Lost in Time? A Meta-Learning Framework for Time-Shift-Tolerant Physiological Signal Transformation

TL;DR

This paper introduces ShiftSyncNet, a meta-learning framework that automatically corrects time misalignments in physiological signal transformation, significantly improving accuracy in translating signals like PPG and BCG into clinically relevant measures.

Contribution

The paper presents a novel meta-learning bi-level optimization approach with a time-shift correction network, effectively addressing temporal misalignments in multimodal physiological signal transformation.

Findings

Outperforms baselines by up to 12.8% in accuracy.

Effectively corrects diverse time shifts in real-world data.

Enhances label quality and transformation precision.

Abstract

Translating non-invasive signals such as photoplethysmography (PPG) and ballistocardiography (BCG) into clinically meaningful signals like arterial blood pressure (ABP) is vital for continuous, low-cost healthcare monitoring. However, temporal misalignment in multimodal signal transformation impairs transformation accuracy, especially in capturing critical features like ABP peaks. Conventional synchronization methods often rely on strong similarity assumptions or manual tuning, while existing Learning with Noisy Labels (LNL) approaches are ineffective under time-shifted supervision, either discarding excessive data or failing to correct label shifts. To address this challenge, we propose ShiftSyncNet, a meta-learning-based bi-level optimization framework that automatically mitigates performance degradation due to time misalignment. It comprises a transformation network (TransNet) and a time-shift correction network (SyncNet), where SyncNet learns time offsets between training pairs and applies Fourier phase shifts to align supervision signals. Experiments on one real-world industrial dataset and two public datasets show that ShiftSyncNet outperforms strong baselines by 9.4%, 6.0%, and 12.8%, respectively. The results highlight its effectiveness in correcting time shifts, improving label quality, and enhancing transformation accuracy across diverse misalignment scenarios, pointing toward a unified direction for addressing temporal inconsistencies in multimodal physiological transformation.