PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 24, 2026Sensors0 citationsOpen Access

Contactless Cardiac Health Monitoring with Millimeter-Wave Radar Based on PMG-SATNet

View Full Paper
TGTianjiao GuoJWJianqi WangNYNianzeng Yuan

Key Result

PMG-SATNet recovered ECG signals from millimeter-wave radar chest vibrations, improving Pearson correlation by up to 3.8% and root mean square error by up to 23.8% over baseline models.

Key Points

  • This research aims to enhance cardiac health monitoring by utilizing non-contact millimeter-wave radar and a novel deep learning approach.
  • Developed PMG-SATNet, a deep learning network with encoder-decoder architecture.
  • Designed seven experimental scenarios for monitoring across wakefulness and sleep.
  • Validated performance on a custom-built dataset comparing against baseline models.
  • PMG-SATNet improved Pearson correlation coefficient by 3.3% and root mean square error by 3.8%.
  • Achieved a 16.4% and 23.8% improvement in signal recovery metrics over baseline models.
  • Demonstrated high fidelity in recovering ECG signals from chest vibrations captured by radar.

Structured PICO

Does PMG-SATNet improve the recovery of ECG signals from millimeter-wave radar compared to baseline models?

P
Population
Self-built dataset covering seven experimental scenarios (wakefulness and sleep)
I
Intervention
PMG-SATNet (a deep learning network consisting of encoder and decoder structures) combined with millimeter-wave radar
C
Comparator
Baseline models
O
Outcome
Pearson correlation coefficient and root mean square error for ECG signal recoverysurrogate

A novel deep learning network, PMG-SATNet, significantly improves the fidelity of contactless ECG signal recovery from millimeter-wave radar compared to baseline models.

Abstract

Cardiovascular diseases are the primary causes of mortality worldwide, often characterized by subtle onset and acute progression. Traditional ECG electrodes may cause skin irritation, limiting routine monitoring and early risk assessment. Relying on the advantages of non-contact monitoring, millimeter-wave radar-based cardiac monitoring combined with deep learning has become a popular research direction recently. To overcome the poor generalization of methods trained from single-source datasets, this study designed seven experimental scenarios covering wakefulness and sleep. A novel deep learning network consisting of encoder and decoder structures named PMG-SATNet was proposed. The encoder comprises a parallel multi-scale feature extraction module and a global temporal relationship modeling module to capture fine-grained local patterns and long-range dependencies. The decoder employs a temporal convolutional network augmented with a spectral attention mechanism to emphasize clinically relevant ECG frequency bands and suppress respiration and body motion interference. After being validated on the self-built dataset, PMG-SATNet outperformed baseline models in terms of Pearson correlation coefficient and root mean square error, with an improvement of 3.3% and 3.8%, and 16.4% and 23.8%, respectively. The validation results imply that PMG-SATNet is capable of recovering ECG signals from millimeter-wave radar-derived chest vibrations with high fidelity and can potentially be implemented in real-life cardiac health monitoring.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Guo et al. (2026) studied this question. PMG-SATNet recovered ECG signals from millimeter-wave radar chest vibrations, improving Pearson correlation by up to 3.8% and root mean square error by up to 23.8% over baseline models.

synapsesocial.com/papers/69eb0cb2553a5433e34b5a74https://doi.org/10.3390/s26092579
Ask AI
Helpful
Bookmark
Share
View Full Paper