Wearable respiratory monitoring during dynamic activities faces challenges from motion artifacts, but hybrid systems and multimodal sensor fusion algorithms offer promising solutions for reliable assessment.
This review highlights the potential of hybrid wearable systems and advanced algorithms to overcome motion artifacts in continuous respiratory monitoring.
Advances in wearable device and sensor technologies progressively shift respiratory monitoring from the clinical setting to real-world conditions. This rapidly developing field allows for more accurate diagnostics. However, reliable monitoring during dynamic activities remains challenging due to artifacts caused by movement, postural changes, electrode drift, and variability in breathing patterns. Therefore, this review focuses on wearable methodologies capable of determining respiratory rate and potentially tidal volume during strenuous physical activities. Direct sensing approaches, including chest and abdominal belts, bioimpedance principles, and inertial sensing units, are complemented by indirect methods derived from ECG and PPG signals. Hybrid systems, which are also discussed, represent a very promising approach. Special attention is paid to signal processing, machine learning, and multimodal sensor fusion algorithms that improve robustness and reliability. By systematically analyzing hardware and software combinations, validation protocols, and current limitations, this article identifies emerging trends in adaptive respiratory monitoring. This review aims to guide the development of next-generation wearable systems.
Pečík et al. (Sat,) conducted a review in Respiratory monitoring. Wearable methodologies was evaluated. Wearable respiratory monitoring during dynamic activities faces challenges from motion artifacts, but hybrid systems and multimodal sensor fusion algorithms offer promising solutions for reliable assessment.