Missing Data Statistics Provide Causal Insights into Data Loss in Diabetes Health Monitoring by Wearable Sensors.

BACKGROUND: Data loss in wearable sensors is an inevitable problem that leads to misrepresentation during diabetes health monitoring. We systematically investigated missing wearable sensors data to get causal insight into the mechanisms leading to missing data. METHODS: Two-week-long data from a continuous glucose monitor and a Fitbit activity tracker recording heart rate (HR) and step count in free-living patients with type 2 diabetes mellitus were used. The gap size distribution was fitted with a Planck distribution to test for missing not at random (MNAR) and a difference between distributions was tested with a Chi-squared test. Significant missing data dispersion over time was tested with the Kruskal-Wallis test and Dunn post hoc analysis. RESULTS: Data from 77 subjects resulted in 73 cleaned glucose, 70 HR and 68 step count recordings. The glucose gap sizes followed a Planck distribution. HR and step count gap frequency differed significantly (p < 0.001), and the missing data were therefore MNAR. In glucose, more missing data were found in the night (23:00-01:00), and in step count, more at measurement days 6 and 7 (p < 0.001). In both cases, missing data were caused by insufficient frequency of data synchronization. CONCLUSIONS: Our novel approach of investigating missing data statistics revealed the mechanisms for missing data in Fitbit and CGM data.

Circadian assessment of heart failure using explainable deep learning and novel multi-parameter polar images.

BACKGROUND AND OBJECTIVE: Heart failure (HF) is a multi-faceted and life-threatening syndrome that affects more than 64.3 million people worldwide. Current gold-standard screening technique, echocardiography, neglects cardiovascular information regulated by the circadian rhythm and does not incorporate knowledge from patient profiles. In this study, we propose a novel multi-parameter approach to assess heart failure using heart rate variability (HRV) and patient clinical information. METHODS: In this approach, features from 24-hour HRV and clinical information were combined as a single polar image and fed to a 2D deep learning model to infer the HF condition. The edges of the polar image correspond to the timely variation of different features, each of which carries information on the function of the heart, and internal illustrates color-coded patient clinical information. RESULTS: Under a leave-one-subject-out cross-validation scheme and using 7,575 polar images from a multi-center cohort (American and Greek) of 303 coronary artery disease patients (median age: 58 years [50-65], median body mass index (BMI): 27.28 kg/m2 [24.91-29.41]), the model yielded mean values for the area under the receiver operating characteristics curve (AUC), sensitivity, specificity, normalized Matthews correlation coefficient (NMCC), and accuracy of 0.883, 90.68%, 95.19%, 0.93, and 92.62%, respectively. Moreover, interpretation of the model showed proper attention to key hourly intervals and clinical information for each HF stage. CONCLUSIONS: The proposed approach could be a powerful early HF screening tool and a supplemental circadian enhancement to echocardiography which sets the basis for next-generation personalized healthcare.

Cardiovascular hardware simulator and artificial aorta-generated central blood pressure waveform database according to various vascular ages for cardiovascular health monitoring applications.

This study presents a database of central blood pressure waveforms according to cardiovascular health conditions, to supplement the lack of clinical data in cardiovascular health research, constructed by a cardiovascular simulator. Blood pressure (BP) is the most frequently measured biomarker, and in addition to systolic and diastolic pressure, its waveform represents the various conditions of cardiovascular health. A BP waveform is formed by overlapping the forward and reflected waves, which are affected by the pulse wave velocity (PWV). The increase in vascular stiffness with aging increases PWV, and the PWV-age distribution curve is called vascular age. For cardiovascular health research, extensive data of central BP waveform is essential, but the clinical data published so far are insufficient and imbalanced in quantity and quality. This study reproduces the central BP waveform using a cardiovascular hardware simulator and artificial aortas, which mimic the physiological structure and properties of the human. The simulator can adjust cardiovascular health conditions to the same level as humans, such as heart rate of 40-100 BPM, stroke volume of 40-100 mL, and peripheral resistance of 12 steps. Also, 6 artificial aortas with vascular ages in the 20-70 were fabricated to reproduce the increase in vascular stiffness due to aging. Vascular age calculated from measured stiffness of artificial aorta and central BP waveform showed an error of less than 3 years from the clinical value. Through this, a total of 636 waveforms were created to construct a central BP waveform database according to controlled various cardiovascular health conditions.

