Tri-axial Accelerometer Data Research Articles

Multivariate Scalar on Multidimensional Distribution Regression With Application to Modeling the Association Between Physical Activity and Cognitive Functions

ABSTRACTWe develop a new method for multivariate scalar on multidimensional distribution regression. Traditional approaches typically analyze isolated univariate scalar outcomes or consider unidimensional distributional representations as predictors. However, these approaches are suboptimal because (i) they fail to utilize the dependence between the distributional predictors and (ii) neglect the correlation structure of the response. To overcome these limitations, we propose a multivariate distributional analysis framework that harnesses the power of multivariate density functions and multitask learning. We develop a computationally efficient semiparametric estimation method for modeling the effect of the latent joint density on the multivariate response of interest. Additionally, we introduce a new conformal prediction algorithm for quantifying the uncertainty of our multivariate predictions based on subject characteristics and individualized distributional predictors, providing valuable insights into the conditional distribution of the response. We validate the effectiveness of our proposed method through comprehensive numerical simulations, clearly demonstrating its superior performance compared to traditional methods. The application of the proposed method is demonstrated on triaxial accelerometer data from the National Health and Nutrition Examination Survey 2011–2014 for modeling the association between cognitive scores across various domains and distributional representation of physical activity among the older adult population. Our results highlight the advantages of the proposed approach, emphasizing the significance of incorporating multidimensional distributional information in the triaxial accelerometer data.

Using Activity Measures and GNSS Data from a Virtual Fencing System to Assess Habitat Preference and Habitat Utilisation Patterns in Cattle.

There has been an increased focus on new technologies to monitor habitat use and behaviour of cattle to develop a more sustainable livestock grazing system without compromising animal welfare. One of the currently used methods for monitoring cattle behaviour is tri-axial accelerometer data from systems such as virtual fencing technology or bespoke monitoring technology. Collection and transmission of high-frequency accelerometer and GNSS data is a major energy cost, and quickly drains the battery in contemporary virtual fencing systems, making it unsuitable for long-term monitoring. In this paper, we explore the possibility of determining habitat preference and habitat utilisation patterns in cattle using low-frequency activity and location data. We achieve this by (1) calculating habitat selection ratios, (2) determining daily activity patterns, and (3) based on those, inferring grazing and resting sites in a group of cattle wearing virtual fencing collars in a coastal setting with grey, wooded, and decalcified dunes, humid dune slacks, and salt meadows. We found that GNSS data, and a measure of activity, combined with accurate mapping of habitats can be an effective tool in assessing habitat preference. The animals preferred salt meadows over the other habitats, with wooded dunes and humid dune slacks being the least preferred. We were able to identify daily patterns in activity. By comparing general trends in activity levels to the existing literature, and using a Gaussian mixture model, it was possible to infer resting and grazing behaviour in the different habitats. According to our inference of behaviour the herd predominantly used the salt meadows for resting and ruminating. The approach used in this study allowed us to use GNSS location data and activity data and combine it with accurate habitat mapping to assess habitat preference and habitat utilisation patterns, which can be an important tool for guiding management decisions.

Robust and optimal alignment of high-dimensional data using maximum likelihood estimation through a random sample consensus framework

AbstractCorrecting spatial orientations of groups of high-dimensional data sets such that they are all in a consistent coordinate system is often a time-consuming and error-prone process. Automation of this process can be accomplished by using Generalized Procrustes Analysis to estimate the relative orientations among a population of high-dimensional data sets. A least squares Procrustes solution is applied through a maximum likelihood estimation and random sample consensus framework for robustness. The likelihood model is comprised of a mixture distribution where inliers are modeled using t-distribution and outliers from a uniform distribution. Applications will focus on a synthetic data set that emulates triaxial acceleration data and also real shock data from a population of triaxial accelerometers. Outliers represent either non-rigid body responses, environmental noise, and/or sensor and data acquisition issues. The intended application for the methodology is to robustly automate the rotation of populations of experimentally collected triaxial accelerometer data sets to a single global coordinate system.

