Machine Learning Models for Classifying Physical Activity in Free-Living Preschool Children

A Data-Driven Approach to Predicting Recreational Activity Participation Using Machine Learning

Purpose: With the popularity of recreational activities, the study aimed to develop prediction models for recreational activity participation and explore the key factors affecting participation in recreational activities. Methods: A total of 12,712 participants, excluding individuals under 20, were selected from the National Health and Nutrition Examination Survey (NHANES) from 2011 to 2018. The mean age of the sample was 46.86 years (±16.97), with a gender distribution of 6,721 males and 5,991 females. The variables included demographic, physical-related variables, and lifestyle variables. This study developed 42 prediction models using six machine learning methods, including logistic regression, Support Vector Machine (SVM), decision tree, random forest, eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM). The relative importance of each variable was evaluated by permutation feature importance. Results: The results illustrated that the LightGBM was the most effective algorithm for predicting recreational activity participation (accuracy: .838, precision: .783, recall: .967, F1-score: .865, AUC: .826). In particular, prediction performance increased when the demographic and lifestyle datasets were used together. Next, as the result of the permutation feature importance based on the top models, education level and moderate-vigorous physical activity (MVPA) were found to be essential variables. Conclusion: These findings demonstrated the potential of a data-driven approach utilizing machine learning in a recreational discipline. Furthermore, this study interpreted the prediction model through feature importance analysis to overcome the limitation of machine learning interpretability.

Comparing cadence-based and machine learning based estimates for physical activity intensity classification: The UK Biobank

OBJECTIVES

Cadence thresholds have been widely used to categorize physical activity intensity in health-related research. We examined the convergent validity of two cadence-based intensity classification approaches against a machine-learning-based intensity schema in 84,315 participants (≥40 years) with wrist-worn accelerometers.

DESIGN

Validity study.

METHODS

Both cadence-based methods (one-level cadence, two-level cadence) calculated intensity-specific time based on cadence-thresholds while the two-level cadence identified stepping behaviors first. We used an overlapping plot, mean absolute error, and Spearman's correlation coefficient to examine agreements between the cadence-based and machine-learning methods. We also evaluated agreements between methods based on practically-important-difference (moderate-to-vigorous-physical activity: ±20 min/day, moderate-physical activity: ±15, vigorous-physical activity: ±2.5, light-physical activity: ±30).

RESULTS

The group-level (median) minutes of moderate-to-vigorous- and moderate-physical activity estimated by one-level cadence were within the range of practically-important-difference compared to the machine-learning method (bias of median: moderate-to-vigorous-physical activity, -3.5, interquartile range [-15.8, 12.2]; moderate-physical activity, -6.0 [-17.2, 4.1]). The group-level vigorous- and light-physical activity minutes derived by two-level cadence were within practically-important-difference range (vigorous-physical activity: -0.9 [-3.1, 0.5]; light-physical activity, -1.3 [-28.2, 28.9]). The individual-level differences between the cadence-based and machine learning methods were high across intensities (e.g., moderate-to-vigorous-physical activity: mean absolute error [one-level cadence: 24.2 min/day; two-level cadence: 26.2]), with the proportion of participants within the practically-important-difference ranging from 8.4 % to 61.6 %.

CONCLUSIONS

One-level cadence showed acceptable group-level estimates of moderate-to-vigorous and moderate-physical activity while two-level cadence showed acceptable group-level estimates of vigorous- and light-physical activity. The cadence-based methods might not be appropriate for individual-level intensity-specific time estimation.

Artificial intelligence in preventive cardiology

Artificial intelligence (AI) is a field of study that strives to replicate aspects of human intelligence into machines. Preventive cardiology, a subspeciality of cardiovascular (CV) medicine, aims to target and mitigate known risk factors for CV disease (CVD). AI's integration into preventive cardiology may introduce novel treatment interventions and AI-centered clinician assistive tools to reduce the risk of CVD. AI's role in nutrition, weight loss, physical activity, sleep hygiene, blood pressure, dyslipidemia, smoking, alcohol, recreational drugs, and mental health has been investigated. AI has immense potential to be used for the screening, detection, and monitoring of the mentioned risk factors. However, the current literature must be supplemented with future clinical trials to evaluate the capabilities of AI interventions for preventive cardiology. This review discusses present examples, potentials, and limitations of AI's role for the primary and secondary prevention of CVD.

