Wearable Heart Research Articles

Accuracy of wearable heart rate monitors in cardiac rehabilitation.

To assess the accuracy of four wearable heart rate (HR) monitors in patients with established cardiovascular disease enrolled in phase II or III cardiac rehabilitation (CR). Eighty adult patients enrolled in phase II or III CR were monitored during a CR session that included exercise on a treadmill and/or stationary cycle. Participants underwent HR monitoring with standard ECG limb leads, an electrocardiographic (ECG) chest strap monitor (Polar H7), and two randomly assigned wrist-worn HR monitors (Apple Watch, Fitbit Blaze, Garmin Forerunner 235, TomTom Spark Cardio), one on each wrist. HR was recorded at rest and at 3, 5, and 7 minutes of steady-state exercise on the treadmill and stationary cycle. Across all exercise conditions, the chest strap monitor (Polar H7) had the best agreement with ECG (rc=0.99) followed by the Apple Watch (rc=0.80), Fitbit Blaze (rc=0.78), TomTom Spark (rc=0.76) and Garmin Forerunner (rc=0.52). There was variability in accuracy under different exercise conditions. On the treadmill, only the Fitbit Blaze performed well (rc=0.76), while on the stationary cycle, Apple Watch (rc=0.89) and TomTom Spark (rc=0.85) were most accurate. In cardiac patients, the accuracy of wearable, optically based HR monitors varies, and none of those tested was as accurate as an electrode-containing chest monitor. This observation has implications for in-home CR, as electrode-containing chest monitors should be used when accurate HR measurement is imperative.

A Portable Wearable Tele-ECG Monitoring System

This paper introduces a wearable Tele-ECG and heart rate (HR) monitoring system which has a novel architecture including a stretchable singlet redesigned with textile electrodes (TEs), textile threads, snap fasteners, Velcro, sponges, and an ECG circuit. In addition, a Bluetooth low energy (BLE), a smartphone, a server, and a web page have been added to the system for remote monitoring. The TE can be attached to and removed from the singlet by a Velcro, which allows the user to dry-clean the TE easily for long-term use. A new holter-based ECG system has been designed to evaluate the TE-based ECG system and the average correlation between the recorded ECG signals is obtained as 99.23%. A filtered digital signal, with a high signal-to-noise ratio of 45.62 dB, is transmitted to the smartphone via BLE. The ECG signal is plotted, the HR is calculated with 1.83% mean absolute percentage error, and displayed. The data are sent to the server, allowing the patient's physician to analyze the signals in real time through the web page or the smartphone. If HR reaches beyond the normal range or user presses the “HELP” button on the smartphone screen, the physician is informed automatically by an short message service (SMS) with a location pin on the map. The battery lasts approximately 14 days and when it needs replacement, the system automatically alerts the users by an SMS and a flashing LED. This fast and uninterrupted telemonitoring system has the potential to improve the patient's life quality by providing a psychological reassurance.

Accuracy of Consumer Wearable Heart Rate Measurement During an Ecologically Valid 24-Hour Period: Intraindividual Validation Study.