Effect of lower body negative pressure on cardiac and cerebral function in postural orthostatic tachycardia syndrome: A pilot MRI assessment.

Postural orthostatic tachycardia syndrome (POTS) is characterized by an excessive heart rate (HR) response upon standing and symptoms indicative of inadequate cerebral perfusion. We tested the hypothesis that during lower body negative pressure (LBNP), individuals with POTS would have larger decreases in cardiac and cerebrovascular function measured using magnetic resonance (MR) imaging. Eleven patients with POTS and 10 healthy controls were studied at rest and during 20 min of -25 mmHg LBNP. Biventricular volumes, stroke volume (SV), cardiac output (Qc), and HR were determined by cardiac MR. Cerebral oxygen uptake (VO2 ) in the superior sagittal sinus was calculated from cerebral blood flow (CBF; MR phase contrast), venous O2 saturation (SvO2 ; susceptometry-based oximetry), and arterial O2 saturation (pulse oximeter). Regional cerebral perfusion was determined using arterial spin labelling. HR increased in response to LBNP (p < 0.001) with no group differences (HC: +9 ± 8 bpm; POTS: +13 ± 11 bpm; p = 0.35). Biventricular volumes, SV, and Qc decreased during LBNP (p < 0.001). CBF and SvO2 decreased with LBNP (p = 0.01 and 0.03, respectively) but not cerebral VO2 (effect of LBNP: p = 0.28; HC: -0.2 ± 3.7 mL/min; POTS: +1.1 ± 2.0 mL/min; p = 0.33 between groups). Regional cerebral perfusion decreased during LBNP (p < 0.001) but was not different between groups. These data suggest patients with POTS have preserved cardiac and cerebrovascular function.

A Preliminary Investigation about the Influence of WIMU PROTM Location on Heart Rate Accuracy: A Comparative Study in Cycle Ergometer.

Technological development has boosted the use of multi-sensor devices to monitor athletes' performance, but the location and connectivity between devices have been shown to affect data reliability. This preliminary study aimed to determine whether the placement of a multi-sensor device (WIMU PROTM) could affect the heart rate signal reception (GARMINTM chest strap) and, therefore, data accuracy. Thirty-two physical education students (20 men and 12 women) performed 20 min of exercise in a cycle ergometer based on the warm-up of the Function Threshold Power 20 test in laboratory conditions, carrying two WIMU PROTM devices (Back: inter-scapula; Bicycle: bicycle's handlebar-20 cm from the chest) and two GARMINTM chest straps. A one-dimensional statistical parametric mapping test found full agreement between the two situations (inter-scapula vs. bicycle's handlebar). Excellent intra-class correlation values were obtained during the warm-up (ICC = 0.99, [1.00-1.00], p < 0.001), the time trial test (ICC = 0.99, [1.00-1.00], p < 0.001) and the cool-down (ICC = 0.99, [1.00-1.00], p < 0.001). The Bland-Altman plots confirmed the total agreement with a bias value of 0.00 ± 0.1 bpm. The interscapular back placement of the WIMU PROTM device does not affect heart rate measurement accuracy with a GARMINTM chest strap during cycling exercise in laboratory conditions.

Pulse rate variability and health-related quality of life assessment with the Short Form-8 Japanese version in the general Japanese population.

We aimed to investigate the association between pulse rate variability (PRV) and health-related quality of life (HRQOL) in the general population. A cross-sectional study was conducted with 5908 Japanese men and women aged 30-79 years. PRV was assessed at rest using 5-min recordings of pulse waves with a photoplethysmographic signal from a fingertip sensor, and the time and frequency domains of PRV were determined. HRQOL was assessed with the Short Form-8 (SF-8) Japanese version, and poor HRQOL was defined as an SF-8 sub-scale score < 50. A test for nonlinear trends was performed with the generalized additive model with a smoothing spline adjusted for confounders. The lowest multivariable-adjusted odds ratios for poor physical component score were found in those who had second or third quartile levels of standard deviation of normal-to-normal intervals (SDNN) and root mean square of successive difference (RMSSD), and high-frequency (HF) power and trended slightly upward in the higher levels. PRV-derived parameters were nonlinearly associated with poor physical component scores. In conclusion, reduced PRV-derived SDNN, RMSSD and HF power were associated with poor HRQOL in the domain of physical function. Higher levels of these parameters did not necessarily translate into better HRQOL.

