Wireless, non-invasive, wearable device for continuous remote monitoring of hemodynamic parameters in a swine model of controlled hemorrhagic shock

Improved U-Net Model to Estimate Cardiac Output Based on Photoplethysmography and Arterial Pressure Waveform

We aimed to estimate cardiac output (CO) from photoplethysmography (PPG) and the arterial pressure waveform (ART) using a deep learning approach, which is minimally invasive, does not require patient demographic information, and is operator-independent, eliminating the need to artificially extract a feature of the waveform by implementing a traditional formula. We aimed to present an alternative to measuring cardiac output with greater accuracy for a wider range of patients. Using a publicly available dataset, we selected 543 eligible patients and divided them into test and training sets after preprocessing. The data consisted of PPG and ART waveforms containing 2048 points with the corresponding CO. We achieved an improvement based on the U-Net modeling framework and built a two-channel deep learning model to automatically extract the waveform features to estimate the CO in the dataset as the reference, acquired using the EV1000, a commercially available instrument. The model demonstrated strong consistency with the reference values on the test dataset. The mean CO was 5.01 ± 1.60 L/min and 4.98 ± 1.59 L/min for the reference value and the predicted value, respectively. The average bias was -0.04 L/min with a -1.025 and 0.944 L/min 95% limit of agreement (LOA). The bias was 0.79% with a 95% LOA between -20.4% and 18.8% when calculating the percentage of the difference from the reference. The normalized root-mean-squared error (RMSNE) was 10.0%. The Pearson correlation coefficient (r) was 0.951. The percentage error (PE) was 19.5%, being below 30%. These results surpassed the performance of traditional formula-based calculation methods, meeting clinical acceptability standards. We propose a dual-channel, improved U-Net deep learning model for estimating cardiac output, demonstrating excellent and consistent results. This method offers a superior reference method for assessing cardiac output in cases where it is unnecessary to employ specialized cardiac output measurement devices or when patients are not suitable for pulmonary-artery-catheter-based measurements, providing a viable alternative solution.

The Pig as a Translational Animal Model for Biobehavioral and Neurotrauma Research

In recent decades, the pig has attracted considerable attention as an important intermediary model animal in translational biobehavioral research due to major similarities between pig and human neuroanatomy, physiology, and behavior. As a result, there is growing interest in using pigs to model many human neurological conditions and injuries. Pigs are highly intelligent and are capable of performing a wide range of behaviors, which can provide valuable insight into the effects of various neurological disease states. One area in which the pig has emerged as a particularly relevant model species is in the realm of neurotrauma research. Indeed, the number of investigators developing injury models and assessing treatment options in pigs is ever-expanding. In this review, we examine the use of pigs for cognitive and behavioral research as well as some commonly used physiological assessment methods. We also discuss the current usage of pigs as a model for the study of traumatic brain injury. We conclude that the pig is a valuable animal species for studying cognition and the physiological effect of disease, and it has the potential to contribute to the development of new treatments and therapies for human neurological and psychiatric disorders.

The identification of blood pressure variation with hypovolemia based on the volume compensation method

Objective: The purpose of this study is to identify the blood pressure variation, which is important in continuous blood pressure monitoring, especially in the case of low blood volume, which is critical for survival. Methods: A pilot study was conducted to identify blood pressure variation with hypovolemia using five Landrace pigs. New multi-dimensional morphological features of Photoplethysmography (PPG) were proposed based on experimental study of hemorrhagic shock in pigs, which were strongly correlated with blood pressure changes. Five machine learning methods were compared to develop the blood pressure variation identification model. Results: Compared with the traditional blood pressure variation identification model with single characteristic based on single period area of PPG, the identification accuracy of mean blood pressure variation based on the proposed multi-feature random forest model in this paper was up to 90%, which was 17% higher than that of the traditional blood pressure variation identification model. Conclusion: By the proposed multi-dimensional features and the identification method, it is more accurate to detect the rapid variation in blood pressure and to adopt corresponding measures. Significance: Rapid and accurate identification of blood pressure variation under low blood volume ultimately has the potential to effectively avoid complications caused by abnormal blood pressure in patients with clinical bleeding trauma.

