ABSTRACT The growing prominence of personal health issues has made appropriate physical exercise increasingly essential for maintaining health. To ensure adequate workout while preventing overexertion, reliable exercise intensity assessment methods are crucial. However, existing approaches suffer from operational complexity and failure to account for individual physiological differences. In this paper, a multimodal exercise intensity monitoring wristband integrating piezoelectric, thermoelectric, and electromyography technologies is developed. A PVDF‐based piezoelectric sensor is employed to monitor pulse. Bi 0.5 Sb 1.5 Te 3 and Bi 2 Se 0.3 Te 2.7 thermoelectric legs are selected to fabricate a thermoelectric sensor for body temperature monitoring, and sEMG is used for muscle fatigue assessment. These physiological signals (pulse, body temperature, and muscle fatigue) enable effective exercise intensity evaluation. During continuous exercise, it is achieved continuous monitoring of heart rate, body temperature, and sEMG of the subject. Specifically, the subject's heart rate increased from 77.60 to 91.04 bpm, body temperature rose from 35.3 to 36.0°C, and the continuous changes in sEMG are also recorded. Then, through machine learning algorithms, quantitative relationships are established between signal feature value and exercise intensity levels, generating personalized exercise recommendations. This research demonstrates significant potential for applications in intelligent and personalized physiological health monitoring systems.
Shen et al. (Fri,) studied this question.