Self-Supervised and Topological Signal-Quality Assessment for Any PPG Device

2025-09-15

AI Summary

Existing photoplethysmography (PPG) signal quality assessment (SQA) methods rely on fragile heuristic rules or supervised models requiring extensive labeled data, exhibiting poor robustness under motion artifacts, low perfusion, and ambient light interference. To address these limitations, we propose the first fully unsupervised PPG-SQA framework, uniquely integrating self-supervised contrastive learning with topological data analysis (TDA). Specifically, a 1D-ResNet-18 backbone extracts motion-invariant features; persistent homology computes 4D topological representations; and HDBSCAN enables parameter-free clustering. The framework is device-agnostic, requires no labeled data or manual hyperparameter tuning, and operates end-to-end. Evaluated on 10,000 PPG windows, it achieves a silhouette coefficient of 0.72—significantly outperforming established baselines—demonstrating strong consistency, generalizability, and practical applicability across diverse physiological and environmental conditions.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Motion & Tracking

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content

Abstract

Wearable photoplethysmography (PPG) is embedded in billions of devices, yet its optical waveform is easily corrupted by motion, perfusion loss, and ambient light, jeopardizing downstream cardiometric analytics. Existing signal-quality assessment (SQA) methods rely either on brittle heuristics or on data-hungry supervised models. We introduce the first fully unsupervised SQA pipeline for wrist PPG. Stage 1 trains a contrastive 1-D ResNet-18 on 276 h of raw, unlabeled data from heterogeneous sources (varying in device and sampling frequency), yielding optical-emitter- and motion-invariant embeddings (i.e., the learned representation is stable across differences in LED wavelength, drive intensity, and device optics, as well as wrist motion). Stage 2 converts each 512-D encoder embedding into a 4-D topological signature via persistent homology (PH) and clusters these signatures with HDBSCAN. To produce a binary signal-quality index (SQI), the acceptable PPG signals are represented by the densest cluster while the remaining clusters are assumed to mainly contain poor-quality PPG signals. Without re-tuning, the SQI attains Silhouette, Davies-Bouldin, and Calinski-Harabasz scores of 0.72, 0.34, and 6173, respectively, on a stratified sample of 10,000 windows. In this study, we propose a hybrid self-supervised-learning--topological-data-analysis (SSL--TDA) framework that offers a drop-in, scalable, cross-device quality gate for PPG signals.

Problem

• Assessing signal quality in wearable PPG devices without labeled data
• Overcoming motion artifacts and device variability in photoplethysmography
• Developing unsupervised quality assessment for cross-device PPG analytics

Innovation

• Self-supervised contrastive ResNet for embedding
• Topological signatures via persistent homology
• Unsupervised clustering with HDBSCAN for SQI