Ian Shapiro

Papers

• Wearable Accelerometer Foundation Models for Health via Knowledge Distillation
  2024 · arXiv preprint

  This Apple paper shows that an accelerometer foundation model can predict many health targets when it learns from a PPG model through knowledge distillation. Training used 20 million minutes of unlabelled data from about 172K people in the Apple Heart and Movement Study. Accelerometer embeddings retrieved the matching PPG embedding with 99.2% top-1 accuracy, and distilled encoders beat accelerometer-only ones by 23% to 49%.

  • Published: ArXiv preprint, 2024
  • Signals: PPG, Accelerometer
  • Result: The authors observe strong cross-modal alignment on unseen data, e.g., 99.2% top-1 accuracy for retrieving PPG embeddings from accelerometry embeddings.
• Large-scale Training of Foundation Models for Wearable Biosignals
  2023 · arXiv preprint

  This Apple paper trains foundation models on large amounts of wearable PPG and ECG. It starts from the problem that labelled medical data are scarce and small, which limits new digital biomarkers, and uses large unlabelled wearable data to get around it.

  • Published: arXiv preprint, 2023
  • Signals: PPG, ECG
  • Method: self-supervised pretraining on large-scale wearable data

Models

• Apple accelerometer FM via PPG→ACC distillation
• Apple wearable biosignals FM

Co-authors