TriDP-PTM: a three-stage distortion-perception tradeoff guides the pre-training model for radar cardiac sensing

2026-05-25

AI Summary

This study addresses the trade-off between distortion-driven fidelity and perception-driven interpretability in non-contact radar-based cardiac monitoring, which often suffers from semantic ambiguity and poor explainability. The authors propose TriDP-PTM, a three-stage distortion-perception pretraining model featuring a multi-scale dual-path architecture that systematically compares direct (radar-to-task) and indirect (radar-to-ECG-to-task) processing pathways. By integrating a composite loss composed of an ECG generator and a feature discriminator, the framework effectively incorporates electrocardiographic medical priors into downstream tasks. The work reveals, for the first time, that the distortion-perception trade-off exhibits three distinct phases—positive-sum, competitive-cooperative, and negative-sum—with the competitive-cooperative phase yielding optimal clinical accuracy and ECG interpretability. Evaluated on data from 30 subjects, the indirect pathway achieves a mean IoU of 0.80 in waveform segmentation, an average accuracy of 98.3% across four classification tasks, and a 56% reduction in mean absolute error for blood pressure regression compared to the strongest baseline.

Abstract

Cardiovascular diseases (CVDs) remain a leading cause of death globally, necessitating continuous, accurate non-invasive cardiac monitoring. While non-contact radar-based approaches show great promise, they often employ a single "distortion-driven" or "perception-driven" paradigm, frequently facing a trade-off between "low distortion but weak semantic information" and "high perceptual fidelity but poor interpretability." To address this, we propose a Three-stage Distortion-Perception Pre-Training Model (TriDP-PTM), a radar-based multi-scale fusion dual-path framework that systematically compares the "direct radar-to-task" path against an "indirect radar-to-ECG-to-task" path. By integrating an ECG generator with a feature discriminator to form a composite loss function, our approach effectively incorporates medical priors - such as ECG morphology and rhythm - into downstream tasks. Through empirical analysis, we reveal that this trade-off manifests in three distinct phases (Positive-Sum, Coopetitive, and Negative-Sum), showing optimal downstream clinical accuracy typically emerges in the coopetitive stage. Extensive experiments on a dataset involving 30 subjects across 5 physiological states reveal that the indirect path consistently outperforms the direct path in diverse tasks, achieving 0.80 mean IoU in waveform segmentation, 98.3% average classification accuracy across four tasks, and a 56% MAE reduction in blood pressure regression compared to the strongest baselines. These findings validate our framework and indicate that, within the indirect radar-to-ECG pathway, appropriately weighting distortion and perception losses to operate in the coopetitive regime is critical for achieving both clinically interpretable ECG morphology and strong downstream accuracy in non-contact cardiac monitoring.

Problem

• distortion-perception tradeoff
• radar cardiac sensing
• non-contact monitoring
• ECG morphology
• clinical interpretability

Innovation

• distortion-perception tradeoff
• radar cardiac sensing
• ECG prior
• multi-scale fusion
• pre-training model