Multi-View Cross-Fusion Transformer Based on Kinetic Features for Non-Invasive Blood Glucose Measurement Using PPG Signal.

Noninvasive blood glucose (BG) measurement could significantly improve the prevention and management of diabetes. In this paper, we present a robust novel paradigm based on analyzing photoplethysmogram (PPG) signals. The method includes signal pre-processing optimization and a multi-view cross-fusion transformer (MvCFT) network for non-invasive BG assessment. Specifically, a multi-size weighted fitting (MSWF) time-domain filtering algorithm is proposed to optimally preserve the most authentic morphological features of the original signals. Meanwhile, the spatial position encoding-based kinetics features are reconstructed and embedded as prior knowledge to discern the implicit physiological patterns. In addition, a cross-view feature fusion (CVFF) module is designed to incorporate pairwise mutual information among different views to adequately capture the potential complementary features in physiological sequences. Finally, the subject- wise 5- fold cross-validation is performed on a clinical dataset of 260 subjects. The root mean square error (RMSE) and mean absolute error (MAE) of BG measurements are 1.129 mmol/L and 0.659 mmol/L, respectively, and the optimal Zone A in the Clark error grid, representing none clinical risk, is 87.89%. The results indicate that the proposed method has great potential for homecare applications.

Video Based Cocktail Causal Container for Blood Pressure Classification and Blood Glucose Prediction.

With the development of modern cameras, more physiological signals can be obtained from portable devices like smartphone. Some hemodynamically based non-invasive video processing applications have been applied for blood pressure classification and blood glucose prediction objectives for unobtrusive physiological monitoring at home. However, this approach is still under development with very few publications. In this paper, we propose an end-to-end framework, entitled cocktail causal container, to fuse multiple physiological representations and to reconstruct the correlation between frequency and temporal information during multi-task learning. Cocktail causal container processes hematologic reflex information to classify blood pressure and blood glucose. Since the learning of discriminative features from video physiological representations is quite challenging, we propose a token feature fusion block to fuse the multi-view fine-grained representations to a union discrete frequency space. A causal net is used to analyze the fused higher-order information, so that the framework can be enforced to disentangle the latent factors into the related endogenous association that corresponds to down-stream fusion information to improve the semantic interpretation. Moreover, a pair-wise temporal frequency map is developed to provide valuable insights into extraction of salient photoplethysmograph (PPG) information from fingertip videos obtained by a standard smartphone camera. Extensive comparisons have been implemented for the validation of cocktail causal container using a Clinical dataset and PPG-BP benchmark. The root mean square error of 1.329±0.167 for blood glucose prediction and precision of 0.89±0.03 for blood pressure classification are achieved in Clinical dataset.

Timed Up and Go test phases as predictors of future falls in community-dwelling older adults / Fases do teste Timed Up and Go como preditoras de quedas futuras em idosos da comunidade

Abstract Introduction The Timed Up and Go (TUG) is a test widely used to assess the risk of falls in older adults. Although it is a complex task, only the total TUG time has been used for evaluation. The widespread use of smartphones has provided the development of applications for monitoring diagnostic procedures. Objective To analyze the ability to predict future falls in older adults. Methods A cohort study (1 year) of 42 participants using the sTUG Doctor. Fall events during 1-year follow-up were monitored by telephone. The number of days between assessment and first fall or last contact was calculated for survival analysis, assessed by unadjusted and adjusted Cox proportional hazards regression models. Tests with p <5 % were considered statistically significant and between 5% and 10% were indicative of significance (Epi-Info™ 7.2). Results Falls were observed in 22 (52.38%) participants (fallers). The results indicated that cognitive impairment, depressive symptoms, women, and participants with fear of falling (FES-I) were more likely to fall. Fallers performed worse on all sTUG Doctor phases. Hazard ratios for predicting falls were significant for total TUG time (1.35; p = 0.029) and total number of steps (1.52; p = 0.057). Total TUG time remained significant when adjusted for sex, age group, FES-I, and depression level. Conclusion The sTUG Doctor was an important tool to predict falls in community-dwelling older adults.

