Stanford University (Ashley Lab)
Multi-task model that jointly assesses signal quality and detects atrial fibrillation from wrist-worn wearable photoplethysmography (PPG), pretrained on roughly one million simulated unlabeled signals before fine-tuning on labeled wearable data. Uses a 1D CNN with separate output heads for signal quality and arrhythmia detection. Developed by Stanford's Ashley Lab.
Architecture
CNN (1D)
1D CNN with unsupervised pretraining on simulated PPG followed by multi-task supervised fine-tuning (signal-quality head + arrhythmia head)
Framework
Keras
Added to catalog
2026-07-10
GPL 3.0
License for model weights only. Associated code may be licensed seperately, check code source for specific terms.
DeepBeat Ambulatory Monitoring External Test Cohort
15-patient ambulatory monitoring dataset used for external evaluation.
~1 million simulated unlabeled PPG signals plus >500K labeled signals from >100 individuals across 3 wearable devices.
Binary classification