Sensor-based Physical Activity Research Articles

Machine Learning-Based Aggression Detection in Children with ADHD Using Sensor-Based Physical Activity Monitoring.

Aggression in children is highly prevalent and can have devastating consequences, yet there is currently no objective method to track its frequency in daily life. This study aims to investigate the use of wearable-sensor-derived physical activity data and machine learning to objectively identify physical-aggressive incidents in children. Participants (n = 39) aged 7 to 16 years, with and without ADHD, wore a waist-worn activity monitor (ActiGraph, GT3X+) for up to one week, three times over 12 months, while demographic, anthropometric, and clinical data were collected. Machine learning techniques, specifically random forest, were used to analyze patterns that identify physical-aggressive incident with 1-min time resolution. A total of 119 aggression episodes, lasting 7.3 ± 13.1 min for a total of 872 1-min epochs including 132 physical aggression epochs, were collected. The model achieved high precision (80.2%), accuracy (82.0%), recall (85.0%), F1 score (82.4%), and area under the curve (89.3%) to distinguish physical aggression epochs. The sensor-derived feature of vector magnitude (faster triaxial acceleration) was the second contributing feature in the model, and significantly distinguished aggression and non-aggression epochs. If validated in larger samples, this model could provide a practical and efficient solution for remotely detecting and managing aggressive incidents in children.

A Multi-Label Based Physical Activity Recognition via Cascade Classifier.

Physical activity recognition is a field that infers human activities used in machine learning techniques through wearable devices and embedded inertial sensors of smartphones. It has gained much research significance and promising prospects in the fields of medical rehabilitation and fitness management. Generally, datasets with different wearable sensors and activity labels are used to train machine learning models, and most research has achieved satisfactory performance for these datasets. However, most of the methods are incapable of recognizing the complex physical activity of free living. To address the issue, we propose a cascade classifier structure for sensor-based physical activity recognition from a multi-dimensional perspective, with two types of labels that work together to represent an exact type of activity. This approach employed the cascade classifier structure based on a multi-label system (Cascade Classifier on Multi-label, CCM). The labels reflecting the activity intensity would be classified first. Then, the data flow is divided into the corresponding activity type classifier according to the output of the pre-layer prediction. The dataset of 110 participants has been collected for the experiment on PA recognition. Compared with the typical machine learning algorithms of Random Forest (RF), Sequential Minimal Optimization (SMO) and K Nearest Neighbors (KNN), the proposed method greatly improves the overall recognition accuracy of ten physical activities. The results show that the RF-CCM classifier has achieved 93.94% higher accuracy than the 87.93% obtained from the non-CCM system, which could obtain better generalization performance. The comparison results reveal that the novel CCM system proposed is more effective and stable in physical activity recognition than the conventional classification methods.

The relationship between self-reported physical frailty and sensor-based physical activity measures in older adults – a multicentric cross-sectional study

BackgroundThe decline in everyday life physical activity reflects and contributes to the frailty syndrome. While especially self-reported frailty assessments have the advantage of reaching large groups at low costs, little is known about the relationship between the self-report and objective measured daily physical activity behavior.The main objective was to evaluate whether and to what extent a self-reported assessment of frailty is associated with daily physical activity patterns.MethodsDaily activity data were obtained from 88 elderly participants (mean 80.6 ± 9.1 years) over up to 21 days. Acceleration data were collected via smartwatch. According to the results of a self-report frailty questionnaire, participants were retrospectively split up into three groups, F (frail, n = 43), P (pre-frail, n = 33), and R (robust, n = 12). Gait- and activity-related measures were derived from the built-in step detector and acceleration sensor and comprised, i.a., standard deviation of 5-s-mean amplitude deviation (MADstd), median MAD (MADmedian), and the 95th percentile of cadence (STEP95). Parameters were fed into a PCA and component scores were used to derive behavioral clusters.ResultsThe PCA suggested two components, one describing gait and one upper limb activity. Mainly gait related parameters showed meaningful associations with the self-reported frailty score (STEP95: R2 = 0.25), while measures of upper limb activity had lower coefficients (MADmedian: R2 = 0.07). Cluster analysis revealed two clusters with low and relatively high activity in both dimensions (cluster 2 and 3). Interestingly, a third cluster (cluster 1) was characterized by high activity and low extent of ambulation. Comparisons between the clusters showed significant differences between activity, gait, age, sex, number of chronic diseases, health status, and walking aid. Particularly, cluster 1 contained a higher number of female participants, whose self-reports tended towards a low health status, the frequent use of a walking aid, and a higher score related to frailty questions.ConclusionsThe results demonstrate that subjective frailty assessments may be a simple first screening approach. However, especially older women using walking aids may classify themselves as frail despite still being active. Therefore, the results of self-reports may be particularly biased in older women.