BackgroundWrist-worn smart watches and fitness monitors (ie, wearables) have become widely adopted by consumers and are gaining increased attention from researchers for their potential contribution to naturalistic digital measurement of health in a scalable, mobile, and unobtrusive way. Various studies have examined the accuracy of these devices in controlled laboratory settings (eg, treadmill and stationary bike); however, no studies have investigated the heart rate accuracy of wearables during a continuous and ecologically valid 24-hour period of actual consumer device use conditions.ObjectiveThe aim of this study was to determine the heart rate accuracy of 2 popular wearable devices, the Apple Watch 3 and Fitbit Charge 2, as compared with the gold standard reference method, an ambulatory electrocardiogram (ECG), during consumer device use conditions in an individual. Data were collected across 5 daily conditions, including sitting, walking, running, activities of daily living (ADL; eg, chores, brushing teeth), and sleeping.MethodsOne participant, (first author; 29-year-old Caucasian male) completed a 24-hour ecologically valid protocol by wearing 2 popular wrist wearable devices (Apple Watch 3 and Fitbit Charge 2). In addition, an ambulatory ECG (Vrije Universiteit Ambulatory Monitoring System) was used as the gold standard reference method, which resulted in the collection of 102,740 individual heartbeats. A single-subject design was used to keep all variables constant except for wearable devices while providing a rapid response design to provide initial assessment of wearable accuracy for allowing the research cycle to keep pace with technological advancements. Accuracy of these devices compared with the gold standard ECG was assessed using mean error, mean absolute error, and mean absolute percent error. These data were supplemented with Bland-Altman analyses and concordance class correlation to assess agreement between devices.ResultsThe Apple Watch 3 and Fitbit Charge 2 were generally highly accurate across the 24-hour condition. Specifically, the Apple Watch 3 had a mean difference of −1.80 beats per minute (bpm), a mean absolute error percent of 5.86%, and a mean agreement of 95% when compared with the ECG across 24 hours. The Fitbit Charge 2 had a mean difference of −3.47 bpm, a mean absolute error of 5.96%, and a mean agreement of 91% when compared with the ECG across 24 hours. These findings varied by condition.ConclusionsThe Apple Watch 3 and the Fitbit Charge 2 provided acceptable heart rate accuracy (<±10%) across the 24 hour and during each activity, except for the Apple Watch 3 during the daily activities condition. Overall, these findings provide preliminary support that these devices appear to be useful for implementing ambulatory measurement of cardiac activity in research studies, especially those where the specific advantages of these methods (eg, scalability, low participant burden) are particularly suited to the population or research question.

Prenatal diagnosis and management of congenital complete heart block.

Congenital complete heart block (CCHB) is a life-threatening medical condition in the unborn fetus with insufficiently validated prenatal interventions. Maternal administration of medications aimed at decreasing the immune response in the fetus and beta-agonists intended to increase fetal cardiac output have shown only marginal benefits. Anti-inflammatory therapies cannot reverse CCHB, but may decrease myocarditis and improve heart function. Advances in prenatal diagnosis and use of strict surveillance protocols for delivery timing have demonstrated small improvements in morbidity and mortality. Ambulatory surveillance programs and wearable fetal heart rate monitors may afford early identification of evolving fetal heart block allowing for emergent treatment. There is also preliminary data suggesting a roll for prevention of CCHB with hydroxychloroquine, but the efficacy and safety is still being studied. To date, intrauterine fetal pacing has not been successful due to the high-risk invasive placement techniques and potential problems with lead dislodgement. The development of a fully implantable micropacemaker via a minimally invasive approach has the potential to pace fetal patients with CCHB and thus delay delivery and allow fetal hydrops to resolve. The challenge remains to establish accepted prenatal interventions capable of successfully managing CCHB in utero until postnatal pacemaker placement is successfully achieved.

Wearable Motion-Based Heart Rate at Rest: A Workplace Evaluation.

This paper studies the feasibility of using low-cost motion sensors to provide opportunistic heart rate assessments from ballistocardiographic signals during restful periods of daily life. Three wearable devices were used to capture peripheral motions at specific body locations (head, wrist, and trouser pocket) of 15 participants during five regular workdays each. Three methods were implemented to extract heart rate from motion data and their performance was compared to those obtained with an FDA-cleared device. With a total of 1358 h of naturalistic sensor data, our results show that providing accurate heart rate estimations from peripheral motion signals is possible during relatively "still" moments. In our real-life workplace study, the head-mounted device yielded the most frequent assessments (22.98% of the time under 5 beats per minute of error) followed by the smartphone in the pocket (5.02%) and the wrist-worn device (3.48%). Most importantly, accurate assessments were automatically detected by using a custom threshold based on the device jerk. Due to the pervasiveness and low cost of wearable motion sensors, this paper demonstrates the feasibility of providing opportunistic large-scale low-cost samples of resting heart rate.

