Stanford University / Cedars-Sinai Medical Center (Ouyang Lab)
Pediatric-specific extension of EchoNet-Dynamic: a video-based deep learning model that segments the left ventricle and estimates ejection fraction (EF) from apical-4-chamber (A4C) and parasternal short-axis (PSAX) pediatric echocardiogram clips. Because adult-trained echo models generalize poorly to children (who vary widely in heart size, rate, and image quality), EchoNet-Peds was trained from scratch on a dedicated pediatric video dataset. It segments the LV with a Dice similarity coefficient of 0.89 in both views, estimates EF with a mean absolute error of 3.66%, and identifies pediatric systolic dysfunction with an AUC of 0.95, significantly outperforming an adult-trained model applied to the same pediatric data.
Architecture
Hybrid
DeepLabV3-style 2D CNN for frame-wise LV semantic segmentation in A4C and PSAX views, aggregated with a 3D-CNN video-classification component to estimate ejection fraction by the 5/6 area-length method
Framework
PyTorch
Added to catalog
2026-08-13
Other
License for model weights only. Associated code may be licensed seperately, check code source for specific terms.
1,923 children (43% female), ages 0-18 years, drawn from routine care at Lucile Packard Children's Hospital Stanford.
LV segmentation in pediatric A4C and PSAX echo views
Ejection fraction (5/6 area-length method) in pediatric patients
Detection of reduced EF (<55%) / pediatric systolic dysfunction