Telemedicine in Adult Congenital Heart Disease: Usefulness of Digital Health Technology in the Assistance of Critical Patients

The number of adults with congenital heart disease (ACHD) has progressively increased in recent years to surpass that of children. This population growth has produced a new demand for health care. Moreover, the 2019 coronavirus pandemic has caused significant changes and has underlined the need for an overhaul of healthcare delivery. As a result, telemedicine has emerged as a new strategy to support a patient-based model of specialist care. In this review, we would like to highlight the background knowledge and offer an integrated care strategy for the longitudinal assistance of ACHD patients. In particular, the emphasis is on recognizing these patients as a special population with special requirements in order to deliver effective digital healthcare.

Developing a real-time detection tool and an early warning score using a continuous wearable multi-parameter monitor

Background: Currently-used tools for early recognition of clinical deterioration have high sensitivity, but with low specificity and are based on infrequent measurements. We aimed to develop a pre-symptomatic and real-time detection and warning tool for potential patients’ deterioration based on multi-parameter real-time warning score (MPRT-WS).

Methods: A total of more than 2 million measurements were collected, pooled, and analyzed from 521 participants, of which 361 were patients in general wards defined at high-risk for deterioration and 160 were healthy participants allocation as controls. The risk score stratification was based on cutoffs of multiple physiological parameters predefined by a panel of specialists, and included heart rate, blood oxygen saturation (SpO2), respiratory rate, cuffless systolic and diastolic blood pressure (SBP and DBP), body temperature, stroke volume (SV), cardiac output, and systemic vascular resistance (SVR), recorded every 5 min for a period of up to 72 h. The data was used to define the various risk levels of a real-time detection and warning tool, comparing it with the clinically-used National Early Warning Score (NEWS).

Results: When comparing risk levels among patients using both tools, 92.6%, 6.1%, and 1.3% of the readings were defined as “Low”, “Medium”, and “High” risk with NEWS, and 92.9%, 6.4%, and 0.7%, respectively, with MPRT-WS (p = 0.863 between tools). Among the 39 patients that deteriorated, 30 patients received ‘High’ or ‘Urgent’ using the MPRT-WS (42.7 ± 49.1 h before they deteriorated), and only 6 received ‘High’ score using the NEWS. The main abnormal vitals for the MPRT-WS were SpO2, SBP, and SV for the “Urgent” risk level, DBP, SVR, and SBP for the “High” risk level, and DBP, SpO2, and SVR for the “Medium” risk level.

Conclusion: As the new detection and warning tool is based on highly-frequent monitoring capabilities, it provides medical teams with timely alerts of pre-symptomatic and real-time deterioration.

Assessing the use of a noninvasive monitoring system providing multiple cardio-pulmonary parameters following revascularization in STEMI patients

Background

Continuous monitoring of ECG, respiratory rate, systolic and diastolic blood pressure, pulse rate, cardiac output, and cardiac index is important in patients with ST-elevation myocardial infarction (STEMI) admitted to the intensive cardiac care unit (ICCU). However, monitoring these parameters in this setting and in these patients using noninvasive, wireless devices has not been conducted so far. We aimed to assess the use of a novel noninvasive continuous monitoring device in STEMI patients admitted to the ICCU.

Methods

Participants included STEMI patients that were admitted to the ICCU after primary percutaneous coronary intervention (PPCI). Patients were continuously monitored using a novel wearable chest patch monitor.

Results

Fifteen patients with STEMI who underwent PPCI were included in this study. The median age was 52.8 years, the majority were males, and the median body mass index (BMI) was 25.7. Monitoring lasted for 66  ±  16 hours, and included the automatic collection and recording of all vitals, freeing the nursing staff to focus on other tasks. The user experience of nurses as reflected in filled questionnaires showed high satisfaction rates in all aspects.

Conclusion

A novel noninvasive, wireless device showed high feasibility in continuously monitoring multiple crucial parameters in STEMI patients admitted to the ICCU after PPCI.

Diurnal cardio-respiratory changes in ambulatory individuals deciphered using a multi-parameter wearable device

Background

Recent technological developments enable big data-driven insights on diurnal changes. This study aimed to describe the trajectory of multiple and advanced parameters using a medical-grade wearable remote patient monitor.

Methods

Parameters were monitored for 24 h in 256 ambulatory participants who kept living their normal life. Parameters included heart rate, blood pressure, stroke volume, cardiac index, systemic vascular resistance, blood oxygen saturation, and respiratory rate. Diurnal variations were evaluated, and analyses were stratified based on sex, age, and body mass index.

Results

All parameters showed diurnal changes (p < 0.001). Females demonstrated higher heart rate and cardiac index with lower systemic vascular resistance. Obese participants had a higher blood pressure, and lower stroke volume and cardiac index. Systemic vascular resistance was higher among the elderly. Diurnal changes corresponded with awake-sleep hours and differed between sex, age, and body mass index groups.