Contactless face video based vital signs detection framework for continuous health monitoring using feature optimization and hybrid neural network.

Continuous monitoring of vital signs such as respiration and heart rate is essential to detect and predict conditions that may affect the patient's well-being. To detect these vital signs most medical systems use contact sensors. They are not feasible for long term monitoring and are not repeatable. Vital signs using facial video-noncontact monitoring are becoming increasingly important. Researchers in the last few years although considerable progress has been made, challenging datasets absence timing of assessment process and the technology still has some limitations such as time consuming nature and lack of computer portability. To solve those problems, we propose a contactless video based vital signs detection framework for continuous health monitoring using feature optimization and hybrid neural network. In the proposed technique, modified war strategy optimization algorithm is proposed to segment the face portion from the input video frames. Then, we utilize the known data acquisition models to extract vital signs from the segmented face portions are heart rate, blood pressure, respiratory rate and oxygen saturation. An improved neural network structure (Lifting Net) is further used to achieve the adaptive extraction of deep hidden features for specific signs, for realizing the high precision of human health monitoring. The Hughes effect or dimensionality issue affects detection accuracy in sign classification when there are fewer training instances relative to the number of spectral features. The problem can be overcome through feature optimization here Northern goshawk optimization algorithm is used to select optimal best features which reduces the data dimensionality issue. Furthermore, hybrid deep ensemble reinforcement learning classifier is proposed for the human vital sign detection and classification which ensures the early detection of patient abnormality. Finally, we validate our framework using benchmark video datasets such as TokyoTechrPPG, PURE and COHFACE. To proves the effectiveness of proposed technique using simulation results and comparative analysis.

Preclinical cardiovascular safety assessment of pharmacology-toxicology relationship for a set of novel kinase inhibitors.

Cardiovascular toxicity is one of the more common causes of attrition in preclinical and clinical drug development. Preclinical cardiovascular safety assessment involves numerous in vitro and in vivo endpoints which are being continually reviewed and improved to lower the incidence of cardiovascular toxicity that manifests only after the initiation of clinical trials. An example of notable preclinical toxicity is necrosis in the papillary muscle of the left ventricle in dogs that is induced by exaggerated pharmacological effects of vasodilators or positive inotropic/vasodilating off-target drug effects. Two distinct, small-molecule inhibitors that target an intracellular kinase, Compound A and Compound B, were profiled in 2-week dose-range finding and 4-week toxicity studies. Serum cardiac troponin (cTnI) was evaluated after a single dose and after 2-week and 4-week repeat dose studies with each kinase inhibitor. Acute effects on hemodynamic (heart rate, blood pressures, left ventricular contractility) and electrocardiographic (QTcV, PR, QRS intervals) endpoints by each inhibitor were assessed in an anesthetized dog cardiovascular model. Cardiovascular degeneration/necrosis with and without fibrosis was observed in dogs and correlated to increases in serum cTnI in repeat-dose toxicity studies. At the same doses used in toxicologic assessments, both kinase inhibitors produced sustained increases in heart rate, left ventricular contractility, and cardiac output, and decreases in mean arterial pressure. Cardiac pathology findings associated with these 2 kinase inhibitors were accompanied not only by cardiac troponin elevations but also associated with hemodynamic changes, highlighting the importance of the link of the physiologic-toxicologic interplay in cardiovascular safety assessment.

Autonomic activity and cardiovascular system risk assessment in pediatric patients with hemolytic uremic syndrome.

