Monitoring Pulse Rate in the Background Using Front Facing Cameras of Mobile Devices.

We propose a novel framework to passively monitor pulse rate during the time spent by users on their personal mobile devices. Our framework is based on passively capturing the user's pulse signal using the front-facing camera. Signal capture is performed in the background, while the user is interacting with the device as he/she normally would, e.g., watch movies, read emails, text, and play games. The framework does not require subject participation with the monitoring procedure, thereby addressing the well-known problem of low adherence with such procedures. We investigate various techniques to suppress the impact of spontaneous user motion and fluctuations in ambient light conditions expected in non-participatory environments. Techniques include traditional signal processing, machine learning classifiers, and deep learning methods. Our performance evaluation is based on a clinical study encompassing 113 patients with a history of atrial fibrillation (Afib) who are passively monitored at home using a tablet for a period of two weeks. Our results show that the proposed framework accurately monitors pulse rate, thereby providing a gateway for long-term monitoring without relying on subject participation or the use of a dedicated wearable device.

Assessment of facial video-based detection of atrial fibrillation across human complexion.

Background: Early self-detection of atrial fibrillation (AF) can help delay and/or prevent significant associated complications, including embolic stroke and heart failure. We developed a facial video technology, videoplethysmography (VPG), to detect AF based on the analysis of facial pulsatile signals. Objective: The purpose of this study was to evaluate the accuracy of a video-based technology to detect AF on a smartphone and to test the performance of the technology in AF patients across the whole spectrum of skin complexion and under various recording conditions. Methods: The performance of video-based monitoring depends on a set of factors such as the angle and the distance between the camera and the patient's face, the strength of illumination, and the patient's skin tone. We conducted a clinical study involving 60 subjects with a confirmed diagnosis of AF. A continuous electrocardiogram was used as the gold standard for cardiac rhythm annotation. The VPG technology was fine-tuned on a smartphone for the first 15 subjects. Validation recordings were then done using 7053 measurements collected from the remaining 45 subjects. Results: The VPG technology detected the presence of AF using the video camera from a common smartphone with sensitivity and specificity ≥90%. The ambient level of illumination needs to be ≥100 lux for the technology to deliver consistent performance across all skin tones. Conclusion: We demonstrated that facial video-based detection of AF provides accurate outpatient cardiac monitoring including high pulse rate accuracy and medical-grade performance for AF detection.

Dynamic Off-Body Rician Channel Modeling for Indoor Wireless Body Area Networks.

Wireless off-body channel is commonly treated as a traditional communication system assuming large and small scale fading to be independent. In this work, we challenge this approach and experimentally show that in a typical WBAN environment of a single room, this assumption is not accurate. We propose an off-body channel model where path loss and multipath fading are entirely determined by Rician distribution and linked through a stochastic Rician factor K. Treating K as a random variable, whose density is determined by the scenario under consideration, allows our model to naturally address different types of body motion along with correlated shadowing. By means of simulation supported by experiments, we show that our model is able to accurately describe probability densities of instantaneous received power in various static and dynamic settings. In addition, we demonstrate that the proposed approach can predict the off-body diversity gains due to correlated antenna reception when selection combining scheme is applied.

On the Benefits of Creeping Wave Antennas in Reducing Interference Between Neighboring Wireless Body Area Networks.

We consider the on-body, off-body, and body-to-body channels in wireless body area networks utilizing creeping wave antennas. Experimental setups are used to gather measurements in the 2.4 GHz band with body area networks operating in an office environment. Data packets providing received signal strength indicators are used to assess the performance of the creeping wave antenna in reducing interference at a neighboring on-body access point while supporting reliable on-body communications. Results demonstrate that creeping wave antennas provide reliable on-body communications while significantly reducing inter-network interference; the inter-network interference is shown to be 10 dB weaker than the on-body signal. In addition, the inter-network interference when both networks utilize creeping wave antennas is shown to be 3 dB weaker than the interference when monopole antennas are used.

Securing While Sampling in Wireless Body Area Networks With Application to Electrocardiography.

Stringent resource constraints and broadcast transmission in wireless body area network raise serious security concerns when employed in biomedical applications. Protecting data transmission where any minor alteration is potentially harmful is of significant importance in healthcare. Traditional security methods based on public or private key infrastructure require considerable memory and computational resources, and present an implementation obstacle in compact sensor nodes. This paper proposes a lightweight encryption framework augmenting compressed sensing with wireless physical layer security. Augmenting compressed sensing to secure information is based on the use of the measurement matrix as an encryption key, and allows for incorporating security in addition to compression at the time of sampling an analog signal. The proposed approach eliminates the need for a separate encryption algorithm, as well as the predeployment of a key thereby conserving sensor node's limited resources. The proposed framework is evaluated using analysis, simulation, and experimentation applied to a wireless electrocardiogram setup consisting of a sensor node, an access point, and an eavesdropper performing a proximity attack. Results show that legitimate communication is reliable and secure given that the eavesdropper is located at a reasonable distance from the sensor node and the access point.