Reliability and validity of rapid assessment tools for measuring 24-hour movement behaviours in children aged 0-5 years: the Movement Behaviour Questionnaire Baby (MBQ-B) and child (MBQ-C)

BACKGROUND

The development of validated "fit-for-purpose" rapid assessment tools to measure 24-hour movement behaviours in children aged 0-5 years is a research priority. This study evaluated the test-retest reliability and concurrent validity of the open-ended and closed-ended versions of the Movement Behaviour Questionnaire for baby (MBQ-B) and child (MBQ-C).

METHODS

300 parent-child dyads completed the 10-day study protocol (MBQ-B: N = 85; MBQ-C: N = 215). To assess validity, children wore an accelerometer on the non-dominant wrist (ActiGraph GT3X+) for 7 days and parents completed 2 × 24-hour time use diaries (TUDs) recording screen time and sleep on two separate days. For babies (i.e., not yet walking), parents completed 2 × 24-hour TUDs recording tummy time, active play, restrained time, screen time, and sleep on days 2 and 5 of the 7-day monitoring period. To assess test-retest reliability, parents were randomised to complete either the open- or closed-ended versions of the MBQ on day 7 and on day 10. Test-retest intraclass correlation coefficients (ICC's) were calculated using generalized linear mixed models and validity was assessed via Spearman correlations.

RESULTS

Test-retest reliability for the MBQ-B was good to excellent with ICC's ranging from 0.80 to 0.94 and 0.71-0.93 for the open- and closed-ended versions, respectively. For both versions, significant positive correlations were observed between 24-hour diary and MBQ-B reported tummy time, active play, restrained time, screen time, and sleep (rho = 0.39-0.87). Test-retest reliability for the MBQ-C was moderate to excellent with ICC's ranging from 0.68 to 0.98 and 0.44-0.97 for the open- and closed-ended versions, respectively. For both the open- and closed-ended versions, significant positive correlations were observed between 24-hour diary and MBQ-C reported screen time and sleep (rho = 0.44-0.86); and between MBQ-C reported and device-measured time in total activity and energetic play (rho = 0.27-0.42).

CONCLUSIONS

The MBQ-B and MBQ-C are valid and reliable rapid assessment tools for assessing 24-hour movement behaviours in infants, toddlers, and pre-schoolers. Both the open- and closed-ended versions of the MBQ are suitable for research conducted for policy and practice purposes, including the evaluation of scaled-up early obesity prevention programs.

An automated, data-driven approach to children's social dynamics in space and time

Most children first enter social groups of peers in preschool. In this context, children use movement as a social tool, resulting in distinctive proximity patterns in space and synchrony with others over time. However, the social implications of children's movements with peers in space and time are difficult to determine due to the difficulty of acquiring reliable data during natural interactions. In this article, we review research demonstrating that proximity and synchrony are important indicators of affiliation among preschoolers and highlight challenges in this line of research. We then argue for the advantages of using wearable sensor technology and machine learning analytics to quantify social movement. This technological and analytical advancement provides an unprecedented view of complex social interactions among preschoolers in natural settings, and can help integrate young children's movements with others in space and time into a coherent interaction framework.

Emerging methods for measuring physical activity using accelerometry in children and adolescents with neuromotor disorders: a narrative review

BACKGROUND

Children and adolescents with neuromotor disorders need regular physical activity to maintain optimal health and functional independence throughout their development. To this end, reliable measures of physical activity are integral to both assessing habitual physical activity and testing the efficacy of the many interventions designed to increase physical activity in these children. Wearable accelerometers have been used for children with neuromotor disorders for decades; however, studies most often use disorder-specific cut points to categorize physical activity intensity, which lack generalizability to a free-living environment. No reviews of accelerometer data processing methods have discussed the novel use of machine learning techniques for monitoring physical activity in children with neuromotor disorders.

