Large-scale Training of Foundation Models for Wearable Biosignals
read the original abstract
Tracking biosignals is crucial for monitoring wellness and preempting the development of severe medical conditions. Today, wearable devices can conveniently record various biosignals, creating the opportunity to monitor health status without disruption to one's daily routine. Despite widespread use of wearable devices and existing digital biomarkers, the absence of curated data with annotated medical labels hinders the development of new biomarkers to measure common health conditions. In fact, medical datasets are usually small in comparison to other domains, which is an obstacle for developing neural network models for biosignals. To address this challenge, we have employed self-supervised learning using the unlabeled sensor data collected under informed consent from the large longitudinal Apple Heart and Movement Study (AHMS) to train foundation models for two common biosignals: photoplethysmography (PPG) and electrocardiogram (ECG) recorded on Apple Watch. We curated PPG and ECG datasets from AHMS that include data from ~141K participants spanning ~3 years. Our self-supervised learning framework includes participant level positive pair selection, stochastic augmentation module and a regularized contrastive loss optimized with momentum training, and generalizes well to both PPG and ECG modalities. We show that the pre-trained foundation models readily encode information regarding participants' demographics and health conditions. To the best of our knowledge, this is the first study that builds foundation models using large-scale PPG and ECG data collected via wearable consumer devices $\unicode{x2013}$ prior works have commonly used smaller-size datasets collected in clinical and experimental settings. We believe PPG and ECG foundation models can enhance future wearable devices by reducing the reliance on labeled data and hold the potential to help the users improve their health.
This paper has not been read by Pith yet.
Forward citations
Cited by 12 Pith papers
-
Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning
xMAE pretrains biosignal representations via masked cross-modal reconstruction of temporally ordered signals like ECG and PPG, outperforming baselines on 15 of 19 downstream tasks including cardiovascular prediction a...
-
Device Passport: Enabling Spatio-Temporal Pretrained Models to Generalize Across Input Layouts
Device Passport improves cross-layout transfer for biosignal models by learning expert mixture models from each channel's functional activity and metadata, outperforming baselines in transfer regimes.
-
Next-Token Prediction Learns Generalisable Representations of Sleep Physiology
Next-token prediction on multi-modal tokenized sleep signals yields embeddings that match supervised performance with far less labels and generalize to daytime heart data.
-
A robust PPG foundation model using multimodal physiological supervision
A PPG foundation model pretrained via multimodal ECG/respiratory contrastive sample selection on ICU data improves performance on 14 of 15 downstream tasks including field-like data while using 3x fewer subjects.
-
AURORA: Contextual Orthogonalization for Geometric Representation Learning in Healthcare Foundation Models
AURORA is a representation learning framework that uses contextual orthogonalization and relational alignment to create disentangled, geometrically interpretable latent spaces in healthcare foundation models.
-
Event Fields: Learning Latent Event Structure for Waveform Foundation Models
Event-centric waveform foundation models are learned via self-supervised consistency on latent event structures and interactions, yielding improved performance and label efficiency over sequence-based baselines on phy...
-
Emergent Symbolic Structure in Health Foundation Models: Extraction, Alignment, and Cross-Modal Transfer
Health foundation model embeddings contain an interpretable symbolic organization shared across modalities that supports cross-domain transfer without joint training.
-
OmniMouse: Scaling properties of multi-modal, multi-task Brain Models on 150B Neural Tokens
OmniMouse demonstrates data-driven scaling in multi-task brain models on a 150B-token neural dataset, achieving SOTA across prediction, decoding, and forecasting while model size gains saturate.
-
Uncertainty-Aware Foundation Models for Clinical Data
The work introduces uncertainty-aware foundation models for clinical data by learning set-valued patient representations that enforce consistency across partial observations and integrate multimodal self-supervised ob...
-
WISTERIA: Learning Clinical Representations from Noisy Supervision via Multi-View Consistency in Electronic Health Records
WISTERIA learns robust clinical representations from noisy EHR labels by enforcing consistency across multiple weak supervision views plus ontology regularization.
-
FunduSegmenter: Leveraging the RETFound Foundation Model for Joint Optic Disc and Optic Cup Segmentation in Retinal Fundus Images
FunduSegmenter adapts RETFound with custom adapters and attention modules to reach 90.51% average Dice score for optic disc and cup segmentation, outperforming baselines on internal and external tests across five datasets.
-
The Bioelectrical Information Theory: Investigating the theoretical compression limit of bioelectrical signals under artificial intelligence
Bioelectrical compression is reframed as a model- and task-conditioned quantity via a three-level hierarchy of signal purification, physiological encoding, and semantic reduction using deep learning.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.