Resumo Introdução O Timed Up and Go (TUG) é um teste bastante utilizado para avaliar o risco de quedas em idosos. Embora seja uma tarefa complexa, apenas o tempo total do TUG (TTUGT) tem sido utilizado para avaliação. A propagação dos smartphones proporcionou o surgimento de aplicativos para monitoramento de procedimentos diagnósticos. Objetivo Analisar a capacidade de predição de quedas futuras em idosos através das fases do TUG utilizando o teste sTUG Doctor. Métodos Estudo de coorte (1 ano) com 42 participantes utilizando o sTUG Doctor. O evento de queda durante um ano de seguimento foi monitorado por meio de contato telefônico. O número de dias entre a avaliação e a primeira queda ou último contato foi calculado para a análise de sobrevida avaliada por modelos não ajustados e ajustados através de modelos de regressão de risco de Cox. Testes com p < 5% foram considerados estatisticamente significantes e entre 5 e 10% indicati-vos de significância (Epi-InfoTM 7.2). Resultados As quedas foram observadas em 22 (52,38%) participantes (caidores). Os resultados indicaram que nível cognitivo, sintomas depressivos, mulheres e participantes com medo de cair (FES-I) são mais propensos a cair. Caidores tiveram pior desempenho em todas as fases do sTUG Doctor. As razões de chance para previsão de queda foram significativas para TTUGT (1,35; p = 0,029) e número total de passos (1,52; p = 0,057). O TTUGT permaneceu significativo ajustando-se ao sexo, faixa etária, FES-I e nível depressivo. Conclusão O sTUG Doctor foi uma ferramenta importante para prever quedas em idosos da comunidade.

Can mHealth Technology Help Mitigate the Effects of the COVID-19 Pandemic?

Goal: The aim of the study herein reported was to review mobile health (mHealth) technologies and explore their use to monitor and mitigate the effects of the COVID-19 pandemic. Methods: A Task Force was assembled by recruiting individuals with expertise in electronic Patient-Reported Outcomes (ePRO), wearable sensors, and digital contact tracing technologies. Its members collected and discussed available information and summarized it in a series of reports. Results: The Task Force identified technologies that could be deployed in response to the COVID-19 pandemic and would likely be suitable for future pandemics. Criteria for their evaluation were agreed upon and applied to these systems. Conclusions: mHealth technologies are viable options to monitor COVID-19 patients and be used to predict symptom escalation for earlier intervention. These technologies could also be utilized to monitor individuals who are presumed non-infected and enable prediction of exposure to SARS-CoV-2, thus facilitating the prioritization of diagnostic testing.

Exploring the uptake and implementation of tele-monitored home-exercise programmes in adults with Parkinson's disease: A mixed-methods pilot study.

BACKGROUND: People with Parkinson's disease experience numerous barriers to exercise participation at fitness facilities. Advances in tele-monitoring technologies create alternative channels for managing and supervising exercise programmes in the home. However, the success of these programmes will depend on participants' perceptions of using the technology and their exercise adherence. Thus, this pilot explored the uptake and implementation of two common methods of Internet-exercise training in Parkinson's disease. METHODS: Twenty adults with Parkinson's disease were randomized into either: telecoach-assisted exercise (TAE) or self-regulated exercise (SRE) groups. Both groups received the same eight-week exercise prescription (combined strength and aerobic exercise) and telehealth system that streamed and recorded vital signs and exercise data. TAE participants exercised under a telecoach's supervision via videoconferencing. SRE participants independently managed their exercise training. Quantitative data were described and qualitative data underwent thematic analysis. RESULTS: Quantitative results demonstrated that TAE participants achieved strong attendance (99.2%), whereas SRE participants demonstrated 35.9% lower attendance, 48% less total time exercising, and 74.5% less time exercising at moderate intensity. Qualitatively, TAE participants reported overtly favourable programme experiences and that assistance from a telecoach enhanced their exercise motivation. SRE participants noted several challenges that impeded adherence. CONCLUSION: Findings demonstrate that adults with Parkinson's disease acknowledge benefits of exercising through a telehealth system and are open to utilizing this channel as a means of exercise. However, human-interactive support may be required to overcome unique impediments to participation. Study findings warrant validation in larger trials that can transfer the success of TAE towards more scalable methods of delivery.