In pediatric patients with hemolytic uremic syndrome (HUS), cardiac involvement and autonomic nervous system function can be evaluated by a non-invasive method called heart rate variability (HRV). This study aims to evaluate heart rate variability and electrocardiography findings in patients with HUS by comparing a healthy group. Patients who are diagnosed with HUS at a university hospital from December 2020 to June 2022 are screened by electrocardiography (ECG), echocardiography, and 24-h Holter ECG. A healthy control group, compatible in age and gender with the patient group, was selected from healthy subjects. HRV parameters, laboratory values, and ECG findings were analyzed and compared with the healthy group and each other. There were 25 patients with HUS and 51 participants in the healthy control group. Statistically significant differences were found in some HRV parameters: standard deviation of normal to normal intervals, the mean of the 5-min RR interval standard deviations, the standard deviation of 5-min RR interval means, the triangular interpolation of normal to normal interval, and very-low-frequency power. HUS patients had impaired and declined HRV values compared to the healthy group. There was a significant decrease in the PR distance, while a significant increase in the corrected QT and QT dispersion values was detected in the electrocardiographic findings of the patient group. HRV values impaired as renal failure parameters increased.  Conclusion: Patients with HUS may have autonomic nervous system dysfunction. HRV measurement is a non-invasive method that can evaluate this. It can be thought that there may be an increased risk of cardiovascular events and arrhythmias in some patients with HUS. ECG should be also considered to detect arrhythmia. What is Known: • Hemolytic uremic syndrome (HUS) primarily effects the hematologic parameters and kidney. • Secondary cardiomyopathy with hypertension and renal failure could be observed in these patients. • Rhythm problems are not expected primarily in these patients. • There is very limited data in evaluating autonomic function and arrhythmia risk for these patients. What is New: • Patients with HUS may have autonomic nervous system dysfunction. • HRV measurement is a non-invasive method that can evaluate this. • Cardiovascular events and arrhythmias due to the deterioration of the balance between the sympathetic and parasympathetic systems could manifest in patients with HUS. • An ECG and screening patients for cardiac events, and monitoring them closely should be considered.

A Lightweight Hybrid Model Using Multiscale Markov Transition Field for Real-Time Quality Assessment of Photoplethysmography Signals.

OBJECTIVE: The proliferation of wearable devices has escalated the standards for photoplethysmography (PPG) signal quality. This study introduces a lightweight model to address the imperative need for precise, real-time evaluation of PPG signal quality, followed by its deployment and validation utilizing our integrated upper computer and hardware system. METHODS: Multiscale Markov Transition Fields (MMTF) are employed to enrich the morphological information of the signals, serving as the input for our proposed hybrid model (HM). HM undergoes initial pre-training utilizing the MIMIC-III and UCI databases, followed by fine-tuning the Queensland dataset. Knowledge distillation (KD) then transfers the large-parameter model's knowledge to the lightweight hybrid model (LHM). LHM is subsequently deployed on the upper computer for real-time signal quality assessment. RESULTS: HM achieves impressive accuracies of 99.1% and 96.0% for binary and ternary classification, surpassing current state-of-the-art methods. LHM, with only 0.2 M parameters (0.44% of HM), maintains high accuracy despite a 2.6% drop. It achieves an inference speed of 0.023 s per image, meeting real-time display requirements. Furthermore, LHM attains a 97.7% accuracy on a self-created database. HM outperforms current methods in PPG signal quality accuracy, demonstrating the effectiveness of our approach. Additionally, LHM substantially reduces parameter count while maintaining high accuracy, enhancing efficiency and practicality for real-time applications. CONCLUSION: The proposed methodology demonstrates the capability to achieve high-precision and real-time assessment of PPG signal quality, and its practical validation has been successfully conducted during deployment. SIGNIFICANCE: This study contributes a convenient and accurate solution for the real-time evaluation of PPG signals, offering extensive application potential.

Workload associated with manual assessment of vital signs as compared with continuous wireless monitoring.

BACKGROUND: Vital sign monitoring is considered an essential aspect of clinical care in hospitals. In general wards, this relies on intermittent manual assessments performed by clinical staff at intervals of up to 12 h. In recent years, continuous monitoring of vital signs has been introduced to the clinic, with improved patient outcomes being one of several potential benefits. The aim of this study was to determine the workload difference between continuous monitoring and manual monitoring of vital signs as part of the National Early Warning Score (NEWS). METHODS: Three wireless sensors continuously monitored blood pressure, heart rate, respiratory rate, and peripheral oxygen saturation in 20 patients admitted to the general hospital ward. The duration needed for equipment set-up and maintenance for continuous monitoring in a 24-h period was recorded and compared with the time spent on manual assessments and documentation of vital signs performed by clinical staff according to the NEWS. RESULTS: The time used for continuous monitoring was 6.0 (IQR 3.2; 7.2) min per patient per day vs. 14 (9.7; 32) min per patient per day for the NEWS. Median difference in duration for monitoring of vital signs was 9.9 (95% CI 5.6; 21) min per patient per day between NEWS and continuous monitoring (p < .001). Time used for continuous monitoring in isolated patients was 6.6 (4.6; 12) min per patient per day as compared with 22 (9.7; 94) min per patient per day for NEWS. CONCLUSION: The use of continuous monitoring was associated with a significant reduction in workload in terms of time for monitoring as compared with manual assessment of vital signs.

