A multi-modal smart chest patch for real-time cardiopulmonary monitoring and anomaly detection

• • The SCP achieves 87% classification accuracy for detecting cardiac and respiratory anomalies on a dataset of 5,561 recordings from 475 participants, surpassing conventional single-modality approaches. This high accuracy is critical for clinical translation, reducing false positives that plague existing wearables and enabling reliable early intervention. • • The patch weighs 5.4 g and is 3.6 mm thick, ensuring unobtrusive wearability. Its flexible sensing layer maintains stable signal quality across diverse individuals, addressing the signal-to-noise ratio (SNR) degradation common in peripheral devices like wrist oximeters, which are positioned far from the signal source. • • Real-time exercise monitoring captured dynamic physiological shifts with ΔHR = 21 bpm and ΔPEP = −30 ms, demonstrating superior signal fidelity. These metrics provide actionable insights for optimizing exercise regimens and detecting early signs of cardiopulmonary dysfunction, a capability absent in commercial fitness trackers. • • The MCMF model fuses ECG, HS, and respiratory signals, enabling extraction of 12 cardiopulmonary parameters including respiratory sinus arrhythmia (RSA), pre-ejection period (PEP), and heart rate variability (HRV). This multimodal integration overcomes the limitations of single-modality devices, which cannot capture composite metrics reflective of heart-lung interactions.