Wearables Meet IoT: Synergistic Personal Area Networks (SPANs).

Wearable monitoring and mobile health (mHealth) revolutionized healthcare diagnostics and delivery, while the exponential increase of deployed "things" in the Internet of things (IoT) transforms our homes and industries. "Things" with embedded activity and vital sign sensors that we refer to as "smart stuff" can interact with wearable and ambient sensors. A dynamic, ad-hoc personal area network can span multiple domains and facilitate processing in synergistic personal area networks-SPANs. The synergy of information from multiple sensors can provide: (a) New information that cannot be generated from existing data alone, (b) user identification, (c) more robust assessment of physiological signals, and (d) automatic annotation of events/records. In this paper, we present possible new applications of SPANs and results of feasibility studies. Preliminary tests indicate that users interact with smart stuff-in our case, a smart water bottle-dozens of times a day and sufficiently long to collect vital signs of the users. Synergistic processing of sensors from the smartwatch and objects of everyday use may provide user identification and assessment of new parameters that individual sensors could not generate, such as pulse wave velocity (PWV) and blood pressure. As a result, SPANs facilitate seamless monitoring and annotation of vital signs dozens of times per day, every day, every time the smart object is used, without additional setup of sensors and initiation of measurements. SPANs creates a dynamic "opportunistic bubble" for ad-hoc integration with other sensors of interest around the user, wherever they go. Continuous long-term monitoring of user's activity and vital signs can provide better diagnostic procedures and personalized feedback to motivate a proactive approach to health and wellbeing.

Development of an Automated 30 Second Chair Stand Test Using Smartwatch Application.

This paper presents development of the smartwatch application for automation of the standard 30 Second Chair Stand Test (30SCST). 30SCST is primarily used to test leg strength and endurance, but also speed and mobility and assess risk of falls. We use inertial signals on smartwatch to detect and count stands during the test. The application notifies the user to start and stop the test using vibration on the smartwatch. Synchronization of notifications and signal acquisition allows assessment of user's response time during the test. Our application monitors baseline heart rate before the test, heart rate increase during the test, and heart rate recovery after the test that might allow assessment of cardiovascular fitness of the user. The application is developed using Wear OS and tested on two smartwatch platforms: Fossil G4 and Polar M600. Pilot test included 12 subjects, six male and six female (mean age 39.1, S.D. 19 years). Overall accuracy of detection of the number of standups is 98.8%. Smartwatch application can be used for automated testing in clinical setups as well as for self-monitoring at home.

A New Method to Prevent Unintentional Child Poisoning.

Unintentional child poisoning represents an increasingly important health issue in the United States and worldwide, partially due to increased use of drugs and food supplements. Biometric authentication is complex for pill bottles, but we propose a new method of user identification using touch capacitance during bottle-opening attempts. A smart pill bottle could generate an immediate warning to deter a child from opening the bottle and send an alert to parents/guardians. In this paper, we present principle of operation and implementation of a prototype "safe bottle We present the results of pilot testing with 5 adults and 3 children using support vector machine (SVM) and neural network (NN). From 232 bottle-opening events, our optimized NN generated no false detections of children as adults and four false detections of adults as children. Preliminary results indicate that smart pill bottles can be used to reliably detect children trying to open pill bottles and reduce risk of child-poisoning events.

Vital Sign Monitoring Using Capacitive Sensing.

