Michael Chan

Acute psychological stress has a multifaceted impact on cardiovascular physiology, including increases in chronotropy, inotropy, and vascular tone. Chronic exposure to stress may greatly increase cardiovascular risk. To examine the cardiovascular impact of acute stress comprehensively, new multimodal portable monitoring solutions are needed. We examined the feasibility of using a compact multimodal wearable patch to measure laboratory-based stress-induced cardiovascular responses in a diverse sample with recent myocardial infarction (MI) and healthy participants (N = 37, 28 MI) during a protocol with a public-speaking stressor. Using the electrocardiogram, seismocardiogram, and photoplethysmogram captured from the device, we found several significant (p < 0.05) autonomic changes in the pooled sample suggesting stress activation: increases in heart rate, chest photoplethysmogram amplitude, and perfusion index and decreases in heart rate variability, left ventricular ejection time, pulse arrival time, and pulse transit time. We, thus, demonstrated that a single wearable device can capture stress-induced cardiovascular changes, enabling simultaneous examination of stress-induced inotropic, chronotropic, and vascular effects. This portable, wireless chest patch may be useful in comprehensively and unobtrusively examining stress-induced cardiovascular effects in-lab. Given the public health importance of psychological stress and cardiovascular disease, future studies should assess the device’s full clinical potential in larger groups with longer monitoring periods.

Objective: To develop a person-specific dynamic modeling approach to quantify the time course of multiple peripheral markers of the autonomic nervous system (ANS) in response to acute stressors and potential stress-reducing interventions. Methods: We curated data (N = 50 participants) from a double-blind, randomized, sham-controlled trial of non-invasive vagus nerve stimulation (nVNS) for posttraumatic stress disorder (PTSD). For each participant, a multi-input, multi-output, linear state space model (SSM) was trained on $\sim$3500 s of cardiovascular and respiratory marker time series data and tested for predictive validity on $\sim$2000 s of held-out data. The inputs to each SSM indicated the presence or absence of potential stressors (e.g., traumatic memories) and interventions (e.g., nVNS or sham stimulation). We analyzed the SSMs' step responses to each input and compared the responses to real data. We then simulated the effects of just-in-time nVNS delivered during a traumatic memory. Results: The SSMs outperformed baseline forecasting methods on held-out data (P $< $ .05). Responses to nVNS were in the opposite direction of responses to traumatic memories, with neutral conditions (included for comparison) remaining in between. For participants with PTSD, just-in-time nVNS attenuated–and briefly reversed–the response to traumatic memories along the principal axis of variance, which explained $\sim$50% or more variance and mirrored expected ANS changes. Just-in-time sham stimulation produced no attenuation. Conclusion: Our methods capture latent ANS dynamics during increasing and decreasing stress levels associated with traumatic memories and nVNS, respectively. Significance: Traumatic memories can cause pathological stress responses during daily life that just-in-time non-invasive neuromodulation may potentially help mitigate.

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.

Objective: This study explores the potential of active vibrational sensing as a digital biomarker to identify and characterize inflammatory symptomatology in the Achilles tendon and its entheses in juvenile idiopathic arthritis (JIA), particularly enthesitis related arthritis (ERA), a subcategory of JIA. Methods: Active vibrational data were non-invasively recorded using a miniature coin vibration motor and accelerometer. Twenty active vibration recordings from children diagnosed with JIA were used in the analysis. Machine learning algorithms were leveraged to classify the vibrational signatures according to the corresponding subject groups. Subjects were classified into symptomatic ERA (sxERA), asymptomatic ERA (asxERA), and asymptomatic JIA (non-ERA) (asxNERA) groups based on clinical evaluations and ILAR criteria. Results: Distinct vibrational signatures were observed during tiptoe standing, providing differentiation between subject groups. Feature-based and waveform-based approaches effectively classified the sxERA group against asxNERA and asxERA groups using leave-one-subject-out (LOSO-CV) and 3-fold cross-validation. For the 3-fold cross-validation, the mean accuracies for distinguishing sxERA from asxNERA were 81% (feature-based) and 81% (waveform-based), while the accuracies for discriminating sxERA against asxERA were 73% (feature-based) and 74% (waveform-based). Conclusion: Active vibrational sensing demonstrates promise as a tool for identifying Achilles tendon inflammation in JIA, potentially aiding in early diagnosis and disease monitoring. Significance: Developing active vibrational sensing as a diagnostic modality could address challenges in diagnosing ERA and facilitate timely intervention and personalized care for JIA, potentially enhancing long-term patient outcomes.

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.

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.

Respiratory rate (RR) is an important vital sign to monitor outside the clinic, particularly during physiological challenges such as exercise; unfortunately, ambulatory measurement devices for RR are typically obtrusive and inaccurate. The objective of this work is to allow for accurate and robust RR monitoring with a convenient and small chest-worn wearable patch during walking and exercise recovery periods. Methods: To estimate RR from the wearable patch, respiratory signals were first extracted from electrocardiogram (ECG), photoplethysmogram (PPG), and seismocardiogram (SCG) signals. The optimal channel in each signal was adaptively selected using respiratory quality index based on Fast Fourier Transform (RQI _FFT ). Next, we proposed modality attentive (MA) fusion–which merged spectral-temporal information from different modalities–to address motion artifacts during walking. The fused output was subsequently denoised using a U-Net-based deep learning model and used for final estimation. A dataset of N = 17 subjects was collected to validate RR estimated during three types of activities: stationary activities, walking (including six-minute-walk-test), and running. Major Results: Combining and denoising ECG and PPG data using MA fusion and the U-Net achieved the lowest mean absolute error (2.21 breaths per minute [brpm]) during walking. After rejecting a small portion of the data (coverage=84.43%) using RQI _FFT , this error was further reduced to 1.59 brpm, which was comparable to the state-of-the-art methods. Conclusion: Applying adaptive channel selection, MA fusion, and U-Net denoising achieved accurate RR estimation from a small chest-worn wearable patch. Significance: This work can enable cardiopulmonary monitoring applications in less controlled setting.

Treating opioid use disorder (OUD) is a significant healthcare challenge in the United States. Remaining abstinent from opioids is challenging for individuals with OUD due to withdrawal symptoms that include restlessness. However, to our knowledge, studies of acute withdrawal have not quantified restlessness using involuntary movements. We hypothesized that wearable accelerometry placed mid-sternum could be used to detect withdrawal-related restlessness in patients with OUD. To study this, 23 patients with OUD undergoing active withdrawal participated in a protocol involving wearable accelerometry, opioid cues to elicit craving, and non-invasive Vagal Nerve Stimulation (nVNS) to dampen withdrawal symptoms. Using accelerometry signals, we analyzed how movements correlated with changes in acute withdrawal severity, measured by the Clinical Opioid Withdrawal Scale (COWS). Our results revealed that patients demonstrating sinusoidal–i.e., predominantly single-frequency oscillation patterns in their motion almost exclusively demonstrated an increase in the COWS, and a strong relationship between the maximum power spectral density and increased withdrawal over time, measured by the COWS (R = 0.92, p = 0.029). Accelerometry may be used in an ambulatory setting to indicate the increased intensity of a patient’s withdrawal symptoms, providing an objective, readily-measurable marker that may be captured ubiquitously.

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.