Search papers, labs, and topics across Lattice.
CardioFusion-AI introduces a robust framework for fusing ECG and PPG signals, addressing the challenges posed by signal degradation due to motion artifacts and sensor dropout. By employing advanced signal-processing techniques and validating the system on extensive real-world data, the framework demonstrates superior performance in maintaining accurate physiological monitoring, achieving a heart-rate mean absolute error of 1.61 bpm for ECG and 2.78 bpm for PPG. Notably, the attention fusion strategy outperformed other methods, achieving the lowest overall error and revealing that adaptive gates can effectively prioritize healthier signals during complete modality loss, while indicating a disconnect between signal quality and gate weight under graded degradation.
Attention fusion can significantly enhance ECG-PPG signal accuracy, achieving a mean absolute error of just 1.66 bpm even under severe degradation conditions.
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.