The Need for Objective Physical Activity Measurements in Routine Bariatric Care

PurposeThis study aims to (1) quantify physical behavior through self-reports and sensor-based measures, (2) examine the correlation between self-reported and sensor-based physical activity (PA) and (3) assess whether bariatric patients adhere to PA guidelines.MethodsA Fitbit accelerometer was used to collect minute-to-minute step count and heart rate data for 14 consecutive days. Total physical activity levels (PAL), moderate-to-vigorous intensity physical activity (MVPA) and sedentary behavior (SB) were used to quantify physical behavior. Self-reported PA was assessed with the International Physical Activity Questionnaire (IPAQ). To analyze the association between sensor-based and self-reported PA, Spearman’s correlation was used. A minimum of 150 MVPA minutes per week was considered as compliance with the PA guidelines.ResultsFitbit data of 37 pre- and 18 post-surgery patients was analyzed. Participants averaged 7403 ± 3243 steps/day and spent most of their time sedentary (832 min, IQR: 749 – 879), especially in prolonged periods of ≥ 30 min (525, IQR: 419 – 641). Median MVPA time was 5.6 min/day (IQR: 1.7 – 10.6). Correlations between self-reported and sensor-based MVPA and SB were respectively 0.072 and 0.455. Only 17.1% was objectively adherent to MVPA guidelines ≥ 150 min/week, while 94.3% met the guidelines in case of self-reports.ConclusionPA quantification confirmed that bariatric patients are highly sedentary and rarely engage in MVPA, despite a relatively high daily step count. Moreover, bariatric patients are not able to assess MVPA and moderately their SB by self-reports. Our results indicate the need for sensor-based PA monitoring in routine bariatric care.Graphical abstract

Physical Activity Behavior of Patients at a Skilled Nursing Facility: Longitudinal Cohort Study.

BackgroundOn-body wearable sensors have been used to predict adverse outcomes such as hospitalizations or fall, thereby enabling clinicians to develop better intervention guidelines and personalized models of care to prevent harmful outcomes. In our previous work, we introduced a generic remote patient monitoring framework (Sensing At-Risk Population) that draws on the classification of human movements using a 3-axial accelerometer and the extraction of indoor localization using Bluetooth low energy beacons, in concert. Using the same framework, this paper addresses the longitudinal analyses of a group of patients in a skilled nursing facility. We try to investigate if the metrics derived from a remote patient monitoring system comprised of physical activity and indoor localization sensors, as well as their association with therapist assessments, provide additional insight into the recovery process of patients receiving rehabilitation.ObjectiveThe aim of this paper is twofold: (1) to observe longitudinal changes of sensor-based physical activity and indoor localization features of patients receiving rehabilitation at a skilled nursing facility and (2) to investigate if the sensor-based longitudinal changes can complement patients’ changes captured by therapist assessments over the course of rehabilitation in the skilled nursing facility.MethodsFrom June 2016 to November 2017, patients were recruited after admission to a subacute rehabilitation center in Los Angeles, CA. Longitudinal cohort study of patients at a skilled nursing facility was followed over the course of 21 days. At the time of discharge from the skilled nursing facility, the patients were either readmitted to the hospital for continued care or discharged to a community setting. A longitudinal study of the physical therapy, occupational therapy, and sensor-based data assessments was performed. A generalized linear mixed model was used to find associations between functional measures with sensor-based features. Occupational therapy and physical therapy assessments were performed at the time of admission and once a week during the skilled nursing facility admission.ResultsOf the 110 individuals in the analytic sample with mean age of 79.4 (SD 5.9) years, 79 (72%) were female and 31 (28%) were male participants. The energy intensity of an individual while in the therapy area was positively associated with transfer activities (β=.22; SE 0.08; P=.02). Sitting energy intensity showed positive association with transfer activities (β=.16; SE 0.07; P=.02). Lying down energy intensity was negatively associated with hygiene activities (β=–.27; SE 0.14; P=.04). The interaction of sitting energy intensity with time (β=–.13; SE 0.06; P=.04) was associated with toileting activities.ConclusionsThis study demonstrates that a combination of indoor localization and physical activity tracking produces a series of features, a subset of which can provide crucial information to the story line of daily and longitudinal activity patterns of patients receiving rehabilitation at a skilled nursing facility. The findings suggest that detecting physical activity changes within locations may offer some insight into better characterizing patients’ progress or decline.

