Pediatric echocardiography, or cardiac ultrasound, is the most widely used and readily available imaging modality to assess and manage congenital and acquired heart disease in children. Echocardiography is portable, efficient, and non-invasive, while still providing high quality imaging, making it the foremost frontline diagnostic tool in the United States for cardiac disease. Assessment of left ventricular function is of paramount importance in monitoring disease progression and targeting treatment for a wide range of pediatric diseases, including patients with cancer receiving chemotherapy, arrythmia management, heart failure, post-surgical ventricular function, genetic abnormalities, and acquired heart disease. In addition to our deep learning model, we introduce a new large video dataset of echocardiograms for computer vision research. The EchoNet-Peds database includes 7,643 labeled echocardiogram videos and human expert annotations (measurements, tracings, and calculations) to provide a baseline to study cardiac motion and chamber sizes. The database includes patients ranging from 0-18 years (43% female) with a wide range of sizes.
Machine learning has shown significant promise in adults, including the ability to improve the assessment of left ventricular function, as demonstrated by EchoNet-Dynamic. However, studies in pediatrics are often limited due to the challenges associated with obtaining large scale datasets and the rarity of congenital and acquired heart disease. Children also have different environmental factors to take into consideration, such as broader range of size, age, heart rate, and patient cooperation, which are known to impact image quality. Machine learning in pediatric cardiology has been somewhat limited by all the aforementioned factors, as well as the lack of open databases that allow for collaboration, ensure reproducibility, and create infrastructure to promote generalizability without bias. Moreover, it is unclear that models trained only on adult data would be adequate for pediatric patients. With EchoNet-Peds, we set out to build upon the work of EchoNet-Dynamic by training and validating a model solely on pediatric data, and adding echocardiographic views that are pertinent to the pediatric community for assessment of left ventricular function. We present the EchoNet-Peds dataset of 7,643 echocardiography videos, spanning the range of typical echocardiography lab imaging acquisition conditions, with corresponding labeled measurements including ejection fraction, left ventricular volume at end-systole and end-diastole, and human expert tracings of the left ventricle as an aid for studying machine learning approaches to evaluate cardiac function. We additionally present the performance of our model with 3-dimensional convolutional neural network architecture for video classification. This model is used semantically segment the left ventricle and to assess ejection fraction to expert human performance and as a benchmark for further collaboration, comparison, and creation of task-specific architectures. To the best of our knowledge, this is the largest labeled pediatric medical video dataset made available publicly to researchers and medical professionals.
Echocardiogram Videos: A standard full resting echocardiogram study consists of a series of videos and images visualizing the heart from different angles, positions, and image acquisition techniques. The dataset contains 3,176 apical-4-chamber echocardiography videos and 4,424 parasternal short axis echocardiography videos from individuals who underwent imaging between 2014 and 2021 as part of routine clinical care at Lucile Packard Children’s Hospital at Stanford. Each video was cropped and masked to remove text and information outside of the scanning sector. The resulting images were then downsampled by cubic interpolation into standardized 112x112 pixel videos.
Measurements: In addition to the video itself, each study is linked to clinical measurements and calculations obtained by a registered sonographer and verified by an expert physician echocardiographer in the standard clinical workflow. A central metric of cardiac function is the left ventricular ejection fraction, which is used to diagnose cardiomyopathy, assess eligibility for certain chemotherapies, and determine indication for medical devices. The ejection fraction method utilized for the pediatric dataset is the “Bullet method” or the “5/6 Area Length Method,” where the left ventricular volumes are derived from both the apical and parasternal short axis views. Ejection fraction is expressed as a percentage and is the ratio of left ventricular end systolic volume (ESV) and left ventricular end diastolic volume (EDV) determined by (EDV - ESV) / EDV.
In our dataset, for each video, the left ventricle is traced at the endocardial border at two separate time points representing end-systole and end-diastole. Each tracing is used to estimate ventricular volume by integration of ventricular area over the length of the major axis of the ventricle. The expert tracings are represented by a collection of paired coordinates corresponding to each human tracing. The first pair of coordinates represent the length and direction of the long axis of the left ventricle, and subsequent coordinate pairs represent short axis linear distances starting from the apex of the heart to the mitral apparatus. Each coordinate pair is also listed with a video file name and frame number to identify the representative frame from which the tracings match.