Comparability of 24-hr Activity Cycle Outputs From ActiGraph Counts Generated in ActiLife and RStudio

Data from ActiGraph accelerometers have long been imported into ActiLife software, where the company’s proprietary “activity counts” were generated in order to understand physical behavior metrics. In 2022, ActiGraph released an open-source method to generate activity counts from any raw, triaxial accelerometer data using Python, which has been translated into RStudio packages. However, it is unclear if outcomes are comparable when generated in ActiLife and RStudio. Therefore, the authors’ technical note systematically compared activity counts and related physical behavior metrics generated from ActiGraph accelerometer data using ActiLife or available packages in RStudio and provides example code to ease implementation of such analyses in RStudio. In addition to comparing triaxial activity counts, physical behavior outputs (sleep, sedentary behavior, light-intensity physical activity, and moderate- to vigorous-intensity physical activity) were compared using multiple nonwear algorithms, epochs, cut points, sleep scoring algorithms, and accelerometer placement sites. Activity counts and physical behavior outcomes were largely the same between ActiLife and the tested packages in RStudio. However, peculiarities in the application of nonwear algorithms to the first and last portions of a data file (that occurred on partial, first or last days of data collection), differences in rounding, and handling of counts values on the borderline of activity intensities resulted in small but inconsequential differences in some files. The hope is that researchers and both hardware and software manufacturers continue to push efforts toward transparency in data analysis and interpretation, which will enhance comparability across devices and studies and help to advance fields examining links between physical behavior and health.

PSIV-23 Activity Classification of Grazing Cows on Desert Rangeland in the Southwestern United States

Abstract Monitoring cattle on rangelands is a daunting task that can be improved by using wearable sensors that are capable of transmitting motion and position data in real time and at low cost. This study tested the performance of machine learning (ML) classifiers to discriminate among foraging activities of cows based on triaxial accelerometer data collected in real-time by LoRa WAN networks. Trials were conducted at the New Mexico State University Chihuahuan Desert Rangeland Research Center and the USDA Jornada Experimental Range in Doña Ana County, NM. A total of 24 Brangus, Brahman, Raramuri Criollo and Angus x Hereford mature cows fitted with LoRa WAN tracking collars housing GPS and triaxial accelerometers were monitored across four periods during the 2022 summer and fall seasons on desert rangeland pastures. Trackers integrated and transmitted activity count (Ac) data from accelerometers at one-minute intervals. Video recording of focal cows (n = 24) was undertaken during daylight hours (0630 to 2000 h) from a distance of ~30 m to minimize interference with natural behaviors. A total of 168 hours of video were recorded and inspected by an experienced observer to label video files according to a classification tree of four main activities: grazing (GR), walking (WA), resting (RE) and ruminating (RU), and two states: active (AC) or static (ST). Individualized activities and states were considered when cows performed the same predefined activity or state for more than 30 secs. Retrieved sensor data from collar trackers were labeled by state and activity according to labels collected from video records. This classification resulted in a dataset containing 9,222 events, including 3,928 for GR, 2,286 for WA, 2,032 for RE, and 976 for RU, as well as 6,214 labels for AC and 3,008 labels for ST. Deep learning through Multilayer Perceptron Classifiers (MLPC) were coded and implemented using a split configuration of 70% of the data for training and 30% for testing, respectively. In preliminary runs, models had reduced ability to properly discriminate among RE (F1 = 0.42) and RU (F1 = 0.43) Thus, RE and RU were merged on subsequent tests, resulting in 3,928 labels for GR, 2,286 labels for WA, and 3,008 labels for merged RE. Deep learning models successfully classified between AC vs. ST behavior with an overall F1 performance score of 0.96. Further use of the same deep learning models successfully classified among GR, WA, and RE activities with an overall F1 performance score of 0.91, suggesting satisfactory application of the trained models to monitor cattle grazing activities on desert rangeland.