Conclusion

Wearable monitoring platforms could decipher hemodynamic changes in subgroups of individuals, and might help with efforts to provide personalized medicine, pre-symptomatic diagnosis and prevention, and drug development.

Advanced Hemodynamic Monitoring Allows Recognition of Early Response Patterns to Diuresis in Congestive Heart Failure Patients

There are no clear guidelines for diuretic administration in heart failure (HF), and reliable markers are needed to tailor treatment. Continuous monitoring of multiple advanced physiological parameters during diuresis may allow better differentiation of patients into subgroups according to their responses. In this study, 29 HF patients were monitored during outpatient intravenous diuresis, using a noninvasive wearable multi-parameter monitor. Analysis of changes in these parameters during the course of diuresis aimed to recognize subgroups with different response patterns. Parameters did not change significantly, however, subgroup analysis of the last quartile of treatment showed significant differences in cardiac output, cardiac index, stroke volume, pulse rate, and systemic vascular resistance according to gender, and in systolic blood pressure according to habitus. Changes in the last quartile could be differentiated using k-means, a technique of unsupervised machine learning. Moreover, patients' responses could be best clustered into four groups. Analysis of baseline parameters showed that two of the clusters differed by baseline parameters, body mass index, and diabetes status. To conclude, we show that physiological changes during diuresis in HF patients can be categorized into subgroups sharing similar response trends, making noninvasive monitoring a potential key to personalized treatment in HF.

Ultra-Thin Flexible Encapsulating Materials for Soft Bio-Integrated Electronics

Recently, bioelectronic devices extensively researched and developed through the convergence of flexible biocompatible materials and electronics design that enables  more precise diagnostics and therapeutics in human health care and opens up the potential to expand into various fields, such as clinical medicine and biomedical research. To establish an accurate and stable bidirectional bio-interface, protection against the external environment and high mechanical deformation is essential for wearable bioelectronic devices. In the case of implantable bioelectronics, special encapsulation materials and optimized mechanical designs and configurations that provide electronic stability and functionality are required for accommodating various organ properties, lifespans, and functions in the biofluid environment. Here, this study introduces recent developments of ultra-thin encapsulations with novel materials that can preserve or even improve the electrical performance of wearable and implantable bio-integrated electronics by supporting safety and stability for protection from destruction and contamination as well as optimizing the use of bioelectronic systems in physiological environments. In addition, a summary of the materials, methods, and characteristics of the most widely used encapsulation technologies is introduced, thereby providing a strategic selection of appropriate choices of recently developed flexible bioelectronics.

Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos

Shock is one of the major killers in intensive care units, and early interventions can potentially reverse it. In this study, we advance a noncontact thermal imaging modality for continuous monitoring of hemodynamic shock working on 1,03,936 frames from 406 videos recorded longitudinally upon 22 pediatric patients. Deep learning was used to preprocess and extract the Center-to-Peripheral Difference (CPD) in temperature values from the videos. This time-series data along with the heart rate was finally analyzed using Long-Short Term Memory models to predict the shock status up to the next 6 h. Our models achieved the best area under the receiver operating characteristic curve of 0.81 ± 0.06 and area under the precision-recall curve of 0.78 ± 0.05 at 5 h, providing sufficient time to stabilize the patient. Our approach, thus, provides a reliable shock prediction using an automated decision pipeline that can provide better care and save lives.

Influence of Sex, BMI, and Skin Color on the Accuracy of Non-Invasive Cuffless Photoplethysmography-Based Blood Pressure Measurements

Vital signs obtained by photoplethysmography-based devices might be influenced by subcutaneous fat and skin color. This observational comparison study aimed to test the accuracy of blood pressure (BP) measurements between a photoplethysmography-based device and cuff-based BP device in ambulatory individuals, coming for a routine BP checkup. Systolic BP (SBP) and diastolic BP (DBP) measurements were stratified based on sex, BMI (<25; 25 ≤BMI<30; 30 ≤kg/m²), and skin color (types 1–3 and 4–6 by the Fitzpatrick scale). A total of 1548 measurements were analyzed. Correlations of SBP and DBP between the devices among males/females were between 0.914–0.987 (p < 0.001), and Bland-Altman analysis showed a bias of less than 0.5 mmHg for both sexes. Correlations of SBP and DBP between the devices among BMI groups were between 0.931–0.991 (p < 0.001), and Bland-Altman analysis showed a bias of less than 1 mmHg for all. Correlations of SBP and DBP between the devices among the skin color groups were between 0.936–0.983 (p < 0.001), and Bland-Altman analysis showed a bias of less than 1 mmHg for all. This study shows similar and high agreements between BP measurements obtained using a PPG-based non-invasive cuffless BP device and a cuff-based BP device across sex, BMI, and skin color groups.

