Objective: At-home monitoring of respiration is of critical urgency especially in the era of the global pandemic due to COVID-19. Electrocardiogram (ECG) and seismocardiogram (SCG) signals—measured in less cumbersome contact form factors than the conventional sealed mask that measures respiratory air flow—are promising solutions for respiratory monitoring. In particular, respiratory rates (RR) can be estimated from ECG-derived respiratory (EDR) and SCG-derived respiratory (SDR) signals. Yet, non-respiratory artifacts might still be present in these surrogates of respiratory signals, hindering the accuracy of the RRs estimated. Methods: In this paper, we propose a novel U-Net-based cascaded framework to address this problem. The EDR and SDR signals were transformed to the spectro-temporal domain and subsequently denoised by a 2D U-Net to reduce the non-respiratory artifacts. Major Results: We have shown that the U-Net that fused an EDR input and an SDR input achieved a low mean absolute error of 0.82 breaths per minute (bpm) and a coefficient of determination (R ² ) of 0.89 using data collected from our chest-worn wearable patch. We also qualitatively provided insights on the complementariness between EDR and SDR signals and demonstrated the generalizability of the proposed framework. Conclusion: ECG and SCG collected from a chest-worn wearable patch can complement each other and yield reliable RR estimation using the proposed cascaded framework. Significance : We anticipate that convenient and comfortable ECG and SCG measurement systems can be augmented with this framework to facilitate pervasive and accurate RR measurement.

In light of the recent Coronavirus disease (COVID-19) pandemic, peripheral oxygen saturation (SpO₂) has shown to be amongst the vital signs most indicative of deterioration in persons with COVID-19. To allow for the continuous monitoring of SpO₂, we attempted to demonstrate accurate SpO₂ estimation using our custom chest-based wearable patch biosensor, capable of measuring electrocardiogram (ECG) and photoplethysmogram (PPG) signals with high fidelity. Through a breath-hold protocol, we collected physiological data with a wide dynamic range of SpO₂ from 20 subjects. The ratio of ratios (R) used in pulse oximetry to estimate SpO₂ was robustly extracted from the red and infrared PPG signals during the breath-hold segments using novel feature extraction and PPG_green-based outlier rejection algorithms. Through subject independent training, we achieved a low root-mean-square error (RMSE) of 2.64 ± 1.14% and a Pearson correlation coefficient (PCC) of 0.89. With subject-specific calibration, we further reduced the RMSE to 2.27 ± 0.76% and increased the PCC to 0.91. In addition, we showed that calibration is more efficiently accomplished by standardizing and focusing on the duration of breath-hold rather than the resulting range in SpO₂. The accurate SpO₂ estimation provided by our custom biosensor and the algorithms provide research opportunities for a wide range of disease and wellness monitoring applications.

Hypovolemic shock is one of the leading causes of death in the military. The current methods of assessing hypovolemia in field settings rely on a clinician assessment of vital signs, which is an unreliable assessment of hypovolemia severity. These methods often detect hypovolemia when interventional methods are ineffective. Therefore, there is a need to develop real-time sensing methods for the early detection of hypovolemia. Previously, our group developed a random-forest model that successfully estimated absolute blood-volume status (ABVS) from noninvasive wearable sensor data for a porcine model (n = 6). However, this model required normalizing ABVS data using individual baseline data, which may not be present in crisis situations where a wearable sensor might be placed on a patient by the attending clinician. We address this barrier by examining seven individual baseline-free normalization techniques. Using a feature-specific global mean from the ABVS and an external dataset for normalization demonstrated similar performance metrics compared to no normalization (normalization: R² = 0.82 ± 0.025|0.80 ± 0.032, AUC = 0.86 ± 5.5 × 10⁻³|0.86 ± 0.013, RMSE = 28.30 ± 0.63%|27.68 ± 0.80%; no normalization: R² = 0.81 ± 0.045, AUC = 0.86 ± 8.9 × 10⁻³, RMSE = 28.89 ± 0.84%). This demonstrates that normalization may not be required and develops a foundation for individual baseline-free ABVS prediction.