METHODS

In this narrative review, we discuss traditional measures of physical activity (including questionnaires and objective accelerometry measures), the limitations of standard analysis for accelerometry in this unique population, and the potential benefits of applying machine learning approaches. We also provide recommendations for using machine learning approaches to monitor physical activity.

CONCLUSIONS

While wearable accelerometers provided a much-needed method to quantify physical activity, standard cut point analyses have limitations in children with neuromotor disorders. Machine learning models are a more robust method of analyzing accelerometer data in pediatric neuromotor disorders and using these methods over disorder-specific cut points is likely to improve accuracy of classifying both type and intensity of physical activity. Notably, there remains a critical need for further development of classifiers for children with more severe motor impairments, preschool aged children, and children in hospital settings.

Longitudinal effects of dog ownership, dog acquisition, and dog loss on children's movement behaviours: findings from the PLAYCE cohort study

INTRODUCTION

Regular physical activity is important for children's physical and mental health, yet many children do not achieve recommended amounts of physical activity. Dog ownership has been associated with increased physical activity in children, however, there have been no longitudinal studies examining this relationship. This study used data from the Play Spaces and Environments for Children's Physical Activity (PLAYCE) cohort study to examine the longitudinal effects of dog ownership status on children's movement behaviours.

METHODS

Change in dog ownership from preschool (wave 1, age 2-5) to fulltime school (wave 2, age 5-7) was used as a natural experiment with four distinct dog ownership groups: continuing non-dog owners (n = 307), continuing dog owners (n = 204), dog acquired (n = 58), and dog loss (n = 31; total n = 600). Daily movement behaviours, including physical activity, sedentary time, sleep, and screen time, were measured using accelerometry and parent-report surveys. Differences between groups over time and by sex were tested using linear mixed effects regression models.

RESULTS

Girls who acquired a dog increased their light intensity activities and games by 52.0 min/day (95%CI 7.9, 96.0) and girls who lost a dog decreased their light intensity activities and games by 62.1 min/day (95%CI -119.3, -4.9) compared to no change among non-dog owners. Girls and boys who acquired a dog increased their unstructured physical activity by 6.8 (95%CI 3.2, 10.3) and 7.1 (95%CI 3.9, 10.3) occasions/week, compared to no changes among non-dog owners. Girls and boys who lost a dog reduced their unstructured physical activity by 10.2 (95%CI -15.0, -5.3) and 7.7 (95%CI -12.0, -3.5) occasions/week. Girls who lost a dog decreased their total physical activity by 46.3 min/day (95%CI -107.5, 14.8) compared to no change among non-dog owners. Continuing dog ownership was typically not associated with movement behaviours. Dog ownership group was not associated with changes in sleep and had mixed associations with screen time.

CONCLUSION

The positive influence of dog ownership on children's physical activity begins in early childhood and differs by child sex. Further research should examine the specific contributions dog-facilitated physical activity makes to children's overall physical activity, including the intensity and duration of dog walking and play.

Floating Epoch Length Improves the Accuracy of Accelerometry-Based Estimation of Coincident Oxygen Consumption

Estimation of oxygen consumption (VO2) from accelerometer data is typically based on prediction equations developed in laboratory settings using steadily paced and controlled test activities. These equations may not capture the temporary changes in VO2 occurring in sporadic real-life physical activity. In this study, we introduced a novel floating epoch for accelerometer data analysis and hypothesized that an adaptive epoch length provides a more consistent estimation of VO2 in irregular activity conditions than a 6 s constant epoch. Two different activity tests were conducted: a progressive constant-speed test (CS) performed on a track and a 6 min back-and-forth walk test including accelerations and decelerations (AC/DC) performed as fast as possible. Twenty-nine adults performed the CS test, and sixty-one performed the AC/DC test. The data were collected using hip-worn accelerometers and a portable metabolic gas analyzer. General linear models were employed to create the prediction models for VO2 that were cross-validated using both data sets and epoch types as training and validation sets. The prediction equations based on the CS test or AC/DC test and 6 s epoch had excellent performance (R2 = 89%) for the CS test but poor performance for the AC/DC test (31%). Only the VO2 prediction equation based on the AC/DC test and the floating epoch had good performance (78%) for both tests. The overall accuracy of VO2 prediction is compromised with the constant length epoch, whereas the prediction model based on irregular acceleration data analyzed with a floating epoch provided consistent performance for both activities.

