A newer version of this project is available. See below for other available versions.

Multi-view Deep Learning Improves Detection of Major Cardiac Conditions from Echocardiography

Principal Investigator(s): View help for Principal Investigator(s) Joshua Barrios, University of California, San Francisco; Geoff Tison, University of California, San Francisco

Version: View help for Version V1

Name	File Type	Size	Last Modified
data			12/16/2025 06:04:PM

Project Citation:

Barrios, Joshua, and Tison, Geoff. Multi-view Deep Learning Improves Detection of Major Cardiac Conditions from Echocardiography. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2026-01-22. https://doi.org/10.3886/E241296V1

Project Description

Summary: This dataset provides a minimal, fully de-identified example of the multiview echocardiography inputs used in “Multi-view Deep Learning Improves Detection of Major Cardiac Conditions from Echocardiography.” It contains 30 echocardiography studies, each with three standard non-Doppler views (apical four-chamber, apical two-chamber, and parasternal long-axis) along with a CSV file specifying file paths and study-level labels for ventricular abnormality. Videos were preprocessed exactly as described in the manuscript, including cropping to the echo cone, removing burned-in text, resizing to 224×224 pixels, and selecting the first 64 frames. This small dataset is intended only to illustrate the multiview data format and support reproducibility of model-loading and training code. It is not suitable for clinical modeling. All clips are fully de-identified and contain no protected health information.

Scope of Project

Subject Terms: echocardiogram; cardiology; ventricular dysfunction

Geographic Coverage: San Francisco, CA

Time Period(s): 1/1/2012 – 12/31/2020

Data Type(s): clinical data

Related Publications

Request Information

This material is sensitive in nature and is available as restricted data through ICPSR. Users are required to apply for access, will be required to pay a fee, and will experience a wait time before access is given. The material will be distributed exactly as it arrived from the data depositor. ICPSR does not check or process the material.

Published Versions

V2 [2026-01-27]

V1 [2026-01-22]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as received from the data depositor. As of April 2026, depositors are required to submit study materials in accessible formats. ICPSR has not reviewed, checked, or processed this material. For additional information about the study, please contact the investigator(s) directly. If you have questions about the accessibility of materials distributed by ICPSR or require further assistance, please visit ICPSR's Accessibility Center.