Methodological considerations in calculating heart rate variability based on wearable device heart rate samples

Heart rate variability (HRV) analysis has recently been incorporated into wearable device application. The data source of HRV is the time series of heart beat intervals extracted from electrocardiogram or photoplethysmogram. These intervals are non-uniformly sampled signals and not suitable for spectral HRV analysis, which usually uses uniformly resampled heart beat intervals before calculating the spectral domain parameters. Such a practice is not applicable to heart rate data obtained from wearable devices that usually display and export the beat per minute (BPM) time series data at 1 Hz. The preferred resampling rate to calculate spectral domain parameters is 4 Hz. We compare the spectral HRV results with the 1 Hz and 4 Hz resampling rates in order to validate the use of 1 Hz resampled-RRI data to represent wearable devices BPM time series data for HRV analysis. Our results show that, using a specific combination of signal processing techniques, the lowest mean relative error in spectral domain parameters of normalized low-frequency power (LFnu), normalized high-frequency power (HFnu) and the ratio of normalized low-frequency power to normalized high-frequency power (LFnu/HFnu) between 1 Hz and 4 Hz are 3.7%, 15.3% and 16.4%, respectively. We conclude that using the heart rate data sampled at 1 Hz produces a reasonable estimation of sympathetic activity but a poor estimation of parasympathetic activity.

Human Emotional Care Purposed Automatic Remote Portrait Drawing Generation and Display System Using Wearable Heart Rate Sensor and Smartphone Camera with Depth Perception

We propose a system that automatically generates portrait drawings for the purpose of human emotional care. Our system comprises two parts: a smartphone application and a server. The smartphone application enables the user to take photographs throughout the day while acquiring heart rates from the smartwatch worn by the user. The server collects the photographs and heart rates and displays portrait drawings automatically stylized from the photograph for the most exciting moment of the day. In the system, the user can recall the exciting and happy moment of the day through admiring the drawings and heal the emotion accordingly. To stylize photographs as portrait drawings, we employ nonphotorealistic rendering (NPR) methods, including a portrait etude stylization proposed in this paper. Finally, the effectiveness of our system is demonstrated through user studies.

Influence on postural control and automatic nerve activity using simple virtual reality system

Virtual reality (VR) environments can be used to stimulate movement, motor imagery, and affect. We developed visual stimuli to generate fear under a simple VR environment using a smartphone. This study aimed to evaluate the influence of automatic nerve activity and standing postural sway within this VR environment. Seventeen healthy adults were instructed to stand on a stabilometer with an outward-facing camera mounted to the VR headset. The subjects were evaluated with or without VR stimuli for each of the following conditions: (1) control (no VR), (2) gazing a human in front of oneself on VR, and (3) gazing a ball coming toward oneself on VR. Body sway while standing upright was recorded for 60 seconds with 200 Hz sampling. Heart rate variability was measured by a wearable heart rate sensor, and the low-to-high frequency (LH/HF) ratio was calculated for sympathetic nerve activity during each condition. At the end of each condition, we evaluated “surprises” and “fears” subjectively felt under each condition using a visual analog scale (VAS). We compared the results from each of these conditions with Kruskal–Wallis and Steel–Dwass test. The level of significance was set at P < 0.05. The results revealed that fear-VAS was significantly higher under condition 3 ( P < 0.05). Additionally, the LF/HF tended to higher, and total length of the body sway tended to smaller than other conditions. Using a simple VR system, we observed that fear stimuli induced motor inhibition, such as “freezing” by activating the limbic system.

Smart watches for heart rate assessment in atrial arrhythmias

BackgroundDespite studies demonstrating the accuracy of smart watches (SW) and wearable heart rate (HR) monitors in sinus rhythm, no data exists regarding their utility in arrhythmias. Methods102 hospitalized patients were evaluated at rest using continuous electrocardiogram (ECG) monitoring with concomitant SW-HR (FitBit, FB, Apple Watch, AW) for 30 min. ResultsAcross all devices, 38,616 HR values were recorded. Sinus rhythm cohort demonstrated strong agreement for both devices with a low bias (FB & AW Bias = 1 beat). In atrial arrhythmias, AW demonstrated a stronger correlation than FB (AW rs = 0.83, FB rs = 0.56, both p < 0.01) with a lower bias (Bias AW = −5 beats, FB = −18 beats). Atrial flutter demonstrated strongest agreement in both devices with a mean bias <1 beat. However, in AF, there was significant HR underestimation (Bias FB = −28 beats, AW-8 beats) with wide limits of agreement. Despite HR underestimation in AF, when SW recorded HR ≥ 100 in arrhythmias, 98% of values were within ±10-beats of ECG-HR. ConclusionsSW demonstrate strong agreement for HR estimation in sinus rhythm and atrial flutter but underestimates HR in AF. Tachycardic episodes recorded at rest on a SW may be suggestive of an underlying atrial tachyarrhythmia and warrant further clinical evaluation. Clinical trial registrationAustralian New Zealand Clinical Trials Registry (www.anzctr.org.au) ACTRN: 12616001374459.