Postexercise cardiovascular hemodynamics assessment before and after a 30-minute standing still recovery.

BACKGROUND: Although postexercise syncope usually occurs shortly after physical exercise conclusion, athletes commonly reveal symptoms of postexercise hypotension several tens of minutes after exercise completion. Currently, no studies have investigated central hemodynamic regulation during posture changes occurring several tens of minutes after exercise compared to immediately after cessation. METHODS: This study examined changes in mean arterial pressure (MAP), heart rate (HR), systemic vascular conductance (SVC), cardiac output, and stroke volume during two sets of tilt tests performed before vs. after a 30-minute standing still recovery, respectively. Tilt tests were performed after a short-lasting supramaximal test (WNG) and long-lasting maximal incremental test (INC) in 12 young endurance-trained individuals. RESULTS: The key findings were that, regardless of the exercise type, the 30-minute recovery augmented (P<0.01) the increase in HR and the drop in SVC during the transition from supine to upright, although the MAP drop was similar (P=0.99) after vs. before recovery. INC led to greater increases (P<0.01) in HR and drops (P<0.01) in SVC compared to WNG during postural transitions both before and after the recovery. CONCLUSIONS: These findings suggest that, in a population that tolerates postexercise hypotension, MAP neural control is more challenged after a 30-minute standing still recovery than before, as evidenced by an augmented vasodilation capacity along with an increased HR buffering response during posture changes. Moreover, our data suggest that effective MAP control is resulting from an equally effective HR buffering response on MAP. Therefore, exercises that induce greater systemic vasodilation lead to greater HR buffering responses.

Applicability of heart rate variability for cardiac autonomic assessment in long-term COVID patients: A systematic review.

PURPOSE: To carry out a systematic review to determine the main methods used to study the heart rate variability (HRV) in individuals after the acute phase of COVID-19. METHODS: The study followed the Preferred Items for Reporting for Systematic Reviews and Meta-Analyses (PRISMA) statement. PubMed, Web of Science, Scopus and CINAHAL electronic databases were searched from the inception to November 2022. The studies were included if they used HRV assessment based on linear and non-linear methods in long-term COVID-19 patients. Review studies, theses and dissertations, conference abstracts, longitudinal studies, studies conducted on animals and studies that included individuals in the acute phase of the COVID-19 were excluded. The methodological quality of the studies was analyzed using the Joanna Briggs Institute's critical evaluation checklist for cross-sectional analytical studies. RESULTS: HRV was mainly assessed using 24-h Holter monitoring in 41.6% (5/12) of the studies, and 12­lead ECG was used in 33.3% (4/12). Regarding the type of assessment, 66.6% (8/12) of the studies only used linear analysis, where 25% (3/12) used analysis in the time domain, and 41.6% (5/12) used both types. Non-linear methods were combined with the previously cited linear method in 25% (3/12) of the studies. Moreover, 50% (6/12) of the studies demonstrated post-COVID-19 autonomic dysfunction, with an increase in the predominance of cardiac sympathetic modulation. The average score of the evaluation checklist was 6.6, characterized as having reasonable methodological quality. CONCLUSION: 24-h Holter and 12­lead ECG are considered effective tools to assess HRV in post-COVID-19 patients. Furthermore, the findings reveal diverse effects of COVID-19 on the autonomic nervous system's sympathovagal balance, which might be influenced by secondary factors such as disease severity, patients' overall health, evaluation timing, post-infection complications, ventilatory functions, and age.

Impact of maternal obesity on resting muscle sympathetic nerve activity during uncomplicated pregnancy: a longitudinal assessment.

