Zero-Effort Ambient Heart Rate Monitoring Using Ballistocardiography Detected Through a Seat Cushion: Prototype Development and Preliminary Study

Abstract

Background Cardiovascular diseases are a leading cause of death worldwide and result in significant economic costs to health care systems. The prevalence of cardiovascular conditions that require monitoring is expected to increase as the average age of the global population continues to rise. Although an accurate cardiac assessment can be performed at medical centers, frequent visits for assessment are not feasible for most people, especially those with limited mobility. Monitoring of vital signs at home is becoming an increasingly desirable, accessible, and practical alternative. As wearable devices are not the ideal solution for everyone, it is necessary to develop parallel and complementary approaches. Objective This research aims to develop a zero-effort, unobtrusive, cost-effective, and portable option for home-based ambient heart rate monitoring. Methods The prototype seat cushion uses load cells to acquire a user’s ballistocardiogram (BCG). The analog signal from the load cells is amplified and filtered by a signal-conditioning circuit before being digitally recorded. A pilot study with 20 participants was conducted to analyze the prototype’s ability to capture the BCG during five real-world tasks: sitting still, watching a video on a computer screen, reading, using a computer, and having a conversation. A novel algorithm based on the continuous wavelet transform was developed to extract the heart rate by detecting the largest amplitude values (J-peaks) in the BCG signal. Results The pilot study data showed that the BCG signals from all five tasks had sufficiently large portions to extract heart rate. The continuous wavelet transform–based algorithm for J-peak detection demonstrated an overall accuracy of 91.4% compared with electrocardiography. Excluding three outliers that had significantly noisy BCG data, the algorithm achieved 94.6% accuracy, which was aligned with that of wearable devices. Conclusions This study suggests that BCG acquired through a seat cushion is a viable alternative to wearable technologies. The prototype seat cushion presented in this study is an example of a relatively accessible, affordable, portable, and unobtrusive zero-effort approach to achieve frequent home-based ambient heart rate monitoring.

Monitoring of Cardiorespiratory Parameters during Sleep Using a Special Holder for the Accelerometer Sensor

“…These methods include the use of different types of sensors, as well as differences in the number of sensors and their placement for measurement. Some of them used a hydraulic sensor system filled with water [17], load cells [18], piezoelectric load [19], pressure sensors installed under the mattress [20,21] and others. With the development of highly sensitive accelerometers based on micro-electromechanical systems (MEMS), their application in various measurement fields is growing rapidly [22][23][24][25].…”

Heart rate detection of ballistocardiogram based on improved DAE and template matching method

“…In order to extract heart rate and respiration rate information from BCG signals, heart rate extraction methods based on weak signal processing algorithm, such as discrete wavelet transformation (CWT) and empirical mode decomposition (EMD) have been proposed one after another. Reference [7] used clustering algorithm to achieve adaptive matching of BCG waveform and calculated heart rate by Hilbert transform; Reference [8] and Reference [9] used discrete wavelet transformation and continuous wavelet transformation methods to extract heart rate information from BCG signals, respectively; Reference [10] combined the empirical mode decomposition (EMD) and independent component analysis (ICA) to achieve noise reduction of BCG signal; Reference [11] proposes a BCG denoising method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) combined with permutation entropy (PE). Wu et al fused Bluetooth signals and BCG signals to monitor sedentary behavior by extracting local spectral features and signal differences [12]; Ibrahim Sadek et al investigated the effectiveness of three heart rate detection algorithms, namely maximal overlap discrete wavelet transform (MODWT-MRA), continuous wavelet transform (CWT) and template matching (TM), on four independent BCG datasets [13].…”

Digital Optical Ballistocardiographic System for Activity, Heart Rate, and Breath Rate Determination during Sleep

“…Furthermore, since BCG signals can also accurately detect human respiratory status and predict respiratory diseases [18], the use of these systems is widely applied in several fields outside the medical scope. For instance, the use of BCG for the development of a health monitoring chair was reported [8,19]. In the automotive industry, this technique is being studied as a tool to measure the vital functions of occupants [20].…”