Wearable Sensor Application for Integrated Early Warning and Health Surveillance

Objective: The Wearable Sensor Application developed by Pacific Northwest National Laboratory (PNNL) provides an early warning system for stressors to individual and group health using physiologic and environmental indicators. The application integrates health monitoring parameters from wearable sensors, e.g., temperature and heart rate, with relevant environmental parameters, e.g., weather and landscape data, and calculates the corresponding physiological strain index. The information is presented to the analyst in a group and individual view with real-time alerting of abnormal health parameters. This application is the first of its kind being developed for integration into the Defense Threat Reduction Agency's Biosurveillance Ecosystem (BSVE).Introduction: Wearable devices are a low cost, minimally invasive way to monitor health. Sensor data provides real-time physiological indictors of an individual’s health status without the requirement of health care professionals or facilities. Information gleamed from wearable sensors can be used to better understand physiological stressors and prodromal symptoms. In addition, this data can be used to monitor individuals that are in high risk of health-related problems.However, raw data from wearable sensors can be overwhelming to process and laborious to monitor for an individual and, even more so, for a group of individuals. Often specific combination of ranges of sensor readings are indicative of changes to health status and need to be evaluated together or used to calculate specific signal parameters. In addition, the environment surrounding the individual needs to be considered when interpreting the data. To address these issues, PNNL has developed an application that collects, analyzes, and integrates wearable sensor data with geographic landscape and weather information to provide a real-time early alert and situational awareness tool for monitoring the health of groups and individuals.Methods: The prototype application described here was a product of PNNL’s BSVE Application Development Competition. The final product that will be deployed in the BSVE is currently under development by PNNL and will vary slightly in the exact design and architecture described.Data. Wearable sensor data was collected from the Rim2Rim (R2R) Watch Study of individuals hiking the Grand Canyon in Arizona [1]. Weather information was obtained from nearby weather stations and mapping features were derived from Google Maps.Calculations. A physiological Strain Index (PSI) was calculated using core temperature estimates derived through a Kalman Filter approach and heart rate [2,3].Application. The prototype backend application development was based in Python with a MongoDB. The front-end development was built using a scalable architecture and modular approach with components in React and D3.Results: A prototype application was developed this past summer through the PNNL BSVE App Competition (Fig 1). The application was aimed at visualizing wearable sensor data from the Grand Canyon R2R hike dataset. Simulated real-time analysis was used to calculate health status of individuals hiking based on measured physiological parameters and to alert to individuals with signs of physiologic health stress. Visualization tools were incorporated to enable sensor data for individuals and the group to be viewed simultaneously along with pertinent weather, geographic, and elevation data.Many features described in the prototype application will be incorporated into the final BSVE application. The key changes will be 1) the ability to select given time periods for viewing historical data as well as the real-time data collection, 2) environmental data and map view will come from BSVE internal data sources, and 3) the alerts will provide more information and have their own page for reviewing.Conclusions: The Wearable Sensor Application developed by PNNL for integration into the BSVE provides an early warning system for individual and group health using physiologic and environmental parameters. The application highlights health status from wearable sensors and relevant environmental parameters while monitoring a calculated physiological strain index. With this tool, an analyst can easily monitor the health of individuals and groups with the aid of real-time alerting tool for early detection of abnormal health parameters.

DeepHeart: Semi-Supervised Sequence Learning for Cardiovascular Risk Prediction

We train and validate a semi-supervised, multi-task LSTM on 57,675 person-weeks of data from off-the-shelf wearable heart rate sensors, showing high accuracy at detecting multiple medical conditions, including diabetes (0.8451), high cholesterol (0.7441), high blood pressure (0.8086), and sleep apnea (0.8298). We compare two semi-supervised training methods, semi-supervised sequence learning and heuristic pretraining, and show they outperform hand-engineered biomarkers from the medical literature. We believe our work suggests a new approach to patient risk stratification based on cardiovascular risk scores derived from popular wearables such as Fitbit, Apple Watch, or Android Wear.

