LifWavNet: Lifting Wavelet-based Network for Non-contact ECG Reconstruction from Radar

2025-10-31

AI Summary

This work addresses the problem of non-contact electrocardiogram (ECG) reconstruction from radar Doppler signals. Methodologically, we propose a multi-resolution deep network built upon learnable lifting wavelets: a learnable wavelet unit and its invertible counterpart jointly enable end-to-end mapping from raw radar echoes to physiological ECG waveforms. To enforce reconstruction fidelity and interpretability, we introduce a multi-resolution short-time Fourier transform (STFT) loss, imposing consistency constraints in both time and frequency domains. Compared with state-of-the-art methods, our framework achieves significantly improved waveform fidelity on two public radar-based ECG datasets. Quantitatively, it reduces heart rate estimation error by 32.7% and heart rate variability (HRV) estimation error by 28.4%. The approach establishes a new paradigm for unobtrusive, continuous cardiac monitoring—offering both high accuracy and inherent interpretability through wavelet-based representation learning.

Technology Category

• Machine Learning: Multimodal Learning
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Multi-modal Vision

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Non-contact electrocardiogram (ECG) reconstruction from radar signals offers a promising approach for unobtrusive cardiac monitoring. We present LifWavNet, a lifting wavelet network based on a multi-resolution analysis and synthesis (MRAS) model for radar-to-ECG reconstruction. Unlike prior models that use fixed wavelet approaches, LifWavNet employs learnable lifting wavelets with lifting and inverse lifting units to adaptively capture radar signal features and synthesize physiologically meaningful ECG waveforms. To improve reconstruction fidelity, we introduce a multi-resolution short-time Fourier transform (STFT) loss, that enforces consistency with the ground-truth ECG in both temporal and spectral domains. Evaluations on two public datasets demonstrate that LifWavNet outperforms state-of-the-art methods in ECG reconstruction and downstream vital sign estimation (heart rate and heart rate variability). Furthermore, intermediate feature visualization highlights the interpretability of multi-resolution decomposition and synthesis in radar-to-ECG reconstruction. These results establish LifWavNet as a robust framework for radar-based non-contact ECG measurement.

Problem

• Reconstructing ECG from radar signals non-contactly
• Employing learnable wavelets for adaptive feature capture
• Enhancing reconstruction fidelity with multi-resolution loss functions

Innovation

• Learnable lifting wavelets adaptively capture radar features
• Multi-resolution STFT loss enforces temporal-spectral consistency
• Lifting network synthesizes physiologically meaningful ECG waveforms