The secret life of wild animals revealed by accelerometer data: how landscape diversity and seasonality influence the behavioural types of European hares

ContextLandscape composition and configuration, as well as seasonal landscape dynamics shape the behaviour, movement and energy expenditure of animals, i.e. foraging, hiding or fleeing, and ultimately survival. Especially in highly modified agricultural systems, it is crucial to understand how animal behaviour is influenced by landscape context to develop sustainable land management concepts.ObjectivesWe show how landscape composition and configuration, together with seasonal dynamics affect animal behavioural types, accounting for the different life-history events in both sexes.MethodsWe investigated 34 European hares in two contrasting agricultural landscapes (a simple and a complex landscape) by using tri-axial accelerometer data to classify the animals’ behaviour into five categories: resting, foraging, moving, grooming and standing upright (i.e. vigilance behaviour). We tested whether the amount of behaviours per category changed with landscape composition and configuration, season and sex.ResultsDuring peak breeding, hares in areas of high habitat diversity rested more, moved less and spent less time searching for resources. During winter, hares moved more and rested less. Females rested less and foraged more in areas with large agricultural fields.ConclusionsA complex landscape is particularly important during the breeding season, allowing animals to allocate enough energy into reproduction. In winter, hares in areas of low habitat diversity may not find enough thermal and anti-predator shelter to move as much as they would need to meet their requirements. Hence, high habitat diversity and small field sizes guarantee species persistence in human-altered agricultural areas throughout the year.

Predicting moose behaviors from tri-axial accelerometer data using a supervised classification algorithm

BackgroundMonitoring the behavior of wild animals in situ can improve our understanding of how their behavior is related to their habitat and affected by disturbances and changes in their environment. Moose (Alces alces) are keystone species in their boreal habitats, where they are facing environmental changes and disturbances from human activities. How these potential stressors can impact individuals and populations is unclear, in part due to our limited knowledge of the physiology and behavior of moose and how individuals can compensate for stress and disturbances they experience. We collected data from collar-mounted fine-scale tri-axial accelerometers deployed on captive moose in combination with detailed behavioral observations to train a random forest supervised classification algorithm to classify moose accelerometer data into discrete behaviors. To investigate the generalizability of our model to collared new individuals, we quantified the variation in classification performance among individuals.ResultsOur machine learning model successfully classified 3-s accelerometer data intervals from 12 Alaskan moose (A. a. gigas) and two European moose (A. a. alces) into seven behaviors comprising 97.6% of the 395 h of behavioral observations conducted in summer, fall and spring. Classification performance varied among behaviors and individuals and was generally dependent on sample size. Classification performance was highest for the most common behaviors lying with the head elevated, ruminating and foraging (precision and recall across all individuals between 0.74 and 0.90) comprising 79% of our data, and lower and more variable among individuals for the four less common behaviors lying with head down or tucked, standing, walking and running (precision and recall across all individuals between 0.28 and 0.79) comprising 21% of our data.ConclusionsWe demonstrate the use of animal-borne accelerometer data to distinguish among seven main behaviors of captive moose and discuss generalizability of the results to individuals in the wild. Our results can support future efforts to investigate the detailed behavior of collared wild moose, for example in the context of disturbance responses, time budgets and behavior-specific habitat selection.

Uncovering Patterns in Dairy Cow Behaviour: A Deep Learning Approach with Tri-Axial Accelerometer Data.

The accurate detection of behavioural changes represents a promising method of detecting the early onset of disease in dairy cows. This study assessed the performance of deep learning (DL) in classifying dairy cows' behaviour from accelerometry data acquired by single sensors on the cows' left flanks and compared the results with those obtained through classical machine learning (ML) from the same raw data. Twelve cows with a tri-axial accelerometer were observed for 136 ± 29 min each to detect five main behaviours: standing still, moving, feeding, ruminating and resting. For each 8 s time interval, 15 metrics were calculated, obtaining a dataset of 211,720 observation units and 15 columns. The entire dataset was randomly split into training (80%) and testing (20%) datasets. The DL accuracy, precision and sensitivity/recall were calculated and compared with the performance of classical ML models. The best predictive model was an 8-layer convolutional neural network (CNN) with an overall accuracy and F1 score equal to 0.96. The precision, sensitivity/recall and F1 score of single behaviours had the following ranges: 0.93-0.99. The CNN outperformed all the classical ML algorithms. The CNN used to monitor the cows' conditions showed an overall high performance in successfully predicting multiple behaviours using a single accelerometer.