Wearable Heart Rate Monitor with Emergency Support

Wearable Heart Rate Monitor with Emergency Support

Implementing Wearable Sensors for Continuous Assessment of Daytime Heart Rate Response in Inpatient Rehabilitation.

It is unclear whether wearable heart rate (HR) sensors can be worn continuously in inpatient rehabilitation to assess cardiorespiratory training response. If feasible, these sensors offer a low-cost low-maintenance method for assessing HR response in this setting. We determined feasibility of wearable sensors for assessing HR response to daytime therapy activities in inpatient rehabilitation within a cardiorespiratory training zone equal to 55-80% of maximal HR (target HR [THR]) for at least two 10-min bouts, 3-5 days per week. Secondarily, we determined episodes of excessive HR (EHR >80% of maximal HR). Subjects 44-80 years of age with diagnoses of stroke, cardiac disorders, orthopedic disorders, medically complex conditions, or pulmonary disorders wore wrist-mounted HR sensors day and night throughout inpatient rehabilitation. The proportion of subjects meeting THR thresholds and experiencing EHR episodes was quantified. Multiple regression predicted THR and EHR outcomes from age, sex, length of stay, and motor function at admission and discharge. Across subjects, 97,800 min of HR data were analyzed. Sixty percent of subjects met THR thresholds for cardiorespiratory benefit. Age was the single significant predictor of percent of days meeting the THR threshold (R = 0.58, p = 0.024). Forty-seven percent of subjects experienced EHR episodes on at least 1 day. No subjects experienced sensor-related adverse events, and no protocol deviations occurred from inadvertent sensor removal. Most subjects experienced HR increases sufficient to obtain cardiorespiratory benefit. Likewise, most subjects had episodes of EHR. Wearable sensors were feasible for continuously assessing HR response, suggesting expanded opportunity in inpatient rehabilitation research and treatment.

Acute Effect of Alcohol Intake on Cardiovascular Autonomic Regulation During the First Hours of Sleep in a Large Real-World Sample of Finnish Employees: Observational Study.

BackgroundSleep is fundamental for good health, and poor sleep has been associated with negative health outcomes. Alcohol consumption is a universal health behavior associated with poor sleep. In controlled laboratory studies, alcohol intake has been shown to alter physiology and disturb sleep homeostasis and architecture. The association between acute alcohol intake and physiological changes has not yet been studied in noncontrolled real-world settings.ObjectiveThe aim of this study was to assess the effects of alcohol intake on the autonomic nervous system (ANS) during sleep in a large noncontrolled sample of Finnish employees.MethodsFrom a larger cohort, this study included 4098 subjects (55.81%, 2287/4098 females; mean age 45.1 years) who had continuous beat-to-beat R-R interval recordings of good quality for at least 1 day with and for at least 1 day without alcohol intake. The participants underwent continuous beat-to-beat R-R interval recording during their normal everyday life and self-reported their alcohol intake as doses for each day. Heart rate (HR), HR variability (HRV), and HRV-derived indices of physiological state from the first 3 hours of sleep were used as outcomes. Within-subject analyses were conducted in a repeated measures manner by studying the differences in the outcomes between each participant’s days with and without alcohol intake. For repeated measures two-way analysis of variance, the participants were divided into three groups: low (≤0.25 g/kg), moderate (>0.25-0.75 g/kg), and high (>0.75 g/kg) intake of pure alcohol. Moreover, linear models studied the differences in outcomes with respect to the amount of alcohol intake and the participant’s background parameters (age; gender; body mass index, BMI; physical activity, PA; and baseline sleep HR).ResultsAlcohol intake was dose-dependently associated with increased sympathetic regulation, decreased parasympathetic regulation, and insufficient recovery. In addition to moderate and high alcohol doses, the intraindividual effects of alcohol intake on the ANS regulation were observed also with low alcohol intake (all P<.001). For example, HRV-derived physiological recovery state decreased on average by 9.3, 24.0, and 39.2 percentage units with low, moderate, and high alcohol intake, respectively. The effects of alcohol in suppressing recovery were similar for both genders and for physically active and sedentary subjects but stronger among young than older subjects and for participants with lower baseline sleep HR than with higher baseline sleep HR.ConclusionsAlcohol intake disturbs cardiovascular relaxation during sleep in a dose-dependent manner in both genders. Regular PA or young age do not protect from these effects of alcohol. In health promotion, wearable HR monitoring and HRV-based analysis of recovery might be used to demonstrate the effects of alcohol on sleep on an individual level.

