A heart beat rate detection framework using multiple nanofiber sensor signals

Although electrocardiogram (ECG) is one standard way for monitoring heart beat rate, there are of great interests in exploring other types of biophysical signals. A novel type of nanofiber (NF) sensor signals, as a potential alternative choice to ECG signals for heart beat monitoring, are investigated in this paper. To get the heart beat signal, three nano sensors are deployed at the wrist. However, detecting the heart beat rate (HBR) directly from the raw data is challenging because the signals of interest are masked by different types of noise. To address this concern, a two-step framework based on ensemble empirical mode decomposition (EEMD) and multiset canonical correlation analysis (MCCA) is proposed to extract the interesting signals. Further, a specific HBR detection method is presented based on peak detection and peak filtering. We apply the proposed framework to the real data collected from one subject performing 8 tasks, and the results demonstrate its effectiveness and potential in real applications.