A hierarchy of correlates impacting adults’ sensor-based physical activity and sedentary time

ABSTRACT Investigation into correlates across all levels of the socio-ecological model predictive of objectively measured physical activity has rarely been assessed in adults. While considering a diversity of correlates, we determined which correlates best predict sensor-based moderate-to-vigorous physical activity (MVPA) and sedentary-time (ST) in adults. A Chi-squared Automatic Interaction Detection algorithm was used to hierarchize the correlates associated with high ST (≥66.6thpercentile) and sufficient MVPA (≥150 min/week) in 865 adults. The main correlate predictive of being active was currently partaking in sport/exercise. The following relevant correlates were being male for the exercisers and having trees in the neighbourhood for the non-exercisers. The final correlate to boost male exercisers’ MVPA was having lots of shops in the neighbourhood and not having television in the bedroom for women. The primary correlate for high ST was job activity level, with individuals having highly active jobs being less likely to exhibit high levels of ST; being single, male, and a former athlete also increased the chances of being highly sedentary. To increase adults’ MVPA, promotion of sport participation, neighbourhood landscape planning, shop availability, as well as limiting television in the bedroom must be prioritized. For counteracting ST, increasing workplace activity level is warranted.

Digital Biomarkers of Physical Frailty and Frailty Phenotypes Using Sensor-Based Physical Activity and Machine Learning.

Remote monitoring of physical frailty is important to personalize care for slowing down the frailty process and/or for the healthy recovery of older adults following acute or chronic stressors. Taking the Fried frailty criteria as a reference to determine physical frailty and frailty phenotypes (slowness, weakness, exhaustion, inactivity), this study aimed to explore the benefit of machine learning to determine the least number of digital biomarkers of physical frailty measurable from a pendant sensor during activities of daily living. Two hundred and fifty-nine older adults were classified into robust or pre-frail/frail groups based on the physical frailty assessments by the Fried frailty criteria. All participants wore a pendant sensor at the sternum level for 48 h. Of seventeen sensor-derived features extracted from a pendant sensor, fourteen significant features were used for machine learning based on logistic regression modeling and a recursive feature elimination technique incorporating bootstrapping. The combination of percentage time standing, percentage time walking, walking cadence, and longest walking bout were identified as optimal digital biomarkers of physical frailty and frailty phenotypes. These findings suggest that a combination of sensor-measured exhaustion, inactivity, and speed have potential to screen and monitor people for physical frailty and frailty phenotypes.

Evaluation of a Low-Cost Commercial Actigraph and Its Potential Use in Detecting Cultural Variations in Physical Activity and Sleep

The purpose of the present study was to evaluate the performance of a low-cost commercial smartwatch, the Xiaomi Mi Band (MB), in extracting physical activity and sleep-related measures and show its potential use in addressing questions that require large-scale real-time data and/or intercultural data including low-income countries. We evaluated physical activity and sleep-related measures and discussed the potential application of such devices for large-scale step and sleep data acquisition. To that end, we conducted two separate studies. In Study 1, we evaluated the performance of MB by comparing it to the GT3X (ActiGraph, wGT3X-BT), a scientific actigraph used in research, as well as subjective sleep reports. In Study 2, we distributed the MB across four countries (Austria, Germany, Cuba, and Ukraine) and investigated physical activity and sleep among these countries. The results of Study 1 indicated that MB step counts correlated highly with the scientific GT3X device, but did display biases. In addition, the MB-derived wake-up and total-sleep-times showed high agreement with subjective reports, but partly deviated from GT3X predictions. Study 2 revealed similar MB step counts across countries, but significant later wake-up and bedtimes for Ukraine than the other countries. We hope that our studies will stimulate future large-scale sensor-based physical activity and sleep research studies, including various cultures.