Using Heart Rate and Accelerometry to Define Quantity and Intensity of Physical Activity in Older Adults.

Physical activity (PA) prevents disease and promotes longevity; yet, few older adults meet the recommended daily guidelines. Wearable PA and heart rate monitors provide the opportunity to define age-related differences in the absolute and relative intensity of daily activities, and provide insight into the underlying factors influencing PA in older adults. Participants in the Baltimore Longitudinal Study of Aging (n = 440, 52% male, aged 31 to 88 years) completed a clinical assessment and wore an Actiheart monitor in the free-living environment. The association between age and minutes per day in sedentary, light, moderate, and vigorous PA was assessed using relative intensity, as defined by heart rate reserve, and absolute intensity using activity count thresholds. In cross-sectional analyses, time spent in sedentary and light activities as defined by relative intensity did not differ across age (p > 0.05), whereas time spent in moderate and vigorous relative PA was higher for each 1 year increase in age (p < .01). Using absolute intensity PA thresholds, older adults registered fewer activity counts per day with more sedentary time and lesser amounts of light, moderate, and vigorous PA (p < .05). Persons with higher relative and lower absolute PA intensity had poorer functional performance and higher subclinical disease indicators. These findings suggest that time spent in moderate or higher intensity activities may not be lower with age after considering changes in physiology, functional ability, and subclinical disease burden and highlight the need for more age- and ability-specific PA research to inform future interventions and public health guidelines.

Validation of Heart Rate Monitor Polar RS800 for Heart Rate Variability Analysis During Exercise.

Hernando, D, Garatachea, N, Almeida, R, Casajús, JA, and Bailón, R. Validation of heart rate monitor Polar RS800 for heart rate variability analysis during exercise. J Strength Cond Res 32(3): 716-725, 2018-Heart rate variability (HRV) analysis during exercise is an interesting noninvasive tool to measure the cardiovascular response to the stress of exercise. Wearable heart rate monitors are a comfortable option to measure interbeat (RR) intervals while doing physical activities. It is necessary to evaluate the agreement between HRV parameters derived from the RR series recorded by wearable devices and those derived from an electrocardiogram (ECG) during dynamic exercise of low to high intensity. Twenty-three male volunteers performed an exercise stress test on a cycle ergometer. Subjects wore a Polar RS800 device, whereas ECG was also recorded simultaneously to extract the reference RR intervals. A time-frequency spectral analysis was performed to extract the instantaneous mean heart rate (HRM), and the power of low-frequency (PLF) and high-frequency (PHF) components, the latter centered on the respiratory frequency. Analysis was done in intervals of different exercise intensity based on oxygen consumption. Linear correlation, reliability, and agreement were computed in each interval. The agreement between the RR series obtained from the Polar device and from the ECG is high throughout the whole test although the shorter the RR is, the more differences there are. Both methods are interchangeable when analyzing HRV at rest. At high exercise intensity, HRM and PLF still presented a high correlation (ρ > 0.8) and excellent reliability and agreement indices (above 0.9). However, the PHF measurements from the Polar showed reliability and agreement coefficients around 0.5 or lower when the level of the exercise increases (for levels of O2 above 60%).

Investigation of Noninvasive Wearable Heart Rate Monitor to Support Subjective Requests of Students with developmental disorders

Investigation of Noninvasive Wearable Heart Rate Monitor to Support Subjective Requests of Students with developmental disorders

A Fuzzy Logic-Based Personalized Method to Classify Perceived Exertion in Workplaces Using a Wearable Heart Rate Sensor

Knowing the perceived exertion of workers during their physical activities facilitates the decision-making of supervisors regarding the worker allocation in the appropriate job, actions to prevent accidents, and reassignment of tasks, among others. However, although wearable heart rate sensors represent an effective way to capture perceived exertion, ergonomic methods are generic and they do not consider the diffuse nature of the ranges that classify the efforts. Personalized monitoring is needed to enable a real and efficient classification of perceived individual efforts. In this paper, we propose a heart rate-based personalized method to assess perceived exertion; our method uses fuzzy logic as an option to manage imprecision and uncertainty in involved variables. We applied some experiments to cleaning staff and obtained results that highlight the importance of a custom method to classify perceived exertion of people doing physical work.

