SiamQuality: A ConvNet-Based Foundation Model for Imperfect Physiological Signals

TL;DR

This paper introduces SimQuality, a CNN-based foundation model trained on large-scale noisy physiological data, demonstrating improved robustness and performance across multiple heart monitoring tasks from wearable device data.

Contribution

The study presents a novel self-supervised learning framework, SimQuality, for developing foundation models on noisy physiological signals using CNNs, with extensive pre-training and validation.

Findings

Outperforms existing methods on six downstream tasks

Pre-trained on over 36 million PPG pairs

Demonstrates CNNs as effective backbone for noisy data

Abstract

Foundation models, especially those using transformers as backbones, have gained significant popularity, particularly in language and language-vision tasks. However, large foundation models are typically trained on high-quality data, which poses a significant challenge, given the prevalence of poor-quality real-world data. This challenge is more pronounced for developing foundation models for physiological data; such data are often noisy, incomplete, or inconsistent. The present work aims to provide a toolset for developing foundation models on physiological data. We leverage a large dataset of photoplethysmography (PPG) signals from hospitalized intensive care patients. For this data, we propose SimQuality, a novel self-supervised learning task based on convolutional neural networks (CNNs) as the backbone to enforce representations to be similar for good and poor quality signals that are from similar physiological states. We pre-trained the SimQuality on over 36 million 30-second PPG pairs and then fine-tuned and tested on six downstream tasks using external datasets. The results demonstrate the superiority of the proposed approach on all the downstream tasks, which are extremely important for heart monitoring on wearable devices. Our method indicates that CNNs can be an effective backbone for foundation models that are robust to training data quality.