Unobtrusive vital sign monitoring is essential for most wearable monitoring applications. This paper presents a new method for unobtrusive vital sign monitoring based on the change in capacitance of passive sensors in contact with the user's body. Unlike traditional capacitive coupling that is commonly used for elimination of the DC component of bioamplifiers, we propose monitoring of heart rate using galvanically isolated capacitive sensors. This paper presents the theory of operation and the implementation of a prototype device for heart rate monitoring. A prototype capacitive sensor was implemented using parallel coplanar plates in contact with the tip of the finger. Preliminary testing allowed us to observe changes in capacitance in the range of 0.1-0.3 pF per heartbeat, or 0.3-0.9% of the measured signal. The detected heart rate was validated using a PPG sensor positioned on the same hand. During a single recording session of 772 s, the detected heart rate was within 3 BPM 42% of the time. Proposed method is sensitive to motion artefacts; however, processing of signals collected from multiple contact points may facilitate removal of motion artifacts. Since many microcontrollers directly support high quality monitoring of capacitance on multiple I/O pins, the proposed method provides a promising solution for a low-power, low-cost, multi-contact monitoring.

A model based analysis of optimality of sit-to-stand transition.

Objective assessment of mobility and effectiveness of interventions remains an open issue. Timed Up and Go (TUG) and 30 Second Chair Stand (30SCS) tests are routinely used in assessing mobility of subjects, but they provide a single parameter. Instrumenting subjects with wearable sensors enables a detailed mobility assessment. Specifically, we argue that instrumented sit-to-stand (S2ST) posture transitions during the TUG and 30SCS tests can be used to assess the strength and balance of subjects. In this paper we develop a personalized three-segment model that quantifies torques/forces on the body and assesses optimality of each sit-to-stand transition. To characterize a S2ST transition we calculate action defined as an integral of mechanical energy over time. The theoretical optimal transition time can thus be determined for each person by finding the minimum action necessary for a S2ST transition. Our model assesses action during the S2ST transition using inputs from smartphone's inertial sensors, and calculates optimum S2ST transition time for a given body composition of a subject. Our experimental evaluation shows that healthy young subjects have posture transition times close to the optimal transition time generated by the model. We hypothesize that the optimality of posture transition provides an objective and potentially more accurate estimation of the mobility. We tested the model by evaluating optimum action and optimum S2ST transition time for 10 geriatric patients undergoing a mobility improvement program by comparing their performance with the optimum performance generated by the model. This paper presents the model and possible use of the results to assess long-term changes in mobility of users.

A Smartphone Application Suite for Assessing Mobility.

Modern smartphones integrate a growing number of inertial and environmental sensors that can enable the development of new mobile health applications. In this paper we introduce a suite of smartphone applications for assessing mobility in elderly population. The suite currently includes applications that automate and quantify the following standardized medical tests for assessing mobility: Timed-Up-and-Go (TUG), 30 Seconds Chair Stand Test (30SCS), and a 4-stage Balance Test (4SBT). For each smartphone application we describe its functionality and a list of parameters extracted by processing signals from smartphone's inertial sensors. The paper shows the results from studies conducted on geriatric patients for TUG tests and from studies conducted in the laboratory on healthy subjects for 30SCS and 4SBT tests.

SmartStuff: A case study of a smart water bottle.

The rapid growth of Internet of Things (IoT) and miniature wearable biosensors have generated new opportunities for personalized eHealth and mHealth services. Smart objects equipped with physiological sensors can provide robust monitoring of activities of daily living and context for wearable physiological sensors. We present a case study of an intelligent water bottle that can precisely measure the amount of liquid in the bottle, monitor activity using inertial sensors, and physiological parameters using a touch and photoplethysmographic sensor. We evaluate two system configurations: a smart water bottle integrated into a personal body sensor network and a cloud based device. This paper presents system organization and the results from preliminary field testing of the prototype device.

Teleexercise for Persons With Spinal Cord Injury: A Mixed-Methods Feasibility Case Series.