Objective Measurement of Posture and Movement in Young Children Using Wearable Sensors and Customised Mathematical Approaches: A Systematic Review

Given the importance of young children's postures and movements to health and development, robust objective measures are required to provide high-quality evidence. This study aimed to systematically review the available evidence for objective measurement of young (0-5 years) children's posture and movement using machine learning and other algorithm methods on accelerometer data. From 1663 papers, a total of 20 papers reporting on 18 studies met the inclusion criteria. Papers were quality-assessed and data extracted and synthesised on sample, postures and movements identified, sensors used, model development, and accuracy. A common limitation of studies was a poor description of their sample data, yet over half scored adequate/good on their overall study design quality assessment. There was great diversity in all aspects examined, with evidence of increasing sophistication in approaches used over time. Model accuracy varied greatly, but for a range of postures and movements, models developed on a reasonable-sized (n > 25) sample were able to achieve an accuracy of >80%. Issues related to model development are discussed and implications for future research outlined. The current evidence suggests the rapidly developing field of machine learning has clear potential to enable the collection of high-quality evidence on the postures and movements of young children.

The association between reallocations of time and health using compositional data analysis: a systematic scoping review with an interactive data exploration interface

BACKGROUND

How time is allocated influences health. However, any increase in time allocated to one behaviour must be offset by a decrease in others. Recently, studies have used compositional data analysis (CoDA) to estimate the associations with health when reallocating time between different behaviours. The aim of this scoping review was to provide an overview of studies that have used CoDA to model how reallocating time between different time-use components is associated with health.

METHODS

A systematic search of four electronic databases (MEDLINE, Embase, Scopus, SPORTDiscus) was conducted in October 2022. Studies were eligible if they used CoDA to examine the associations of time reallocations and health. Reallocations were considered between movement behaviours (sedentary behaviour (SB), light physical activity (LPA), moderate-to-vigorous physical activity (MVPA)) or various activities of daily living (screen time, work, household chores etc.). The review considered all populations, including clinical populations, as well as all health-related outcomes.

RESULTS

One hundred and three studies were included. Adiposity was the most commonly studied health outcome (n = 41). Most studies (n = 75) reported reallocations amongst daily sleep, SB, LPA and MVPA. While other studies reported reallocations amongst sub-compositions of these (work MVPA vs. leisure MVPA), activity types determined by recall (screen time, household chores, passive transport etc.) or bouted behaviours (short vs. long bouts of SB). In general, when considering cross-sectional results, reallocating time to MVPA from any behaviour(s) was favourably associated with health and reallocating time away from MVPA to any behaviour(s) was unfavourably associated with health. Some beneficial associations were seen when reallocating time from SB to both LPA and sleep; however, the strength of the association was much lower than for any reallocations involving MVPA. However, there were many null findings. Notably, most of the longitudinal studies found no associations between reallocations of time and health. Some evidence also suggested the context of behaviours was important, with reallocations of leisure time toward MVPA having a stronger favourable association for health than reallocating work time towards MVPA.

CONCLUSIONS

Evidence suggests that reallocating time towards MVPA from any behaviour(s) has the strongest favourable association with health, and reallocating time away from MVPA toward any behaviour(s) has the strongest unfavourable association with health. Future studies should use longitudinal and experimental study designs, and for a wider range of outcomes.