Predicting bite rate of grazing cattle from accelerometry data via semi-supervised regression

Accurate estimation of bite rate, i.e., the number of bites during a given grazing period, is important for the characterisation of cattle grazing behaviour and feed intake. In this paper, we estimate the bite rate of grazing cattle from accelerometer data using semi-supervised linear regression. We use a mixture of labelled and unlabelled accelerometer data collected during the grazing trial to train the semi-supervised model. The labelled data has associated ground-truth values for the bite rate and was recorded by an expert observer at intermittent times over the duration of the trial. The trial involved ten cattle, each fitted with a tri-axial accelerometer mounted on a neck collar, over a period of five weeks. We calculate various features from tri-axial accelerometer data and utilise several feature selection approaches and regression methods to fit bite-rate prediction models to our data and compare their performance. Our leave-one-animal-out cross-validated results show that a semi-supervised linear regression model with the specifically chosen feature set delivers good bite-rate prediction performance with a root mean squared error of 5.73 bites per minute and an R2 value of 0.73. Other regression models tested here (including orthogonal distance regression) yielded similar results.

Energy Expenditure of Disaster Relief Operations Estimated Using a Tri-Axial Accelerometer and a Wearable Heart Rate Monitor.

The management of nutrition, food, and health for disaster relief personnel is one of the crucial aspects for carrying out effective rescue activities during large-scale natural disasters, such as a big earthquake, flooding, and landslide following heavy rainfall or man-made disasters, such as widespread fire in industrial areas. Rescue workers, such as fire fighters and rescue teams who work on the disaster relief operations, have to work long, hard, and irregular hours that require energy (both intake and expenditure), with especially altered eating patterns. Reliable estimates of the energy expenditure (TEE) for such disaster relief operations have not been fully established. Here, we propose to clarify the energy expenditure for each type of large-scale disaster activity conducted by fire fighters. Thirty fire fighters (survey participants in this research) who participated in the simulation training of large-scale disaster activities wore tri-axial accelerometers and heart rate monitors during training; and, post-training, 28 fire fighters submitted complete activity record tables. An estimation formula combining tri-axial accelerometer and heart rate monitor data was used. Additionally, energy expenditure per hour (excluding resting energy expenditure: REE) (per average body weight of participants) was calculated for 10 types of large-scale disaster response activities. We propose utilization of these data as a reference value for examining the TEE of firefighting and rescue operations in future large-scale disasters.

Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes

Diabetes affects millions of people worldwide and is steadily increasing. A serious condition associated with diabetes is low glucose levels (hypoglycemia). Monitoring blood glucose is usually performed by invasive methods or intrusive devices, and these devices are currently not available to all patients with diabetes. Hand tremor is a significant symptom of hypoglycemia, as nerves and muscles are powered by blood sugar. However, to our knowledge, no validated tools or algorithms exist to monitor and detect hypoglycemic events via hand tremors. In this paper, we propose a noninvasive method to detect hypoglycemic events based on hand tremors using accelerometer data. We analyzed triaxial accelerometer data from a smart watch recorded from 33 patients with type 1 diabetes for 1 month. Time and frequency domain features were extracted from acceleration signals to explore different machine learning models to classify and differentiate between hypoglycemic and nonhypoglycemic states. The mean duration of the hypoglycemic state was 27.31 (SD 5.15) minutes per day for each patient. On average, patients had 1.06 (SD 0.77) hypoglycemic events per day. The ensemble learning model based on random forest, support vector machines, and k-nearest neighbors had the best performance, with a precision of 81.5% and a recall of 78.6%. The results were validated using continuous glucose monitor readings as ground truth. Our results indicate that the proposed approach can be a potential tool to detect hypoglycemia and can serve as a proactive, nonintrusive alert mechanism for hypoglycemic events.