Physical Activity Recognition With Statistical-Deep Fusion Model Using Multiple Sensory Data for Smart Health

Nowadays, enhancing the living standard with smart healthcare via the Internet of Things is one of the most critical goals of smart cities, in which artificial intelligence plays as the core technology. Many smart services, deployed according to wearable sensor-based physical activity recognition, have been able to early detect unhealthy daily behaviors and further medical risks. Numerous approaches have studied shallow handcrafted features coupled with traditional machine learning (ML) techniques, which find it difficult to model real-world activities. In this work, by revealing deep features from deep convolutional neural networks (DCNNs) in fusion with conventional handcrafted features, we learn an intermediate fusion framework of human activity recognition (HAR). According to transforming the raw signal value to pixel intensity value, segmentation data acquired from a multisensor system are encoded to an activity image for deep model learning. Formulated by several novel residual triple convolutional blocks, the proposed DCNN allows extracting multiscale spatiotemporal signal-level and sensor-level correlations simultaneously from the activity image. In the fusion model, the hybrid feature merged from the handcrafted and deep features is learned by a multiclass support vector machine (SVM) classifier. Based on several experiments of performance evaluation, our fusion approach for activity recognition has achieved the accuracy over 96.0% on three public benchmark data sets, including Daily and Sport Activities, Daily Life Activities, and RealWorld. Furthermore, the method outperforms several state-of-the-art HAR approaches and demonstrates the superiority of the proposed intermediate fusion model in multisensor systems.

Predicting decline in clinical gait speed in adults with knee osteoarthritis using baseline physical activity data

Purpose: Accelerated functional mobility decline in people with Knee Osteoarthritis (OA) has been associated with, pain, joint function, and real-life physical activity. Yet the overall impact of physical activity on the functional decline is not fully understood. This study aimed to investigate the real-life physical activity risk factors associated with longitudinal decline in gait speed in people with knee OA using baseline sensor-based physical activity measures.

Sensor-based physical activity, sedentary time, and reported cell phone screen time: A hierarchy of correlates in youth

BackgroundEvidence on correlates relies on subjective metrics and fails to include correlates across all levels of the ecologic model. We determined which correlates best predict sensor-based physical activity (PA), sedentary time (ST), and self-reported cell phone screen time (CST) in a large sample of youth, while considering a multiplicity of correlates. MethodsUsing sensor-based accelerometry, we assessed the PA and ST of 2179 youths. A χ2 automatic interaction detection algorithm was used to hierarchize the correlates associated with too much ST (> 50th percentile), insufficient moderate-to-vigorous PA (MVPA) (<60 min/day), and prolonged CST (≥2 h/day). ResultsAmong youth 10–14 years old, the correlates for being inactive consisted of being a girl, not having sport facilities in the neighborhood, and not perceiving the neighborhood as a safe place, whereas in the youth 15–18 years old, the correlate for being inactive was not performing sports (9.7% chance of being active). The correlates for predicting high ST in the younger group was not performing sports (55.8% chance for high ST), and in the older group, the correlates were not owning a pet, perceiving the neighborhood as safe, and having inactive parents (63.7% chance for high ST). In the younger group, the greatest chances of having high CST were among those who were in the last elementary school years, who were girls, and who did not have friends in the neighborhood (73.1% chance for high CST), whereas in the older group, the greatest chance for having high CST was among those who were girls and had a TV in the bedroom (74.3% chance for high CST). ConclusionTo counteract ST and boost MVPA among youths, a specific focus on girls, the promotion of sport participation and facilities, neighborhood safety, and involvement of family must be prioritized.

Performance Analysis of Boosting Classifiers in Recognizing Activities of Daily Living.