BACKGROUND: Spinal cord injury (SCI) results in significant loss of function below the level of injury, often leading to restricted participation in community exercise programs. To overcome commonly experienced barriers to these programs, innovations in technology hold promise for remotely delivering safe and effective bouts of exercise in the home. OBJECTIVE: To test the feasibility of a remotely delivered home exercise program for individuals with SCI as determined by (1) implementation of the intervention in the home; (2) exploration of the potential intervention effects on aerobic fitness, physical activity behavior, and subjective well-being; and (3) acceptability of the program through participant self-report. METHODS: Four adults with SCI (mean age 43.5 [SD 5.3] years; 3 males, 1 female; postinjury 25.8 [SD 4.3] years) completed a mixed-methods sequential design with two phases: an 8-week intervention followed by a 3-week nonintervention period. The intervention was a remotely delivered aerobic exercise training program (30-45 minutes, 3 times per week). Instrumentation included an upper body ergometer, tablet, physiological monitor, and custom application that delivered video feed to a remote trainer and monitored and recorded exercise data in real time. Implementation outcomes included adherence, rescheduled sessions, minutes of moderate exercise, and successful recording of exercise data. Pre/post-outcomes included aerobic capacity (VO2 peak), the Physical Activity Scale for Individuals with Physical Disabilities (PASIPD), the Satisfaction with Life Scale (SWLS), and the Quality of Life Index modified for spinal cord injury (QLI-SCI). Acceptability was determined by participant perceptions of the program features and impact, assessed via qualitative interview at the end of the nonintervention phase. RESULTS: Participants completed all 24 intervention sessions with 100% adherence. Out of 96 scheduled training sessions for the four participants, only 8 (8%) were makeup sessions. The teleexercise system successfully recorded 85% of all exercise data. The exercise program was well tolerated by all participants. All participants described positive outcomes as a result of the intervention and stated that teleexercise circumvented commonly reported barriers to exercise participation. There were no reported adverse events and no dropouts. CONCLUSION: A teleexercise system can be a safe and feasible option to deliver home-based exercise for persons with SCI. Participants responded favorably to the intervention and valued teleexercise for its ability to overcome common barriers to exercise. Study results are promising but warrant further investigation in a larger sample.

Periodic leg movement (PLM) monitoring using a distributed body sensor network.

Wireless sensors networks represent the architecture of choice for distributed monitoring due to the ease of deployment and configuration. We developed a distributed sleep monitoring system which combines wireless inertial sensors SP-10C by Sensoplex controlled by a custom smartphone application as an extension of the polysomnographic (PSG) monitor SOMNOscreen plus from Somnomedics. While existing activity monitors are wired to the SOMNOscreen, our system allows the use of wireless inertial sensors to improve user's comfort during sleep. The system is intended for monitoring of periodic leg movements (PLM) and user's activity during sleep. Wireless sensors are placed on ankle and toes of the foot in a customized sock. An Android app communicates with wireless sensors over Bluetooth Smart (BTS) link and streams 3D accelerometer values, 4D unit quaternion values and timestamps. In this paper we present a novel method of synchronization of data streams from PSG and inertial sensors, and original method of detection of PLM events. The system was tested using five experiments of simulated PLM, and achieved 96.51% of PLM detection accuracy.

Preliminary analysis of the use of smartwatches for longitudinal health monitoring.

New generations of smartwatches feature continuous measurement of physiological parameters, such as heart rate, galvanic skin resistance (GSR), and temperature. In this paper we present the results of preliminary analysis of the use of Basis Peak smartwatch for longitudinal health monitoring during a 4 month period. Physiological measurements during sleep are validated using Zephyr Bioharness 3 monitor and SOMNOscreen+ polysomnographic monitoring system from SOMNOmedics. Average duration of sequences with no missed data was 49.9 minutes, with maximum length of 17 hours, and they represent 88.88% of recording time. Average duration of the charging event was 221.9 min, and average time between charges was 54 hours, with maximum duration of the charging event of 16.3 hours. Preliminary results indicate that the physiological monitoring performance of existing smartwatches provides sufficient performance for longitudinal monitoring of health status and analysis of health and wellness trends.

A distributed scheme to manage the dynamic coexistence of IEEE 802.15.4-based health-monitoring WBANs.

