Robust Photoplethysmography Signal Denoising via Mamba Networks

2025-10-13

AI Summary

PPG signals in wearable health monitoring are highly susceptible to noise and motion artifacts, degrading heart rate (HR) estimation accuracy. To address this, we propose an end-to-end PPG denoising framework built upon the Mamba architecture—the first application of state space models to physiological signal denoising—enabling efficient long-range temporal modeling. We introduce a scale-invariant signal-to-distortion ratio (SI-SDR) loss combined with HR-aware supervision to jointly optimize waveform fidelity and physiological consistency. Furthermore, we adopt a joint optimization strategy between the primary denoising network and an auxiliary HR predictor to enhance estimation robustness. Evaluated on the BIDMC dataset under both synthetic noise and realistic motion artifact conditions, our method outperforms conventional filters and state-of-the-art neural networks, achieving a 32.7% reduction in mean absolute HR estimation error.

Technology Category

• Machine Learning: Hardware-aware ML
• Intelligent Robots: State Estimation
• Search and Optimization: Non-convex Optimization

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Photoplethysmography (PPG) is widely used in wearable health monitoring, but its reliability is often degraded by noise and motion artifacts, limiting downstream applications such as heart rate (HR) estimation. This paper presents a deep learning framework for PPG denoising with an emphasis on preserving physiological information. In this framework, we propose DPNet, a Mamba-based denoising backbone designed for effective temporal modeling. To further enhance denoising performance, the framework also incorporates a scale-invariant signal-to-distortion ratio (SI-SDR) loss to promote waveform fidelity and an auxiliary HR predictor (HRP) that provides physiological consistency through HR-based supervision. Experiments on the BIDMC dataset show that our method achieves strong robustness against both synthetic noise and real-world motion artifacts, outperforming conventional filtering and existing neural models. Our method can effectively restore PPG signals while maintaining HR accuracy, highlighting the complementary roles of SI-SDR loss and HR-guided supervision. These results demonstrate the potential of our approach for practical deployment in wearable healthcare systems.

Problem

• Denoising PPG signals degraded by noise and motion artifacts
• Preserving physiological information during signal denoising process
• Enhancing reliability of heart rate estimation from wearable PPG

Innovation

• Mamba-based DPNet for temporal modeling
• SI-SDR loss ensures waveform fidelity
• HR predictor maintains physiological consistency