Maternal obesity increases the risk of adverse pregnancy outcomes. The mechanisms that contribute to this elevated risk are unclear but may be related to greater activity of the sympathetic nervous system, which is associated with hypertensive disorders of pregnancy. We hypothesized that resting muscle sympathetic nerve activity (MSNA) would be greater in women with obesity during pregnancy when compared with normal-weight women. Blood pressure, heart rate, and MSNA were recorded during 5 min of supine rest in 14 normal-weight women [body mass index (BMI) 22.1 ± 2.1 (SD) kg/m2] and 14 women with obesity (BMI 33.9 ± 3.5 kg/m2) during (early and late) pregnancy and postpartum. All women had uncomplicated pregnancies. Resting MSNA burst frequency was not different between groups during early (normal weight 17 ± 10 vs. obesity 22 ± 15 bursts/min, P = 0.35) but was significantly greater in the obesity group during late pregnancy (23 ± 13 vs. 35 ± 15 bursts/min, P = 0.031) and not different postpartum (10 ± 6 vs. 9 ± 7 bursts/min, P = 0.74). These findings were also apparent when comparing burst incidence and total activity. Although still within the normotensive range, systolic blood pressure was greater in the obesity group across all time points (P = 0.002). Diastolic blood pressure was lower during pregnancy compared with postpartum (P < 0.001) and not different between groups (P = 0.488). Heart rate increased throughout pregnancy in both groups (P < 0.001). Our findings suggest that maternal obesity is associated with greater increases in sympathetic activity even during uncomplicated pregnancy. Future research is needed to determine if this is linked with an increased risk of adverse outcomes or is required to maintain homeostasis in pregnancy.NEW & NOTEWORTHY The impact of maternal obesity on resting muscle sympathetic nerve activity was examined during (early and late) and after uncomplicated pregnancy. Resting muscle sympathetic nerve activity is not different during early pregnancy or postpartum but is significantly elevated in women with obesity during late pregnancy when compared with normal-weight women. Future research is needed to determine if this is linked with an increased risk of adverse outcomes or is required to maintain homeostasis in pregnancy.

A Discolorable Flexible Synaptic Transistor for Wearable Health Monitoring.

Multifunctional intelligent wearable electronics, providing integrated physiological signal analysis, storage, and display for real-time and on-site health status diagnosis, have great potential to revolutionize health monitoring technologies. Advanced wearable systems combine isolated digital processor, memory, and display modules for function integration; however, they suffer from compatibility and reliability issues. Here, we introduce a flexible multifunctional electrolyte-gated transistor (EGT) that integrates synaptic learning, memory, and autonomous discoloration functionalities for intelligent wearable application. This device exhibits synergistic light absorption coefficient changes during voltage-gated ion doping that modulate the electrical conductance changes for synaptic function implementation. By adaptively changing color, the EGT can differentiate voltage pulse inputs with different frequency, amplitude, and duration parameters, exhibiting excellent reversibility and reliability. We developed a smart wearable monitoring system that incorporates EGT devices and sensors for respiratory and electrocardiogram signal analysis, providing health warnings through real-time and on-site discoloration. This study represents a significant step toward smart wearable technologies for health management, offering health evaluation through intelligent displays.

BCG Signal Quality Assessment Based on Time-Series Imaging Methods.

This paper describes a signal quality classification method for arm ballistocardiogram (BCG), which has the potential for non-invasive and continuous blood pressure measurement. An advantage of the BCG signal for wearable devices is that it can easily be measured using accelerometers. However, the BCG signal is also susceptible to noise caused by motion artifacts. This distortion leads to errors in blood pressure estimation, thereby lowering the performance of blood pressure measurement based on BCG. In this study, to prevent such performance degradation, a binary classification model was created to distinguish between high-quality versus low-quality BCG signals. To estimate the most accurate model, four time-series imaging methods (recurrence plot, the Gramain angular summation field, the Gramain angular difference field, and the Markov transition field) were studied to convert the temporal BCG signal associated with each heartbeat into a 448 × 448 pixel image, and the image was classified using CNN models such as ResNet, SqueezeNet, DenseNet, and LeNet. A total of 9626 BCG beats were used for training, validation, and testing. The experimental results showed that the ResNet and SqueezeNet models with the Gramain angular difference field method achieved a binary classification accuracy of up to 87.5%.