The overlap of transmission ranges between wireless networks as a result of mobility is referred to as dynamic coexistence. The interference caused by coexistence may significantly affect the performance of wireless body area networks (WBANs) where reliability is particularly critical for health monitoring applications. In this paper, we analytically study the effects of dynamic coexistence on the operation of IEEE 802.15.4-based health monitoring WBANs. The current IEEE 802.15.4 standard lacks mechanisms for effectively managing the coexistence of mobile WBANs. Considering the specific characteristics and requirements of health monitoring WBANs, we propose the dynamic coexistence management (DCM) mechanism to make IEEE 802.15.4-based WBANs able to detect and mitigate the harmful effects of coexistence. We assess the effectiveness of this scheme using extensive OPNET simulations. Our results indicate that DCM improves the successful transmission rates of dynamically coexisting WBANs by 20%-25% for typical medical monitoring applications.

Sensors and systems for obesity care and research.

Obesity and overweight decrease the quality and length of life and increases healthcare costs. Dramatic increase of obesity, particularly childhood obesity, in developed countries requires efficient methods and systems for obesity research and monitoring of users. New generation of smart sensors and ubiquitous monitoring systems provide unprecedented opportunities to assess real-life environments, mobility, physical activity, and physiological responses. In this paper we present current trends in sensing, social networks, and systems integration. New monitoring and intervention technologies create new opportunities for remote patient monitoring and intervention.

Research methodology for real-time stress assessment of nurses.

This article presents a research methodology for analysis of stress effects and allostatic load of nurses during daily activities. Stress-related health issues are critical in healthcare workers, in particular nurses. Typical causes of stress include inadequate staffing of nurses for the number and acuity of patients, dealing with difficult patients and families, and lack of autonomy in care delivery decisions. This is all compounded by lack of recovery time while on shift, variable shifts with limited recovery time between days worked, and fatigue from dealing with difficult patients, families, and healthcare workers. Under unresolved stress, the heart rate and other vital parameters may fail to return to the baseline. This study examined the physiological responses of nurses during care on a high-fidelity patient simulation to develop a research methodology and identify physiological parameters suitable for real-time assessment of allostatic load during work. Our results demonstrated that heart rate and heart rate variability can be reliably measured using wearable sensors to assess allostatic load. During this study and our previous related work, we acquired valuable experience regarding selection and deployment of commercially available sensors, system integration, recruitment of subjects, and general research methodology. The research methodology developed and presented in this article can be applied to a number of other applications and experimental protocols.

A mobile system for assessment of physiological response to posture transitions.

Posture changes initiate a dynamic physiological response that can be used as an indicator of the overall health status. We introduce an inconspicuous mobile wellness monitoring system (imWell) that continuously assesses the dynamic physiological response to posture transitions during activities of daily living. imWell utilizes a Zephyr BioHarness 3 physiological monitor that continually reports heart activity and physical activity via Bluetooth to a personal device (e.g. smartphone). The personal device processes the reported activity data in real-time to recognize posture transitions from the accelerometer data and to characterize dynamic heart response to posture changes. It annotates, logs, and uploads the heart activity data to our mHealth server. In this paper we present algorithms for detection of posture transitions and heart activity characterization during a sit-to-stand transition. The proposed system was tested on seven healthy subjects performing a predefined protocol. The total average and standard deviation for sit-to-stand transition time is 2.7 ± 0.69 s, resulting in the change of heart rate of 27.36 ± 9.30 bpm (from 63.3 ± 9.02 bpm to 90.66 ± 10.09 bpm).

Performance analysis of coexisting IEEE 802.15.4-based health monitoring WBANs.

Wireless Body Area Networks (WBANs) for health monitoring systems are required to meet stringent performance demands regarding the tradeoff between reliability, latency, and power efficiency. WBANs feature limited range and bandwidth and they are prone to interference. Considering the life-critical nature of some WBAN systems, we present an in-depth investigation of the situations where the dynamic coexistence of multiple WBANs may severely affect their performances. In this paper, we analytically study the effect of coexistence on the operation of WBANs. We present a mathematical analysis to precisely obtain the probabilities of successful communication and validate this analysis through simulation. Our simulation analysis indicates that in the default mode of operation, coexistence of three WBANs can lead to the loss of 20-85% of data transmissions for typical sensor configurations.