CardioFusion-AI: Robust ECG--PPG Fusion for Multimodal Physiological Monitoring Under Signal Degradation

📄 arXiv: 2608.26000v1 📥 PDF

作者: Navaneetha Krishnan Kamalakannan, Janakiraman Kamalakannan

分类: eess.SP, cs.LG, q-bio.QM

发布日期: 2026-08-26

备注: 7 pages, 4 figures. Under review at IEEE Journal of Biomedical and Health Informatics. Code: https://github.com/ka-cyber/CardioFusion-AI

💡 一句话要点

提出CardioFusion-AI以解决ECG和PPG信号融合中的可靠性问题

🎯 匹配领域: 支柱八：物理动画 (Physics-based Animation) 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 心电图 光电容积脉搏波图 信号融合 生理监测 自适应融合 信号质量评估 多模态监测

📋 核心要点

  1. 现有的ECG和PPG融合方法在信号降解时表现不佳，可能导致融合结果不如单一模态可靠。
  2. CardioFusion-AI框架通过信号处理前端和自适应融合策略，增强了对信号降解的鲁棒性。
  3. 实验结果表明，注意力融合策略在多种降解条件下表现优异，整体误差最低，显示出模态质量的重要性。

📝 摘要（中文）

可穿戴心电图（ECG）和光电容积脉搏波图（PPG）传感器互为补充，但各自容易受到运动伪影、接触不良和传感器失效的影响。假设两种模态同等可信的融合策略在信号降解时可能不如单一清晰模态可靠。本文提出CardioFusion-AI框架，经过53个真实重症监护记录的验证，心率平均绝对误差为ECG 1.61 bpm和PPG 2.78 bpm。通过对八种ECG-PPG融合策略在六种降解条件下进行对比，注意力融合策略实现了最低的整体误差（1.66±0.43 bpm），并指出模态可用性和质量是自适应融合中功能上不同的问题。

🔬 方法详解

问题定义：本文旨在解决ECG和PPG信号融合中的可靠性问题，尤其是在信号降解情况下，现有方法往往假设两种模态同等可信，导致融合效果不佳。

核心思路：CardioFusion-AI框架通过引入信号质量评估和自适应权重分配，增强了对降解信号的处理能力，确保在模态失效时依然能够提供可靠的监测结果。

技术框架：该框架包括信号处理前端（R波和收缩波峰检测、信号质量指数评估、逐拍脉搏传导时间估计）和自适应融合策略，整体流程从信号采集到融合输出。

关键创新：最重要的创新在于引入了信号质量条件的自适应融合策略，使得在模态完全失效时，系统能够自动调整权重，优先使用健康模态的数据。

关键设计：在实验中使用了53个真实重症监护记录，采用了Holm校正的显著性测试，设置了多个训练种子以确保结果的可靠性，关注信号质量与自适应权重之间的关系。实验结果显示，信号质量条件下的自适应融合在缺失PPG时表现出特定的改进。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示，注意力融合策略在多种降解条件下实现了最低的整体误差（1.66±0.43 bpm），并且在缺失PPG的情况下，信号质量条件下的自适应融合表现出特定的改进（1.56±0.59 bpm），接近单模态的最佳表现（1.48 bpm）。

🎯 应用场景

CardioFusion-AI框架在生理监测领域具有广泛的应用潜力，尤其是在重症监护、远程健康监测和移动健康设备中。通过提高ECG和PPG信号融合的可靠性，该技术可以为患者提供更准确的健康状态评估，进而改善临床决策和患者护理质量。

📄 摘要（原文）

Wearable electrocardiogram (ECG) and photoplethysmogram (PPG) sensors are complementary but individually fragile: motion artifact, poor contact, and sensor dropout can degrade one or both signals. Fusion strategies that assume both modalities are equally trustworthy can become less reliable than a single clean modality under degradation. We present CardioFusion-AI, a framework whose signal-processing front end, including R-peak and systolic-peak detection, an Orphanidou-type signal-quality index, and beat-by-beat pulse transit time estimation, is validated on 53 real intensive-care recordings (848 windows; heart-rate mean absolute error 1.61 bpm for ECG and 2.78 bpm for PPG) and a real annotated fetal ECG database (R-peak F1 0.89-0.98). We then conduct a controlled synthetic degradation study comparing eight ECG-PPG fusion strategies across six degradation regimes spanning graded corruption and complete modality loss, using five independent training seeds. Attention fusion achieved the lowest descriptive overall error (1.66+/-0.43 bpm). Both adaptive gates reallocated weight toward the healthy modality under complete modality loss, but showed near-zero correlation between gate weight and signal quality under graded degradation (r = 0.10-0.24). Signal-quality conditioning produced a specific improvement under missing-PPG conditions (1.56+/-0.59 bpm), approaching the 1.48 bpm unimodal ceiling. With only five training seeds, no pairwise comparison survives Holm-corrected significance testing; effect sizes and confidence intervals are therefore reported. These results indicate that modality availability and modality quality are functionally distinct problems for adaptive fusion.