Neural Network Approach to Investigating the Importance of Test Items for Predicting Physical Activity in Chronic Obstructive Pulmonary Disease

Contracting COPD reduces a patient's physical activity and restricts everyday activities (physical activity disorder). However, the fundamental cause of physical activity disorder has not been found. In addition, costly and specialized equipment is required to accurately examine the disorder; hence, it is not regularly assessed in normal clinical practice. In this study, we constructed a machine learning model to predict physical activity using test items collected during the normal care of COPD patients. In detail, we first applied three types of data preprocessing methods (zero-padding, multiple imputation by chained equations (MICE), and k-nearest neighbor (kNN)) to complement missing values in the dataset. Then, we constructed several types of neural networks to predict physical activity. Finally, permutation importance was calculated to identify the importance of the test items for prediction. Multifactorial analysis using machine learning, including blood, lung function, walking, and chest imaging tests, was the unique point of this research. From the experimental results, it was found that the missing value processing using MICE contributed to the best prediction accuracy (73.00%) compared to that using zero-padding (68.44%) or kNN (71.52%), and showed better accuracy than XGBoost (66.12%) with a significant difference (p < 0.05). For patients with severe physical activity reduction (total exercise < 1.5), a high sensitivity (89.36%) was obtained. The permutation importance showed that "sex, the number of cigarettes, age, and the whole body phase angle (nutritional status)" were the most important items for this prediction. Furthermore, we found that a smaller number of test items could be used in ordinary clinical practice for the screening of physical activity disorder.

Sensitivity and specificity of measuring children's free-living cycling with a thigh-worn Fibion® accelerometer

Objective

Cycling is an important part of children's active travel, but its measurement using accelerometry is a challenge. The aim of the present study was to evaluate physical activity duration and intensity, and sensitivity and specificity of free-living cycling measured with a thigh-worn accelerometer.

Methods

Participants were 160 children (44 boys) aged 11.5 ± 0.9 years who wore a triaxial Fibion® accelerometer on right thigh for 8 days, 24 h per day, and reported start time and duration of all cycling, walking and car trips to a travel log. Linear mixed effects models were used to predict and compare Fibion-measured activity and moderate-to-vigorous activity duration, cycling duration and metabolic equivalents (METs) between the travel types. Sensitivity and specificity of cycling bouts during cycling trips as compared to walking and car trips was also evaluated.

Results

Children reported a total of 1,049 cycling trips (mean 7.08 ± 4.58 trips per child), 379 walking trips (3.08 ± 2.81) and 716 car trips (4.79 ± 3.96). There was no difference in activity and moderate-to-vigorous activity duration (p > .105), a lower cycling duration (-1.83 min, p < .001), and a higher MET-level (0.95, p < .001) during walking trips as compared to cycling trips. Both activity (-4.54 min, p < .001), moderate-to-vigorous activity (-3.60 min, p < .001), cycling duration (-1.74 min, p < .001) and MET-level (-0.99, p < .001) were lower during car trips as compared to cycling trips. Fibion showed the sensitivity of 72.2% and specificity of 81.9% for measuring cycling activity type during the reported cycling trips as compared to walking and car trips when the minimum required duration for cycling was less than 29 s.

Conclusions

The thigh-worn Fibion® accelerometer measured a greater duration of cycling, a lower MET-level, and a similar duration of total activity and moderate-to-vigorous activity during free-living cycling trips as compared to walking trips, suggesting it can be used to measure free-living cycling activity and moderate-to-vigorous activity duration in 10-12-year-old children.

The moderating role of parent perceptions in relationships between objectively measured neighbourhood environment attributes and pre-schooler's physical activity: Findings from the PLAYCE study

We examined the moderating effects of parent perceptions of the neighbourhood environment on associations between objectively measured neighbourhood environment attributes and physical activity among pre-schoolers. The number of neighbourhood parks was positively associated with pre-schooler energetic play when parents had above average perceptions of access to services. Objectively measured street connectivity was associated with fewer minutes of energetic play when pedestrian and traffic safety was perceived to be below average by parents. Greater understanding of the role played by parents in pre-schooler's exposure to physically active supportive environments is needed to inform environmental interventions for specific age groups.