Assessing the Usability of a Novel Wearable Remote Patient Monitoring Device for the Early Detection of In-Hospital Patient Deterioration: Observational Study

Background

Patients admitted to general wards are inherently at risk of deterioration. Thus, tools that can provide early detection of deterioration may be lifesaving. Frequent remote patient monitoring (RPM) has the potential to allow such early detection, leading to a timely intervention by health care providers.

Objective

This study aimed to assess the potential of a novel wearable RPM device to provide timely alerts in patients at high risk for deterioration.

Methods

This prospective observational study was conducted in two general wards of a large tertiary medical center. Patients determined to be at high risk to deteriorate upon admission and assigned to a telemetry bed were included. On top of the standard monitoring equipment, a wearable monitor was attached to each patient, and monitoring was conducted in parallel. The data gathered by the wearable monitors were analyzed retrospectively, with the medical staff being blinded to them in real time. Several early warning scores of the risk for deterioration were used, all calculated from frequent data collected by the wearable RPM device: these included (1) the National Early Warning Score (NEWS), (2) Airway, Breathing, Circulation, Neurology, and Other (ABCNO) score, and (3) deterioration criteria defined by the clinical team as a “wish list” score. In all three systems, the risk scores were calculated every 5 minutes using the data frequently collected by the wearable RPM device. Data generated by the early warning scores were compared with those obtained from the clinical records of actual deterioration among these patients.

Results

In total, 410 patients were recruited and 217 were included in the final analysis. The median age was 71 (IQR 62-78) years and 130 (59.9%) of them were male. Actual clinical deterioration occurred in 24 patients. The NEWS indicated high alert in 16 of these 24 (67%) patients, preceding actual clinical deterioration by 29 hours on average. The ABCNO score indicated high alert in 18 (75%) of these patients, preceding actual clinical deterioration by 38 hours on average. Early warning based on wish list scoring criteria was observed for all 24 patients 40 hours on average before clinical deterioration was detected by the medical staff. Importantly, early warning based on the wish list scoring criteria was also observed among all other patients who did not deteriorate.

Conclusions

Frequent remote patient monitoring has the potential for early detection of a high risk to deteriorate among hospitalized patients, using both grouped signal-based scores and algorithm-based prediction. In this study, we show the ability to formulate scores for early warning by using RPM. Nevertheless, early warning scores compiled on the basis of these data failed to deliver reasonable specificity. Further efforts should be directed at improving the specificity and sensitivity of such tools.

Trial Registration

ClinicalTrials.gov NCT04220359; https://clinicaltrials.gov/ct2/show/NCT04220359

Twenty-Four-Hour Ambulatory Blood Pressure Measurement Using a Novel Noninvasive, Cuffless, Wireless Device

BACKGROUND

Ambulatory blood pressure monitoring (ABPM) using cuff-based devices is used for diagnosis and treatment of hypertension. Technical limitations, low compliance, and complex procedures limit their use. The aim of the present study was to test the accuracy of a new photoplethysmography-based, wearable device (Wrist-monitor) as compared with the standard cuff-based ABPM device.

METHODS

Twenty-four-hour (24H) ABPM was performed in parallel for both devices on volunteers aged 18-65 years, while documenting their daily activities. Level of comfort and activity disturbance of both devices were recorded. Linear regression and Bland-Altman were used to evaluate the agreement between devices. Receiver operating characteristic (ROC) curve analysis was used to classify hypertension based on the average Wrist-monitor measurements as compared with a cuff-based ABPM device.

RESULTS

The study included 28 subjects (18 men) mean age 41.5 ± 16.2 years. Bland-Altman analysis resulted in 24H bias of -1.1 mm Hg for both diastolic blood pressure (DBP) and systolic blood pressure (SBP). Mean daytime bias was -1.9 mm Hg for DBP and SBP, while nighttime bias was smaller (0.7 and 0.4 mm Hg for DBP and SBP, respectively). ROC curve analysis yielded a mean area under the curve (AUC) of 1 for SBP and 24H blood pressure measurements. AUCs of 0.994 and 0.955 were found for the daytime DBP and night DBP, respectively. 24H ABPM with the Wrist-monitor caused significantly less inconvenience compared with the cuff-based device (P < 0.001).

