Abstract:
Bioimpedance (BioZ) is a versatile, safe, and non-invasive technique with significant potential across various biomedical applications, including cardiovascular health monitoring and muscle function assessment. This work investigates BioZ measurement approaches by exploring the effects of key factors such as signal frequency, electrode type, contact pressure, and measurement location. Through systematic variation of these parameters, we assess their impact on the reliability and stability of BioZ readings at both the wrist and forearm. The findings demonstrate that high-frequency BioZ measurements at the forearm show significant promise, and the choice of electrode material plays a crucial role in signal stability. Additionally, machine learning techniques were employed to predict hand grip strength (HGS), an established indicator of cardiovascular health, using BioZ data. The inclusion of other physiological health parameters in the measurement setup further enhanced the predictive model. Results indicate that combining wrist and forearm measurements improves prediction accuracy, with forearm-specific measurements showing a stronger correlation with HGS. This research not only establishes optimized protocols for BioZ measurement but also underscores its potential in wearable health monitoring systems. The findings support the broader application of BioZ in wearable biosensors, contributing to advancements in personalized healthcare and continuous physiological monitoring.
