Monitoring Cardiac Activity by Detecting Subtle Head Movements Using MEMS Technology

Ballistocardiography (BCG) is a non-invasive technique that measures the recoil forces of the body in reaction to the cardiac contraction and blood flow through the vessels. This work compares the performance of a virtual reality (VR) headset-embedded gyroscope and of a novel high-performance gyroscope in measuring median HR from the BCG signal obtained from subtle head movements. Nine healthy volunteers were enrolled in this study. Head BCG signals were acquired for 1 minute in supine position using the triaxial gyroscope (VRG) embedded in an Oculus Quest (Facebook) headset, and a monoaxial high-performance gyroscope (HPG). 1-lead ECG signal was acquired simultaneously and used as a gold standard for HR measurement (HRECG). Automatic beat-by-beat identification was performed on the BCG signals, from which median HR was computed (HRVRG and HRHPG). Results obtained with the three sensors were statistically compared, and linear regression and Bland Altman analyses were performed. Pitch and roll head rotations provided more accurate HR estimates compared to the yaw rotation, with more marked peaks in the BCG signal, possibly due to the to the anatomical orientation of the carotid arteries and to how the head is perfused with blood. Also, the HPG outperformed the VRG, thus potentially allowing a more detailed analysis of the BCG signal morphology, with possible application in the extraction of novel biomarkers with clinical utility.

Automated summary

This study compares the performance of a virtual reality headset-embedded gyroscope and a novel high-performance gyroscope in measuring median heart rate from the ballistocardiography (BCG) signal. The results suggest that pitch and roll head rotations provide more accurate heart rate estimates, and the high-performance gyroscope outperforms the virtual reality gyroscope, potentially enabling the extraction of novel biomarkers with clinical utility from the BCG signal.