Physical activity is essential for physical and mental health, and its absence is highly associated with severe health conditions and disorders. Therefore, tracking activities of daily living can help promote quality of life. Wearable sensors in this regard can provide a reliable and economical means of tracking such activities, and such sensors are readily available in smartphones and watches. This study is the first of its kind to develop a wearable sensor-based physical activity classification system using a special class of supervised machine learning approaches called boosting algorithms. The study presents the performance analysis of several boosting algorithms (extreme gradient boosting—XGB, light gradient boosting machine—LGBM, gradient boosting—GB, cat boosting—CB and AdaBoost) in a fair and unbiased performance way using uniform dataset, feature set, feature selection method, performance metric and cross-validation techniques. The study utilizes the Smartphone-based dataset of thirty individuals. The results showed that the proposed method could accurately classify the activities of daily living with very high performance (above 90%). These findings suggest the strength of the proposed system in classifying activity of daily living using only the smartphone sensor’s data and can assist in reducing the physical inactivity patterns to promote a healthier lifestyle and wellbeing.

Opportunistic sensing for inferring in-the-wild human contexts based on activity pattern recognition using smart computing

In recent years, with the evolution of internet-of-things and smart sensing technologies, sensor-based physical activity recognition has gained substantial prominence, and numerous research works have been conducted in this regard. However, the accurate recognition of in-the-wild human activities and the associated contexts remains an open research challenge to be addressed. This research work presents a novel activity-aware human context recognition scheme that explicitly learns human activity patterns in diverse behavioral contexts and infers in-the-wild user contexts based on physical activity recognition. In this aspect, five daily living activities, e.g., lying, sitting, standing, walking, and running, are associated with overall fourteen different behavioral contexts, including phone positions. A public domain dataset, i.e., ExtraSensory, is used for evaluating the proposed scheme using a series of machine learning classifiers. Random Forest classifier achieves the best recognition rate of 88.4% and 89.8% in recognizing five physical activities and the associated behavioral contexts, respectively, which demonstrates the efficacy of the proposed method.

Change of Objectively-Measured Physical Activity during Geriatric Rehabilitation.

This prospective study investigated feasibility and sensitivity of sensor-based physical activity (PA) measures to monitor changes in PA during geriatric rehabilitation and its relation to clinical parameters at admission. PA was routinely measured at day 2 and day 15 after admission in 647 patients (70.2% women, mean age = 82.0 (SD = 7.19) years) of a German geriatric hospital using a thigh-worn accelerometer. Clinical records were used to include age, Barthel Index, diagnosis, mobility, orientation and cognition. Mean values and 95% confidence intervals (95%-CI) of walking duration, walking bout duration and number of sit-to-stand transfers were calculated to quantify different domains of PA. All observed PA parameters improved during rehabilitation, regardless of age, diagnosis or physical and cognitive function at admission. Walking duration increased by 12.1 (95%-CI: 10.3; 13.8) min, walking bout duration by 2.39 (95%-CI: 1.77; 3.00) s, and number of sit-to-stand transfers by 7 (95%-CI: 5; 8). Floor and ceiling effects were not observed. Walking duration at day 2 as well as day 15 was continuously associated with Barthel Index and statistically significant improved for all levels of Barthel Index. In summary, this study showed that sensor-based PA monitoring is feasible to assess the individual progress in geriatric rehabilitation patients.

Day-to-day variability of multiple sensor-based physical activity parameters in older persons with dementia

ObjectiveTo investigate the day-to-day variability of various sensor-based physical activity (PA) parameters and to analyze effects of weekdays vs. weekend days as well as the influence of concomitant factors (gender, living conditions, cognitive status, physical performance, and level of PA) in multi-morbid, older persons with mild-to-moderate stage dementia. MethodsIn 53 participants, PA was recorded on three consecutive days. Day-to-day variability was estimated by calculating intraclass correlation coefficients (ICCs) for two consecutive days each (Friday-Saturday, Saturday-Sunday). ResultsAlmost all non-walking parameters (lying time, sitting time, standing time, active time, inactive time, and sit-to-stand transitions) showed a consistently low day-to-day variability for Friday-Saturday as well as Saturday-Sunday (ICCs: .60–.96) and hence remained almost unaffected by specific days of the week and concomitant factors. Only the sub-analysis by level of PA revealed slightly deviating results (ICCs: .38–.89). The walking parameters (walking time, walking episodes, and steps) revealed a higher day-to-day variability for Friday-Saturday (ICCs: .01–.40) and a generally lower variability for Saturday-Sunday (ICCs: −.08 − .88), also depending on the respective concomitant factors. ConclusionsTwo consecutive days are adequate to reliably assess non-walking parameters, whereas walking parameters showed higher day-to-day variability with a relevant influence of type of days and concomitant factors.