Our code is available here.
By registering for downloads from the EchoNet-Pediatric Dataset, you are agreeing to this Research Use Agreement, as well as to the Terms of Use of the Stanford University School of Medicine website as posted and updated periodically at http://www.stanford.edu/site/terms/.
1. Permission is granted to view and use the EchoNet-Pediatric Dataset without charge for personal, non-commercial research purposes only. Any commercial use, sale, or other monetization is prohibited.
2. Other than the rights granted herein, the Stanford University School of Medicine (“School of Medicine”) retains all rights, title, and interest in the EchoNet-Pediatric Dataset.
3. You may make a verbatim copy of the EchoNet-Pediatric Dataset for personal, non-commercial research use as permitted in this Research Use Agreement. If another user within your organization wishes to use the EchoNet-Pediatric Dataset, they must register as an individual user and comply with all the terms of this Research Use Agreement.
4. YOU MAY NOT DISTRIBUTE, PUBLISH, OR REPRODUCE A COPY of any portion or all of the EchoNet-Pediatric Dataset to others without specific prior written permission from the School of Medicine.
5. YOU MAY NOT SHARE THE DOWNLOAD LINK to the EchoNet-Pediatric dataset to others. If another user within your organization wishes to use the EchoNet-Pediatric Dataset, they must register as an individual user and comply with all the terms of this Research Use Agreement.
6. You must not modify, reverse engineer, decompile, or create derivative works from the EchoNet-Pediatric Dataset. You must not remove or alter any copyright or other proprietary notices in the EchoNet-Pediatric Dataset.
7. The EchoNet-Pediatric Dataset has not been reviewed or approved by the Food and Drug Administration, and is for non-clinical, Research Use Only. In no event shall data or images generated through the use of the EchoNet-Pediatric Dataset be used or relied upon in the diagnosis or provision of patient care.
8. THE ECHONET-PEDIATRIC DATASET IS PROVIDED "AS IS," AND STANFORD UNIVERSITY AND ITS COLLABORATORS DO NOT MAKE ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, NOR DO THEY ASSUME ANY LIABILITY OR RESPONSIBILITY FOR THE USE OF THIS ECHONET-PEDIATRIC DATASET.
9. You will not make any attempt to re-identify any of the individual data subjects. Re-identification of individuals is strictly prohibited. Any re-identification of any individual data subject shall be immediately reported to the School of Medicine.
10. Any violation of this Research Use Agreement or other impermissible use shall be grounds for immediate termination of use of this EchoNet-Pediatric Dataset. In the event that the School of Medicine determines that the recipient has violated this Research Use Agreement or other impermissible use has been made, the School of Medicine may direct that the undersigned data recipient immediately return all copies of the EchoNet-Pediatric Dataset and retain no copies thereof even if you did not cause the violation or impermissible use.
In consideration for your agreement to the terms and conditions contained here, Stanford grants you permission to view and use the EchoNet-Pediatric Dataset for personal, non-commercial research. You may not otherwise copy, reproduce, retransmit, distribute, publish, commercially exploit or otherwise transfer any material.
You may use EchoNet-Pediatric Dataset for legal purposes only.
You agree to indemnify and hold Stanford harmless from any claims, losses or damages, including legal fees, arising out of or resulting from your use of the EchoNet-Pediatric Dataset or your violation or role in violation of these Terms. You agree to fully cooperate in Stanford’s defense against any such claims. These Terms shall be governed by and interpreted in accordance with the laws of California.
Video-Based Deep Learning for Automated Assessment of Left Ventricular Ejection Fraction in Pediatric Patients.
Charitha Reddy, Leo Lopez, David Ouyang, James Y. Zou, Bryan He Journal of the American Society of Echocardiography (2022)
For inquiries, contact us at reddyc@stanford.edu.