CONCLUSIONS

The cuffless device provides comparable measurements to those obtained with the currently used cuff-based ABPM device, with significantly less inconvenience to the subject.

CLINICAL TRIALS REGISTRATION

Trial Number NCT03810586.

Exploring Early Pre-Symptomatic Detection of Influenza Using Continuous Monitoring of Advanced Physiological Parameters during a Randomized Controlled Trial

Early detection of influenza may improve responses against outbreaks. This study was part of a clinical study assessing the efficacy of a novel influenza vaccine, aiming to discover distinct, highly predictive patterns of pre-symptomatic illness based on changes in advanced physiological parameters using a novel wearable sensor. Participants were frequently monitored 24 h before and for nine days after the influenza challenge. Viral load was measured daily, and self-reported symptoms were collected twice a day. The Random Forest classifier model was used to classify the participants based on changes in the measured parameters. A total of 116 participants with ~3,400,000 data points were included. Changes in parameters were detected at an early stage of the disease, before the development of symptomatic illness. Heart rate, blood pressure, cardiac output, and systemic vascular resistance showed the greatest changes in the third post-exposure day, correlating with viral load. Applying the classifier model identified participants as flu-positive or negative with an accuracy of 0.81 ± 0.05 two days before major symptoms appeared. Cardiac index and diastolic blood pressure were the leading predicting factors when using data from the first and second day. This study suggests that frequent remote monitoring of advanced physiological parameters may provide early pre-symptomatic detection of flu.

Continuous Remote Patient Monitoring Shows Early Cardiovascular Changes in COVID-19 Patients

COVID-19 exerts deleterious cardiopulmonary effects, leading to a worse prognosis in the most affected. This retrospective multi-center observational cohort study aimed to analyze the trajectories of key vitals amongst hospitalized COVID-19 patients using a chest-patch wearable providing continuous remote patient monitoring of numerous vital signs. The study was conducted in five COVID-19 isolation units. A total of 492 COVID-19 patients were included in the final analysis. Physiological parameters were measured every 15 min. More than 3 million measurements were collected including heart rate, systolic and diastolic blood pressure, cardiac output, cardiac index, systemic vascular resistance, respiratory rate, blood oxygen saturation, and body temperature. Cardiovascular deterioration appeared early after admission and in parallel with changes in the respiratory parameters, showing a significant difference in trajectories within sub-populations at high risk. Early detection of cardiovascular deterioration of COVID-19 patients is achievable when using frequent remote patient monitoring.

A Pilot Study of Blood Pressure Monitoring After Cardiac Surgery Using a Wearable, Non-invasive Sensor

Background: Continuous blood pressure (BP) measurement in intensive care units is based on arterial line (AL) transducers, sometimes associated with clinical complications. Our objective was to evaluate continuous BP measurements obtained from a non-invasive, wireless photoplethysmography (PPG)-based device using two distinct configurations (wristwatch and chest-patch monitors) compared to an AL. Methods: In this prospective evaluation study, comparison of the PPG-based devices to the AL was conducted in 10 patients immediately following cardiac surgery. Pulse rate (PR), systolic BP (SBP), diastolic BP (DBP), and mean arterial pressure (MAP) were recorded using both the AL and the PPG-based devices simultaneously for an average of 432 ± 290 min starting immediately after cardiac surgery. Bland-Altman plots and Pearson's correlations were used to assess the accuracy and degree of agreement between techniques. Results: A total of ~4,000 data points were included in the final analysis. AL measurements for PR, SBP, DBP and MAP were significantly (p < 0.001) and strongly correlated with both the wristwatch (r = 0.99, r = 0.94, r = 0.93 and r = 0.96, respectively) and the chest-patch (r = 0.99, r = 0.95, r = 0.93 and r = 0.95, respectively) monitors. Both configurations showed a marginal bias of <1 mmHg for BP measurements and <1 beat/min for PR [95% limits of agreement -3,3 beat/min; BP measurements: (-6)-(-10), 6-10 mmHg] compared to AL measurements. Conclusion: The PPG-based devices offer a high level of accuracy for cardiac-related parameters compared to an AL in post-cardiac surgery patients. Such devices could provide advanced monitoring capabilities in a variety of clinical settings, including immediate post-operative and intensive care unit settings. Clinical Trial Registration:www.clinicaltrials.gov, NCT03603860.