AnimalAccML: An open-source graphical user interface for automated behavior analytics of individual animals using triaxial accelerometers and machine learning

Automated collection of accelerometer data and subsequent machine learning modeling are prevalent combined methods for animal behavior recognition. However, there is a lack of customized tools for user-friendly machine learning model development. Meanwhile, existing models in previous research could not be directly used for behavior interpretation. The objective of this study was to design and develop a tool for customized machine learning model development and animal behavior analysis using triaxial accelerometer data. A graphical user interface was programmed with Python and saved in a public repository for open access. The interface mainly consists of pages of ‘Manage Project’, ‘Preprocess Data’, ‘Develop Models’, and ‘Analyze Behavior’. An open dataset containing triaxial accelerometer data of six beef cattle was used to test the developed interface. The main results show that users can customize appropriate machine learning models for behavior analytics through several mouse clicks on the interface. A total of 15 models can be selected and trained to determine an optimal one, and model performance can be optimized by adjusting parameters of window size, step size, and training-to-validation ratio. Data imbalance can be solved by merging minority classes into one. The newly developed model has the capacity to analyze overall behavior time budget, statistics (e.g., mean, minimum, maximum, and standard deviation) of each behavior duration, and frequency of behavior sequences. The tool is supportive for automated animal behavior analytics critical to enhancing animal welfare, housing environment, genetics selection, and flock management.

Accelerometer-Measured Physical Behavior and Cardiorespiratory Fitness as Indicators of Work Ability.

Work ability (WA) reflects an individual's resources, work demands, and related environment. Self-reports have shown that higher physical activity (PA) is associated with better WA. This study investigated associations of accelerometer-measured (24/7) physical behavior and cardiorespiratory fitness (CRF) with WA. In the FinFit2017-population-based study, the physical behavior of 20-69-year-old working Finns was measured in terms of PA, standing, and sedentariness using validated MAD-APE algorithms based on raw triaxial accelerometer data. During waking hours, the accelerometer was hip-worn, while during the time in bed (TIB), it was worn on the non-dominant wrist. CRF was measured with a 6 min walk test. WA was assessed by four questions excerpted from the Work Ability Index (WAI), called the short WAI (sWAI). Participants (n = 1668, mean age 46.6, SD = 10.9, 57% women) scored on average 23.3 on the sWAI (range 6-27), with a higher value indicating a better WA. More minutes in standing (p = 0.001) and in moderate (p = 0.004) and vigorous PA (p < 0.001) as well as a higher step number (p < 0.001) and better CRF (p < 0.001) were associated with a higher sWAI value. More time spent lying down (p < 0.001) and in high-movement (p < 0.001) and total TIB (p = 0.001) was associated with a lower sWAI. Detailed analysis of 24/7 physical behavior can be utilized in identifying individual-related indicators of WA.

An efficient deep learning-based approach for human activity recognition using smartphone inertial sensors

Human activity recognition (HAR) has recently witnessed outstanding growth in health and entertainment applications. Owing to the availability of smartphones, many new methods and protocols for using the data from smartphones’ embedded sensors are emerging. Nonetheless, the methods carried out and published in the literature leave a wide area for improvement, in terms of accuracy, resource economy, and adaptation to real-world nuisances. On top of that, a novel classification method that is more economical and efficient is proposed in this paper using both 1D convolutional neural network (1D-CNN) parameters and handcrafted temporal and frequency features with the proficiency of a multilayer perceptron neural network (MLP) classifier. The method proposed requires only tri-axial accelerometer data, allowing it to be deployed even into lower equipment devices; it was tested within the two well-known benchmark datasets: UCI-HAR and Uni-MIB SHAR. Experimental results yield a classification accuracy exceeding 99%, outperforming many of the methods recently shown in the literature.

