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

Predictive Analytics for Glaucoma Using Data From the All of Us Research Program.

Sally L Baxter1, Bharanidharan Radha Saseendrakumar1, Paulina Paul2

  • 1From the Viterbi Family Department of Ophthalmology and Shiley Eye Institute, University of California San Diego, (S.L.B., B.R.S.), La Jolla, California; UCSD Health Department of Biomedical Informatics, University of California San Diego, (S.L.B., B.R.S., P.P., J.K., L.B., T.-T.K., L.O.-M.), La Jolla, California.

American Journal of Ophthalmology
|January 26, 2021
PubMed

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

Changes in health-related social needs and in depression and anxiety symptoms in Massachusetts low-income, community health center patients: an observational cohort study.

Preventive medicine reports·2026
Same author

Return of results is important to heterogeneous research participants: A single-site survey.

Journal of clinical and translational science·2026
Same author

The All of Us Research Program's wearables dataset.

Nature medicine·2026
Same author

Blood mitochondrial heteroplasmic variants and cognitive performance in late midlife: REGARDS study.

BMC neurology·2026
Same author

Leveraging the <i>All of Us</i> Research Program to Advance Women's Health: Addressing Conditions that Affect Women Differently, Disproportionately, and Uniquely.

Journal of women's health (2002)·2026
Same author

Community Engagement in Long Covid: Insights From the Boston COVID Recovery Cohort.

Health expectations : an international journal of public participation in health care and health policy·2025
Same journal

Forging Ahead: The Need for Improved Representation in Academic Ophthalmology.

American journal of ophthalmology·2026
Same journal

Clinical Utility of Ultra-Widefield Swept-Source OCT for Intraocular Tumors: Comparison With Ultrasonography, SD-OCT, and MRI.

American journal of ophthalmology·2026
Same journal

Therapeutic Advances in Corneal Scar management: Topical Treatments, Mesenchymal Cell Therapy and Stromal Transplantation.

American journal of ophthalmology·2026
Same journal

Increased Risk for Ocular Surface Neoplasia in Recipients of Solid Organ Transplant.

American journal of ophthalmology·2026
Same journal

Aflibercept With vs Without Reduced-Fluence Photodynamic Therapy for Polypoidal Choroidal Vasculopathy: Optical Coherence Tomography Angiographic changes from a randomized clinical trial.

American journal of ophthalmology·2026
Same journal

Posterior Segment Risk Factors for Penetrating Keratoplasty Failure.

American journal of ophthalmology·2026
See all related articles
Summary
This summary is machine-generated.

New machine learning models trained on All of Us (AoU) data significantly improved predictions for glaucoma surgery needs. This demonstrates AoU

Area of Science:

  • Ophthalmology
  • Medical Informatics
  • Machine Learning

Background:

  • Glaucoma surgery prediction models are crucial for patient management.
  • Existing models may lack generalizability due to limited cohort diversity.
  • The All of Us (AoU) research program offers a large, diverse dataset for health research.

Purpose of the Study:

  • Validate a single-center glaucoma surgery prediction model using AoU data.
  • Develop and evaluate novel machine learning models with AoU data.
  • Assess the utility of AoU as a data source for ophthalmic research.

Main Methods:

  • Extracted electronic health record data for 1,231 primary open-angle glaucoma patients from AoU.
  • Externally validated a prior single-center model.

Related Experiment Videos

  • Trained new predictive models using logistic regression, neural networks, and random forests with cross-validation.
  • Evaluated model performance using AUC, accuracy, precision, and recall.

    • Main Results:

    • The single-center model showed poor performance on AoU data (AUC=0.49).
    • Models trained on AoU data achieved significantly higher AUCs, ranging from 0.80 to 0.99.
    • The AoU cohort (mean age 69.1 years) had greater representation of women (57.3%) and Black individuals (33.5%) compared to single-center cohorts.

    Conclusions:

    • Machine learning models trained with diverse, national AoU data outperform single-center models for predicting glaucoma surgery.
    • AoU data presents a valuable resource for ophthalmic research, offering advantages over claims data.
    • Further research utilizing AoU is warranted to advance glaucoma care.