Photoplethysmography model learns features from short to long signals

RAE-PPG: Duration-Grounded Retain-and-Extend Pretraining for PPG Foundation Models

Machine Learning

Summary

Different features of pulse signals from photoplethysmography (PPG) need varying amounts of time to measure. The authors propose a method where a single model learns features progressively from short to long signal durations, reusing what it learned before while adding new information. They train a special Transformer model on progressively longer signals (10, 30, and 240 seconds), ensuring that features from shorter signals stay accessible when analyzing longer ones. Their approach improves performance on many tasks compared to other models.

What this means in practice

  • For medical device developers: Develop better PPG-based health monitors by using models that analyze features reliably across multiple signal durations.$Commercial implications: Enables advanced wearable and clinical PPG sensors that interpret varied duration signals for improved health monitoring.
  • For biomedical data engineers: Improve pipelines that analyze PPG signals by incorporating models trained to retain and extend features over different time spans.

Authors

Suyeong Lee, Hochang Lee, Seokyong Sheem, Daekyum Kim

Abstract

Signal features derived from photoplethysmography (PPG) require different signal durations to characterize. Existing PPG foundation models treat duration as a pretraining or evaluation condition rather than using the different durations required by PPG features to organize self-supervision. We hypothesize that self-supervision should expand with signal duration, allowing a single encoder to progressively acquire additional features while preserving and reusing earlier learning. We introduce Retain-and-Extend PPG (RAE-PPG), which trains a single Transformer encoder successively on 10 s, 30 s, and 240 s inputs, adding supervision for signal features supported by each longer observation. The encoder is partitioned into duration-specific parameter groups, allowing later stages to reuse earlier groups while updating only the group assigned to the current stage. Selected earlier targets are reused to supervise later stages, encouraging the corresponding features to remain accessible in longer-input representations. Direct decoding from the final encoder shows that earlier features remain recoverable from longer-input representations, while later-stage features show higher mean decoding performance at their introduction durations. Controlled comparisons further show that prior-stage learning provides a better basis for learning newly introduced features at both transitions. Across 18 tasks from eight datasets, the final frozen encoder achieves the best observed score on 12 tasks compared with five existing PPG foundation models.