On the benefits of alternative color spaces for noncontact heart rate measurements using standard red-green-blue cameras.

Existing video plethysmography methods use standard red-green-blue (sRGB) video recordings of the facial region to estimate heart pulse rate without making contact with the person being monitored. Methods achieving high estimation accuracy require considerable signal-processing power and result in significant processing latency. High processing power and latency are limiting factors when real-time pulse rate estimation is required or when the sensing platform has no access to high processing power. We investigate the use of alternative color spaces derived from sRGB video recordings as a fast light-weight alternative to pulse rate estimation. We consider seven color spaces and compare their performance with state-of-the-art algorithms that use independent component analysis. The comparison is performed over a dataset of 41 video recordings from subjects of varying skin tone and age. Results indicate that the hue channel provides better estimation accuracy using extremely low computation power and with practically no latency.

Patient-specific 12-lead ECG reconstruction from sparse electrodes using independent component analysis.

We propose and evaluate a method of 12-lead electrocardiogram (ECG) reconstruction from a three-lead set. The method makes use of independent component analysis and results in adaptive patient-specific transforms. The required calibration process is short and makes use of a single beat. We apply the method to two sets of leads: leads I, II, V2 and the Frank XYZ leads. Performance is evaluated via percent correlation calculations between reconstructed and original leads from a publicly available database of 549 ECG recordings. Results depict percent correlation exceeding 96% for almost all leads. Adaptability of the method's transform is shown to compensate for changes in signal propagation conditions due to breathing, resulting in reduced variance of reconstruction accuracy across beats. This implies that the method is robust to changes that occur after the time of calibration. Accurate and adaptive reconstruction has the potential to augment the clinical significance of wireless ECG systems since the number of sensor nodes placed on the body is limited and the subject could be mobile.

On increasing network lifetime in body area networks using global routing with energy consumption balancing.

Global routing protocols in wireless body area networks are considered. Global routing is augmented with a novel link cost function designed to balance energy consumption across the network. The result is a substantial increase in network lifetime at the expense of a marginal increase in energy per bit. Network maintenance requirements are reduced as well, since balancing energy consumption means all batteries need to be serviced at the same time and less frequently. The proposed routing protocol is evaluated using a hardware experimental setup comprising multiple nodes and an access point. The setup is used to assess network architectures, including an on-body access point and an off-body access point with varying number of antennas. Real-time experiments are conducted in indoor environments to assess performance gains. In addition, the setup is used to record channel attenuation data which are then processed in extensive computer simulations providing insight on the effect of protocol parameters on performance. Results demonstrate efficient balancing of energy consumption across all nodes, an average increase of up to 40% in network lifetime corresponding to a modest average increase of 0.4 dB in energy per bit, and a cutoff effect on required transmission power to achieve reliable connectivity.

Constrained independent component analysis approach to nonobtrusive pulse rate measurements.

Nonobtrusive pulse rate measurement using a webcam is considered. We demonstrate how state-of-the-art algorithms based on independent component analysis suffer from a sorting problem which hinders their performance, and propose a novel algorithm based on constrained independent component analysis to improve performance. We present how the proposed algorithm extracts a photoplethysmography signal and resolves the sorting problem. In addition, we perform a comparative study between the proposed algorithm and state-of-the-art algorithms over 45 video streams using a finger probe oxymeter for reference measurements. The proposed algorithm provides improved accuracy: the root mean square error is decreased from 20.6 and 9.5 beats per minute (bpm) for existing algorithms to 3.5 bpm for the proposed algorithm. An error of 3.5 bpm is within the inaccuracy expected from the reference measurements. This implies that the proposed algorithm provided performance of equal accuracy to the finger probe oximeter.

Reconstructing ECG precordial leads from a reduced lead set using independent component analysis.

In this paper, precordial lead reconstruction from a reduced set of leads is considered. We propose the use of independent component analysis to train patient-specific transforms from a reduced lead set to the six precordial leads of the standard 12-lead electrocardiogram. The proposed approach is applied to a publicly available database comprising 549 ECG recordings of patients with varying cardiovascular conditions. The fidelity of reconstruction is measured using percent correlation between the actual and reconstructed signals following a 30 seconds time lapse. The mean correlation is over 95% with a standard deviation under 12.7% for all reconstructed leads. The results demonstrate the potential of the suggested approach to provide a reliable solution to precordial leads reconstruction.