A “one-size-fits-most” walking recognition method for smartphones, smartwatches, and wearable accelerometers

The ubiquity of personal digital devices offers unprecedented opportunities to study human behavior. Current state-of-the-art methods quantify physical activity using “activity counts,” a measure which overlooks specific types of physical activities. We propose a walking recognition method for sub-second tri-axial accelerometer data, in which activity classification is based on the inherent features of walking: intensity, periodicity, and duration. We validate our method against 20 publicly available, annotated datasets on walking activity data collected at various body locations (thigh, waist, chest, arm, wrist). We demonstrate that our method can estimate walking periods with high sensitivity and specificity: average sensitivity ranged between 0.92 and 0.97 across various body locations, and average specificity for common daily activities was typically above 0.95. We also assess the method’s algorithmic fairness to demographic and anthropometric variables and measurement contexts (body location, environment). Finally, we release our method as open-source software in Python and MATLAB.

Impact of test, vaccinate and remove protocol on home ranges and nightly movements of badgers in a medium density population

In the British Isles, the European badger (Meles meles) is thought to be the primary wildlife reservoir of bovine tuberculosis (bTB), an endemic disease in cattle. Test, vaccinate or remove (‘TVR’) of bTB test-positive badgers, has been suggested to be a potentially useful protocol to reduce bTB incidence in cattle. However, the practice of removing or culling badgers is controversial both for ethical reasons and because there is no consistent observed effect on bTB levels in cattle. While removing badgers reduces population density, it may also result in disruption of their social behaviour, increase their ranging, and lead to greater intra- and inter-species bTB transmission. This effect has been recorded in high badger density areas, such as in southwest England. However, little is known about how TVR affects the behaviour and movement of badgers within a medium density population, such as those that occur in Northern Ireland (NI), which the current study aimed to examine. During 2014–2017, badger ranging behaviours were examined prior to and during a TVR protocol in NI. Nightly distances travelled by 38 individuals were determined using Global Positioning System (GPS) measurements of animal tracks and GPS-enhanced dead-reckoned tracks. The latter was calculated using GPS, tri-axial accelerometer and tri-axial magnetometer data loggers attached to animals. Home range and core home range size were measured using 95% and 50% autocorrelated kernel density estimates, respectively, based on location fixes. TVR was not associated with measured increases in either distances travelled per night (mean = 3.31 ± 2.64 km) or home range size (95% mean = 1.56 ± 0.62 km2, 50% mean = 0.39 ± 0.62 km2) over the four years of study. However, following trapping, mean distances travelled per night increased by up to 44% eight days post capture. Findings differ from those observed in higher density badger populations in England, in which badger ranging increased following culling. Whilst we did not assess behaviours of individual badgers, possible reasons why no differences in home range size were observed include higher inherent ‘social fluidity’ in Irish populations whereby movements are less restricted by habitat saturation and/or that the numbers removed did not reach a threshold that might induce increases in ranging behaviour. Nevertheless, short-term behavioural disruption from trapping was observed, which led to significant increases in the movements of individual animals within their home range. Whether or not TVR may alter badger behaviours remains to be seen, but it would be better to utilise solutions such as oral vaccination of badgers and/or cattle as well as increased biosecurity to limit bTB transmission, which may be less likely to cause interference and thereby reduce the likelihood of bTB transmission.

Joint use of location and acceleration data reveals influences on transitions among habitats in wintering birds

Quantifying relationships between animal behavior and habitat use is essential to understanding animal decision-making. High-resolution location and acceleration data allows unprecedented insights into animal movement and behavior. These data types allow researchers to study the complex linkages between behavioral plasticity and habitat distribution. We used a novel Markov model in a Bayesian framework to quantify the influence of behavioral state frequencies and environmental variables on transitions among landcover types through joint use of location and tri-axial accelerometer data. Data were collected from 56 greater white-fronted geese (Anser albifrons frontalis) across seven ecologically distinct winter regions over two years in midcontinent North America. We showed that goose decision-making varied across landcover types, ecoregions, and abiotic conditions, and was influenced by behavior. We found that time spent in specific behaviors explained variation in the probability of transitioning among habitats, revealing unique behavioral responses from geese among different habitats. Combining GPS and acceleration data allowed unique study of potential influences of an ongoing large-scale range shift in the wintering distribution of a migratory bird across midcontinent North America. We anticipate that behavioral adaptations among variable landscapes is a likely mechanism explaining goose use of highly variable ecosystems during winter in ways which optimize their persistence.

