics
|April 16, 2017
PubMed
Summary
This summary is machine-generated.

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

A comparison of Fast Healthcare Interoperability Resources and Observational Medical Mutcomes Partnership electronic health record data within the All of Us Research Program.

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

Heterogeneity of Treatment Effects Across Nine Glucose-Lowering Drug Classes in Type 2 Diabetes: Extension of the LEGEND-T2DM Network Study.

Diabetes, obesity & metabolism·2026
Same author

Semaglutide and Neovascular Age-Related Macular Degeneration Among Adults with Type 2 Diabetes: An OHDSI Network Study.

Ophthalmology·2026
Same author

Comparative Cardiovascular Effectiveness of Glucagon-Like Peptide 1 Receptor Agonists and Sodium-Glucose Cotransporter 2 Inhibitors in Diabetes Mellitus.

Journal of the American College of Cardiology·2026
Same author

Toward AI-Powered Cancer Etiology Research.

Cancer discovery·2026
Same author

Real-world evidence for comparative safety of second-line antihyperglycemic agents in older adults with type 2 diabetes.

Nature communications·2026

This study introduces bulk learning for efficient data-driven phenotyping from electronic health records (EHR). This method uses ensemble learning to identify disease cohorts with sparse data, improving scalability and accuracy.

Area of Science:

  • Computational biology
  • Medical informatics
  • Machine learning

Background:

  • Data-driven phenotyping from electronic health records (EHR) is crucial for cohort stratification.
  • Conventional methods rely on manual knowledge engineering or supervised learning, which are costly, time-consuming, and lack scalability.
  • Rare or acute conditions pose challenges due to insufficient data for accurate statistical modeling.

Purpose of the Study:

  • To develop a scalable and efficient method for data-driven phenotyping using electronic health records (EHR).
  • To address the limitations of manual annotation and data scarcity in identifying clinical phenotypes, particularly for rare diseases.
  • To demonstrate a hierarchical learning approach for bulk phenotyping using ensemble learning and feature abstraction.

Main Methods:

Keywords:
Disease modelingEHR phenotypingEnsemble learningFeature learningKnowledge representationStacked generalization

Related Experiment Videos

  • A hierarchical learning method based on ensemble learning is proposed for feature abstraction.
  • Bulk learning framework trains and evaluates multiple phenotypes simultaneously using a sparse annotation set.
  • Disease cohort definitions are learned from an abstract feature space derived from multiple diseases and diagnostic codes as surrogates.
  • Surrogate labels enable model training and evaluation with sparse annotated samples.

Main Results:

  • The proposed method allows for training and evaluation using a sparse annotated sample.
  • The approach enables the creation of statistical models from an abstract feature space of low dimensionality.
  • Shared clinical traits of target diseases are encapsulated within the abstract feature space for effective bulk learning.

Conclusions:

  • The bulk learning framework offers a scalable solution for data-driven phenotyping from EHR.
  • This approach effectively handles data scarcity and improves the identification of diverse clinical phenotypes.
  • The method demonstrates the potential of feature abstraction and ensemble learning in medical informatics.