An Automatic Remote Health Risk Assessment system based on LSTM for elderly.

To address the challenges posed by the aging process, we designed and validated an LSTM-based automatic remote health risk assessment system for the elderly. This system consists of a wireless physiological parameter sensing unit, a vital sign prediction unit and a pre-defined risk scoring criteria unit. The vital sign prediction module is composed of five 5-input-1-output neural networks based on the LSTM architecture, which are responsible for predicting the vital signs collected by wireless sensors, including: systolic blood pressure (SBP), pulse rate (PR), respiratory rate (RR), temperature (TEMP), and oxygen saturation (SPO2). The pre-defined health risk scoring criteria is a simplified version of the National Early Warning Score (NEWS), which is responsible for calculating the risk level based on the predicted values. This allows the care team to respond to the medical needs of the elderly in a timely manner. Through experiments, our system can achieve a risk identification accuracy of 74% and MAEs of the predicted values for each parameter are in an acceptable range. Our results suggest that an automated remote health risk assessment system for the elderly using deep learning could be a viable new strategy for home-based monitoring systems.

On the Validity of Single Regression Strategy for Granger Causality Assessment in Cardiovascular and Cardiorespiratory Control Studies.

Granger causality (GC) analysis is based on the comparison between prediction error variances computed over the full and restricted models after identifying the coefficients of appropriate vector regressions. GC markers can be computed via a double regression (DR) approach identifying two separate, independent models and a single regression (SR) strategy optimizing the description of the dynamics of the target over the full model and, then, reusing some parts of it in the restricted model. The present study compares the SR and DR strategies over heart period (HP), systolic arterial pressure (SAP) and respiration (R) beat-to-beat series collected during a graded orthostatic challenge induced by head-up tilt in 17 healthy individuals (age: 21-36 yrs; median: 29 yrs; 9 females and 8 males). We found that the DR approach was more powerful than the SR one in detecting the expected stronger involvement of the baroreflex during the challenge, while the expected weaker cardiorespiratory coupling was identified by both SR and DR strategies. The less powerful ability of the SR approach was the result of the greater variance of GC markers compared to the DR strategy. We conclude that, contrary to the suggestions present in literature, the SR approach is not necessarily associated with a smaller dispersion of GC markers. Moreover, we suggest that additional factors, such as the strategy utilized to build embedding spaces and metric utilized to compare prediction error variances, might play an important role in differentiating SR and DR approaches.

SMART System in the Assessment of Exercise Tolerance in Adults.

Health-oriented physical activity should meet two key criteria: safety and an optimal level of exercise. The system of monitoring and rationalization of training (SMART) was designed to meet them. SMART integrates a custom-configured inertial measurement unit (IMU) and a sensor with real-time heart rate measurement (HR) using a proprietary computer application. SMART was used to evaluate the safety and exercise load with 115 study participants: 51 women (44.35%) and 64 men (55.65%) aged 19 to 65 years. The exercise test was the 6MWT test. In 35% of the participants, the mean HR exceeded the recognized safe limit of HR 75% max. Ongoing monitoring of HR allows for optimal exercise and its safety. Step count data were collected from the SMART system. The average step length was calculated by dividing the distance by the number of steps. The aim of the present study was to assess the risk of excessive cardiovascular stress during the 6MWT test using the SMART system.

Physical Noninvasive Attacks on Photoplethysmogram by Computer Controlled Blood Pressure Cuff.

Sensor data has been used in social security and welfare infrastructures such as insurance and medical care to provide personalized products and services; there is a risk that attackers can alter sensor data to obtain unfair benefits. We consider that one of the attack methods to modify sensor data is to attack the wearer's body to modify biometric information. In this study, we propose a noninvasive attack method to modify the sensor value of a photoplethysmogram. The proposed method can disappear pulse wave peaks by pressurizing the upper arm with air pressure to control blood volume. Seven subjects experiencing a rest environment and five subjects experiencing an after-exercise environment wore five different models of smartwatches, and three pressure patterns were performed. It was confirmed in both situations that the displayed heart rate decreased from the true heart rate.