CHAP-Adult: A Reliable and Valid Algorithm to Classify Sitting and Measure Sitting Patterns Using Data From Hip-Worn Accelerometers in Adults Aged 35.

Hip-worn accelerometers are commonly used, but data processed using the 100 counts per minute cut point do not accurately measure sitting patterns. We developed and validated a model to accurately classify sitting and sitting patterns using hip-worn accelerometer data from a wide age range of older adults. Deep learning models were trained with 30-Hz triaxial hip-worn accelerometer data as inputs and activPAL sitting/nonsitting events as ground truth. Data from 981 adults aged 35-99 years from cohorts in two continents were used to train the model, which we call CHAP-Adult (Convolutional Neural Network Hip Accelerometer Posture-Adult). Validation was conducted among 419 randomly selected adults not included in model training. Mean errors (activPAL - CHAP-Adult) and 95% limits of agreement were: sedentary time -10.5 (-63.0, 42.0) min/day, breaks in sedentary time 1.9 (-9.2, 12.9) breaks/day, mean bout duration -0.6 (-4.0, 2.7) min, usual bout duration -1.4 (-8.3, 5.4) min, alpha .00 (-.04, .04), and time in ≥30-min bouts -15.1 (-84.3, 54.1) min/day. Respective mean (and absolute) percent errors were: -2.0% (4.0%), -4.7% (12.2%), 4.1% (11.6%), -4.4% (9.6%), 0.0% (1.4%), and 5.4% (9.6%). Pearson's correlations were: .96, .92, .86, .92, .78, and .96. Error was generally consistent across age, gender, and body mass index groups with the largest deviations observed for those with body mass index ≥30 kg/m2. Overall, these strong validation results indicate CHAP-Adult represents a significant advancement in the ambulatory measurement of sitting and sitting patterns using hip-worn accelerometers. Pending external validation, it could be widely applied to data from around the world to extend understanding of the epidemiology and health consequences of sitting.

Forecasting gait freezing event in Parkinson's patients utilizing machine learning approach from accelerometer signals

Neurological illnesses are one of the most common medical conditions affecting human races and societies worldwide. Parkinson's disease is a neurological ailment caused by absence of dopamine in the human brain and has an impact on the afflicted person's daily routine. Gait freezing event is the most concerning symptom of Parkinson's disease, and it affects around half of people with severe Parkinson's. Machine learning methods are used in this study to detect and forecast gait freezing events. Two hundred thirty seven gait freezing instances from eight patients were collected from tri-axial accelerometer data set and used to train four machine learning classification models. After comparing different performance measures of the four classification models it was found that the Random forest classification model was the most suitable one for predicting gait freezing events in Parkinson disease as it had the best accuracy, sensitivity ,selectivity and least error among the four models. Keywords: Parkinson’s disease , Gait freezing , machine learning , tri-axial accelerometer , Random forest classification model.

Automated classification of sleep and wake from single day triaxial accelerometer data.

Actigraphy allows for the remote monitoring of subjects' activity for clinical and research purposes. However, most standard methods are built for proprietary measures from specific devices that are not widely used. In this study, we develop an algorithm for classifying sleep and awake using a single day of triaxial accelerometer data, which can be acquired from all smart devices. This algorithm consists of two stages, clustering and hidden Markov modeling, and outperforms standard algorithms in sensitivity (94%), specificity (93 %), and overall accuracy (93%) across seven subjects. This method can help automate actigraphy analyses at scale using widely available technology using even a single day's worth of data. Clinical Relevance- Automated monitoring of patients' activity at home can help track recovery trajectories after surgery and injury, disease progression, treatment response.