Adherence to the World Health Organization's physical activity recommendation in preschool-aged children: a systematic review and meta-analysis of accelerometer studies

BACKGROUND

The World Health Organization (WHO) recommend that preschool-aged children should engage in 180 min of total physical activity (TPA) including 60 min of moderate-to-vigorous physical activity (MVPA) each day. No systematic reviews or meta-analyses have pooled adherence to the recommendation across multiple studies. This study aimed to estimate the prevalence of preschool-aged children achieving the WHO's physical activity recommendation for young children, and determine if the prevalence differed between boys and girls.

METHODS

Primary literature searches were conducted on six online databases and a machine learning assisted systematic review was used to identify relevant studies. Studies written in English reporting on the prevalence of children aged 3-5 years achieving overall WHO physical activity recommendation or the individual TPA or MVPA aspects of the recommendation measured using accelerometers were eligible for inclusion. Random effects meta-analysis was used to determine the prevalence of preschools achieving the overall WHO recommendation and the individual TPA and MVPA aspect of the recommendation, and to determine difference in prevalence between boys and girls.

RESULTS

Forty-eight studies reporting on 20,078 preschool-aged children met the inclusion criteria. Based on the most commonly employed accelerometer cut-points across all aspects of the recommendation, 60% (95% Confidence Interval [CI] = 37%, 79%) of preschool-aged children adhered to the overall physical activity recommendation, 78% (95% CI = 38%, 95%) adhered to the TPA aspect of the recommendation, and 90% (95% CI = 81%, 95%) adhered to the MVPA aspect of the recommendation. There was substantial variability is prevalence estimates between different accelerometer cut-points. Girls were significantly less likely to achieve the overall recommendation and the MVPA aspect of the recommendation than boys were.

CONCLUSIONS

Although there was substantial variability in estimated prevalence of preschool-aged children adhering the WHO physical activity recommendation between various accelerometer cut-points, the weight of available evidence suggests that the majority of young children are adhering to the overall recommendation and the individual TPA and MVPA aspects of the recommendation. Large-scale, intercontinental surveillance studies are needed to further strengthen the evidence regarding the prevalence of preschool-aged children achieving physical activity recommendation globally.

Sensors for Human Physical Behaviour Monitoring

The understanding and measurement of physical behaviours that occur in everyday life are essential not only for determining their relationship with health, but also for interventions, physical activity monitoring/surveillance of the population and specific groups, drug development, and developing public health guidelines and messages [...].

Parental Influences on Physical Activity and Screen Time among Preschool Children from Low-Income Families in Brazil

Background: Children from low-middle income countries (LMIC) are disproportionately affected by obesity, and low physical activity (PA) and high screen time (ST) are major contributors. Parents are key influencers on children's PA and ST, yet, no study has investigated relationships between parenting practices and children's PA and ST in LMIC families. This study examined parental influences on PA and ST among preschool-aged children from low-income families in Brazil. Methods: Parents completed a validated, culturally adapted interviewer-administered survey assessing child ST and parenting practices. Child sedentary time, total movement, and energetic play were measured by accelerometery. Results: Data were available on 77 parent-child dyads [mean age 4.6 years (standard deviation = 0.8), 53% male, and 41% mixed-race]. Parenting practices associated with greater PA were use of PA to reward/control behavior (rho = 0.34-0.49), limiting or monitoring ST (rho = 0.30), explicit modeling/enjoyment of PA (rho = 0.24), verbal encouragement for PA (rho = 0.30), and importance and value of PA (rho = 0.24-0.38; p < 0.05). Parenting practices associated with higher ST were rules around active play indoor (rho = 0.23), use of ST to reward/control behavior (rho = 0.30), exposure to screens (rho = 0.40), and explicit modeling/enjoyment of ST (rho = 0.50; p < 0.05). Conclusion: Recognized parenting practices such as explicit modeling of PA and ST, monitoring and limiting ST, and rules and restrictions about PA and ST are associated with young children's PA and ST in low-income Brazilian families. The findings identify potential targets for family-based interventions to promote healthy lifestyle behaviors and prevent childhood obesity.