An Energy Efficient, Minimally Intrusive Multisensor Intelligent System For Health Monitoring Of Elderly People

abstract Most of the existing systems for elderly health monitoring deploy a large number of cognitive sensors including wearable sensors for physiological parameter measurement. Increasing number of sensors not only make the system power consuming and expensive but also intrusive in nature. However, there exists very limited research on power saving algorithms in such systems incorporating customer friendly features. In this paper, we report a modified health monitoring system which addresses both these issues. The central controller unit has an in-built algorithm based on two level adaptive branch prediction techniques to detect the period of inactivity of sensor nodes. Further, only one wearable heart rate sensor node is included in the system which measures the heart rate and detects abnormality. The central controller signals an alarm to the user to wear this predicting the sleeping time. This makes the system minimally intrusive and user friendly. Thus the multi-sensor network consists of motion sensor, current sensor and a wearable heart rate sensor along with a central controller unit. The prototype of the whole system has been installed in the house of elderly person and it has been observed that the time of prediction was close to the actual time for more than 90% of the days for a test period of one month. An average of 68% power saving has been achieved in the modified system.

A Novel Framework for Motion-Tolerant Instantaneous Heart Rate Estimation by Phase-Domain Multiview Dynamic Time Warping.

long-term wearable instantaneous heart rate (IHR) monitoring is essential to enable pervasive heart health and fitness management. In this paper, a novel framework is proposed to robustly estimate the IHR from electrocardiogram (ECG) signals corrupted by large amounts of daily motion artifacts, which are one of the major impediments against the long-term IHR monitoring. the corrupted ECG signals are first projected to a high-dimensional phase space, where the constructed phase portraits of heartbeats are of many new geometrical properties and are expected to be powerful patterns more immune to the motion artifacts. Afterwards, a multiview dynamic time warping approach is applied on the constructed phase portraits, to effectively capture motion artifacts-induced inconsistencies and reveal heartbeats-related consistencies from corrupted signals. Finally, the phase portraits of heartbeats in the multidimensional phase space can be identified, and then, the IHR estimates are achieved. the proposed framework is evaluated on a wrist-ECG dataset acquired by a semicustomized platform and also a public ECG dataset. With a signal-to-noise ratio as low as -9dB, the mean absolute error and root mean square error of the estimated IHR are 2.5 beats per minute (BPM) and 7.0BPM, respectively. these results demonstrate that our framework can effectively identify the heartbeats from ECG signals continuously corrupted by intense and random motion artifacts and estimate the IHR. the proposed framework greatly outperforms previously reported approaches and is expected to contribute to long-term IHR monitoring. long-term wearable instantaneous heart rate (IHR) monitoring is essential to enable pervasive heart health and fitness management. In this paper, a novel framework is proposed to robustly estimate the IHR from electrocardiogram (ECG) signals corrupted by large amounts of daily motion artifacts, which are one of the major impediments against the long-term IHR monitoring. the corrupted ECG signals are first projected to a high-dimensional phase space, where the constructed phase portraits of heartbeats are of many new geometrical properties and are expected to be powerful patterns more immune to the motion artifacts. Afterwards, a multiview dynamic time warping approach is applied on the constructed phase portraits, to effectively capture motion artifacts-induced inconsistencies and reveal heartbeats-related consistencies from corrupted signals. Finally, the phase portraits of heartbeats in the multidimensional phase space can be identified, and then, the IHR estimates are achieved. the proposed framework is evaluated on a wrist-ECG dataset acquired by a semicustomized platform and also a public ECG dataset. With a signal-to-noise ratio as low as -9dB, the mean absolute error and root mean square error of the estimated IHR are 2.5 beats per minute (BPM) and 7.0BPM, respectively. these results demonstrate that our framework can effectively identify the heartbeats from ECG signals continuously corrupted by intense and random motion artifacts and estimate the IHR. the proposed framework greatly outperforms previously reported approaches and is expected to contribute to long-term IHR monitoring.