A Combination of Indoor Localization and Wearable Sensor-Based Physical Activity Recognition to Assess Older Patients Undergoing Subacute Rehabilitation: Baseline Study Results.

BackgroundHealth care, in recent years, has made great leaps in integrating wireless technology into traditional models of care. The availability of ubiquitous devices such as wearable sensors has enabled researchers to collect voluminous datasets and harness them in a wide range of health care topics. One of the goals of using on-body wearable sensors has been to study and analyze human activity and functional patterns, thereby predicting harmful outcomes such as falls. It can also be used to track precise individual movements to form personalized behavioral patterns, to standardize the concept of frailty, well-being/independence, etc. Most wearable devices such as activity trackers and smartwatches are equipped with low-cost embedded sensors that can provide users with health statistics. In addition to wearable devices, Bluetooth low-energy sensors known as BLE beacons have gained traction among researchers in ambient intelligence domain. The low cost and durability of newer versions have made BLE beacons feasible gadgets to yield indoor localization data, an adjunct feature in human activity recognition. In the studies by Moatamed et al and the patent application by Ramezani et al, we introduced a generic framework (Sensing At-Risk Population) that draws on the classification of human movements using a 3-axial accelerometer and extracting indoor localization using BLE beacons, in concert.ObjectiveThe study aimed to examine the ability of combination of physical activity and indoor location features, extracted at baseline, on a cohort of 154 rehabilitation-dwelling patients to discriminate between subacute care patients who are re-admitted to the hospital versus the patients who are able to stay in a community setting.MethodsWe analyzed physical activity sensor features to assess activity time and intensity. We also analyzed activities with regard to indoor localization. Chi-square and Kruskal-Wallis tests were used to compare demographic variables and sensor feature variables in outcome groups. Random forests were used to build predictive models based on the most significant features.ResultsStanding time percentage (P<.001, d=1.51), laying down time percentage (P<.001, d=1.35), resident room energy intensity (P<.001, d=1.25), resident bed energy intensity (P<.001, d=1.23), and energy percentage of active state (P=.001, d=1.24) are the 5 most statistically significant features in distinguishing outcome groups at baseline. The energy intensity of the resident room (P<.001, d=1.25) was achieved by capturing indoor localization information. Random forests revealed that the energy intensity of the resident room, as a standalone attribute, is the most sensitive parameter in the identification of outcome groups (area under the curve=0.84).ConclusionsThis study demonstrates that a combination of indoor localization and physical activity tracking produces a series of features at baseline, a subset of which can better distinguish between at-risk patients that can gain independence versus the patients that are rehospitalized.

Fall Risk in Relation to Individual Physical Activity Exposure in Patients with Different Neurodegenerative Diseases: a Pilot Study.

Falls in patients with neurodegenerative diseases (NDDs) have enormous detrimental consequences. A better understanding of the interplay between physical activity (PA) and fall risk might help to reduce fall frequency. We aimed to investigate the association between sensor-based PA and fall risk in NDDs, using "falls per individual PA exposure time" as a novel measure. Eighty-eight subjects (n = 31 degenerative ataxia (DA), n = 14 Parkinson's disease (PD), n = 12 progressive supranuclear palsy (PSP) and 31 healthy controls) were included in this pilot study. PA was recorded in free-living environments with three-axial accelerometers (activPAL™) over 7days. Falls were prospectively assessed over 12months. Fall incidence was calculated by (i) absolute number of falls per person years (py) and (ii) falls per exposure to individual PA. Absolute fall incidence was high in all three NDDs, with differing levels (DA, 9 falls/py; PD, 14 falls/py; PSP, 29 falls/py). Providing a more fine-grained view on fall risk, correction for individual exposure to PA revealed that measures of low walking PA were associated with higher fall incidence in all three NDDs. Additionally, higher fall incidence was associated with more sit-to-stand transfers in PD and longer walking bouts in PSP. Our results suggest that low walking PA is a risk factor for falls in DA, PD and PSP, indicating the potential benefit of increasing individual PA in these NDDs to reduce fall risk. Moreover, they show that correction for individual exposure to PA yields a more differentiated view on fall risk within and across NDDs.