In neurodegenerative conditions like Parkinson’s disease (PD) and multiple system atrophy (MSA), cardiovascular autonomic dysfunction (CVAD) is associated with several poor longterm health outcomes. CVAD commonly manifests as orthostatic hypotension (OH), a sustained drop in blood pressure upon standing that can cause syncope and falls. Conventional screening methods for OH are suboptimal and formal autonomic testing is limited to specialized centers. This study explores a multimodal wearable sensing patch for remote monitoring of CVAD. We collected wave-form data during clinical autonomic testing and a 24-hour period at home from 20 participants with synucleinopathies (12 with OH) and 6 healthy controls. We developed an automated posture detection pipeline that identified 103 at-home orthostatic events. Then, physiomarkers related to heart rate variability, cardiac mechanics, and vasomotor function were derived during the supine and standing periods associated with clinical and at-home orthostatic transitions. Comparisons of baroreflex-related supine physiomarkers revealed significant differences between those with and without OH. We characterized cardiovascular autonomic dynamics while standing, leveraging low-dimensional representations, and found marked differences in the aggregate responses between groups. We also observed significantly higher within-subject similarity between the at-home responses of the OH group. Finally, we examined the discriminative power of the patch’s physiomarkers and demonstrated accurate classification of persons with OH during the clinical stand testing (F1=0.83). This study is the first to couple orthostatic event detection with machine learning analysis of wearable-derived physiomarkers, illustrating that wearable sensing can accurately classify OH and provide novel insights into CVAD outside the clinic.

Seismocardiogram (SCG) signals are noninvasively obtained cardiomechanical signals containing important features for cardiovascular health monitoring. However, these signals are prone to contamination by motion noise, which can significantly impact accuracy and robustness of the measurements. A deep learning model based on the U-Net architecture is proposed to recover SCG signals contaminated by motion noise induced by walking. The model performance was evaluated through qualitative visualization, as well as quantitative analyses. Quantitative analyses included distance-based comparisons before and after applying our model. Analyses also included assessments of the model's efficacy in improving the performance of downstream tasks related to health parameter estimation during walking. Experimental findings revealed that the denoising model improved similarity to clean signals by approximately 90%. The performance of the model in enhancing heart rate estimation demonstrated a mean absolute error of 1.21 BPM and a root-mean-squared error (RMSE) of 1.97 BPM during walking after denoising with 9.16 BPM and 10.38 BPM improvements, respectively, compared to without denoising. Furthermore, the RMSEs of aortic opening and aortic closing time estimation after denoising for one dataset with catheter ground truth were 7.29 ms and 19.71 ms during walking, respectively, with 50.33 ms and 51.91 ms RMSE improvements compared to without denoising. And for another dataset with ICG-derived PEP ground truth, the RMSE of aortic opening time estimation after denoising was 10.21 ms during walking, with 38.74 ms RMSE improvement compared to without denoising. The proposed model attenuates motion noise from corrupted SCG signals while preserving cardiac information. This development paves the way for improved ambulatory cardiac health monitoring using wearable accelerometers during daily activities.

Fall detection (FD) systems are important assistive technologies for healthcare that can detect emergency fall events and alert caregivers. However, it is not easy to obtain large-scale annotated fall events with various specifications of sensors or sensor positions during the implementation of accurate FD systems. Moreover, the knowledge obtained through machine learning has been restricted to tasks in the same domain. The mismatch between different domains might hinder the performance of FD systems. Cross-domain knowledge transfer is very beneficial for machine-learning based FD systems to train a reliable FD model with well-labeled data in new environments. In this study, we propose domain-adaptive fall detection (DAFD) using deep adversarial training (DAT) to tackle cross-domain problems, such as cross-position and cross-configuration. The proposed DAFD can transfer knowledge from the source domain to the target domain by minimizing the domain discrepancy to avoid mismatch problems. The experimental results show that the average F1-score improvement when using DAFD ranges from 1.5% to 7% in the cross-position scenario, and from 3.5% to 12% in the cross-configuration scenario, compared to using the conventional FD model without domain adaptation training. The results demonstrate that the proposed DAFD successfully helps to deal with cross-domain problems and to achieve better detection performance.