Examining sensor-based physical activity recognition and monitoring for healthcare using Internet of Things: A systematic review.

Due to importantly beneficial effects on physical and mental health and strong association with many rehabilitation programs, Physical Activity Recognition and Monitoring (PARM) have been considered as a key paradigm for smart healthcare. Traditional methods for PARM focus on controlled environments with the aim of increasing the types of identifiable activity subjects complete and improving recognition accuracy and system robustness by means of novel body-worn sensors or advanced learning algorithms. The emergence of the Internet of Things (IoT) enabling technology is transferring PARM studies to open and connected uncontrolled environments by connecting heterogeneous cost-effective wearable devices and mobile apps. Little is currently known about whether traditional PARM technologies can tackle the new challenges of IoT environments and how to effectively harness and improve these technologies. In an effort to understand the use of IoT technologies in PARM studies, this paper will give a systematic review, critically examining PARM studies from a typical IoT layer-based perspective. It will firstly summarize the state-of-the-art in traditional PARM methodologies as used in the healthcare domain, including sensory, feature extraction and recognition techniques. The paper goes on to identify some new research trends and challenges of PARM studies in the IoT environments, and discusses some key enabling techniques for tackling them. Finally, this paper consider some of the successful case studies in the area and look at the possible future industrial applications of PARM in smart healthcare.

A semiparametric model for wearable sensor-based physical activity monitoring data with informative device wear.

Wearable sensors provide an exceptional opportunity in collecting real-time behavioral data in free living conditions. However, wearable sensor data from observational studies often suffer from information bias, since participants' willingness to wear the monitoring devices may be associated with the underlying behavior of interest. The aim of this study was to introduce a semiparametric statistical approach for modeling wearable sensor-based physical activity monitoring data with informative device wear. Our simulation study indicated that estimates from the generalized estimating equations showed ignorable bias when device wear patterns were independent of the participants physical activity process, but incrementally more biased when the patterns of device non-wear times were increasingly associated with the physical activity process. The estimates from the proposed semiparametric modeling approach were unbiased both when the device wear patterns were (i) independent or (ii) dependent to the underlying physical activity process. We demonstrate an application of this method using data from the 2003-2004 National Health and Nutrition Examination Survey ($N=4518$), to examine gender differences in physical activity measured using accelerometers. The semiparametric model can be implemented using our R package acc, free software developed for reading, processing, simulating, visualizing, and analyzing accelerometer data, publicly available at the Comprehensive R Archive Network.

Associations between sensor-based physical activity behaviour features and health-related parameters

ObjectivesWearable actimetry devices are used increasingly in cohort and cross-sectional studies to assess physical activity (PA) behaviour objectively. Thus far, the medical relevance of distinct PA groups, as identified by using new methods of sensor data analysis, remains unclear. The objective of this research paper is to evaluate whether such PA groups differ in commonly accepted health risk parameters. MethodsPA sensor data and corresponding outcome data of the NHANES 2005–06 study were obtained. Data pre-processing included elimination of potential outliers, data splitting and the computation of PA parameters, including a novel regularity measure. PA groups were identified using the x-Means clustering algorithm, and groups were evaluated for differences in CRP, BMI and HDL. ResultsData sets of 7334 NHANES participants were analysed, and four distinct PA groups were identified. Statistically significant group differences were found for CRP and BMI (p<0.001), but not for HDL (p=0.67). ConclusionsPA groups derived from objective accelerometer mass data differ in exemplary health-related outcome parameters. The novel PA regularity measure is of particular interest and may become part of future PA assessments, especially when regarding low-intensity, short-lived PA events. Further research in pattern recognition methods and analytic algorithms for PA data from current multi-sensing devices is necessary.