Machine Learning in Nutrition Research

Data currently generated in the field of nutrition are becoming increasingly complex and high-dimensional, bringing with them new methods of data analysis. The characteristics of machine learning (ML) make it suitable for such analysis and thus lend itself as an alternative tool to deal with data of this nature. ML has already been applied in important problem areas in nutrition, such as obesity, metabolic health, and malnutrition. Despite this, experts in nutrition are often without an understanding of ML, which limits its application and therefore potential to solve currently open questions. The current article aims to bridge this knowledge gap by supplying nutrition researchers with a resource to facilitate the use of ML in their research. ML is first explained and distinguished from existing solutions, with key examples of applications in the nutrition literature provided. Two case studies of domains in which ML is particularly applicable, precision nutrition and metabolomics, are then presented. Finally, a framework is outlined to guide interested researchers in integrating ML into their work. By acting as a resource to which researchers can refer, we hope to support the integration of ML in the field of nutrition to facilitate modern research.

Physical Activity among U.S. Preschool-Aged Children: Application of Machine Learning Physical Activity Classification to the 2012 National Health and Nutrition Examination Survey National Youth Fitness Survey

Early childhood is an important development period for establishing healthy physical activity (PA) habits. The objective of this study was to evaluate PA levels in a representative sample of U.S. preschool-aged children. The study sample included 301 participants (149 girls, 3-5 years of age) in the 2012 U.S. National Health and Examination Survey National Youth Fitness Survey. Participants were asked to wear an ActiGraph accelerometer on their wrist for 7 days. A machine learning random forest classification algorithm was applied to accelerometer data to estimate daily time spent in moderate- and vigorous-intensity PA (MVPA; the sum of minutes spent in running, walking, and other moderate- and vigorous-intensity PA) and total PA (the sum of MVPA and light-intensity PA). We estimated that U.S. preschool-aged children engaged in 28 min/day of MVPA and 361 min/day of total PA, on average. MVPA and total PA levels were not significantly different between males and females. This study revealed that U.S. preschool-aged children engage in lower levels of MVPA and higher levels of total PA than the minimum recommended by the World Health Organization.

Systematic review of accelerometer-based methods for 24-h physical behavior assessment in young children (0-5 years old)

BACKGROUND

Accurate accelerometer-based methods are required for assessment of 24-h physical behavior in young children. We aimed to summarize evidence on measurement properties of accelerometer-based methods for assessing 24-h physical behavior in young children.

METHODS

We searched PubMed (MEDLINE) up to June 2021 for studies evaluating reliability or validity of accelerometer-based methods for assessing physical activity (PA), sedentary behavior (SB), or sleep in 0-5-year-olds. Studies using a subjective comparison measure or an accelerometer-based device that did not directly output time series data were excluded. We developed a Checklist for Assessing the Methodological Quality of studies using Accelerometer-based Methods (CAMQAM) inspired by COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN).

RESULTS

Sixty-two studies were included, examining conventional cut-point-based methods or multi-parameter methods. For infants (0-12 months), several multi-parameter methods proved valid for classifying SB and PA. From three months of age, methods were valid for identifying sleep. In toddlers (1-3 years), cut-points appeared valid for distinguishing SB and light PA (LPA) from moderate-to-vigorous PA (MVPA). One multi-parameter method distinguished toddler specific SB. For sleep, no studies were found in toddlers. In preschoolers (3-5 years), valid hip and wrist cut-points for assessing SB, LPA, MVPA, and wrist cut-points for sleep were identified. Several multi-parameter methods proved valid for identifying SB, LPA, and MVPA, and sleep. Despite promising results of multi-parameter methods, few models were open-source. While most studies used a single device or axis to measure physical behavior, more promising results were found when combining data derived from different sensor placements or multiple axes.

CONCLUSIONS

Up to age three, valid cut-points to assess 24-h physical behavior were lacking, while multi-parameter methods proved valid for distinguishing some waking behaviors. For preschoolers, valid cut-points and algorithms were identified for all physical behaviors. Overall, we recommend more high-quality studies evaluating 24-h accelerometer data from multiple sensor placements and axes for physical behavior assessment. Standardized protocols focusing on including well-defined physical behaviors in different settings representative for children's developmental stage are required. Using our CAMQAM checklist may further improve methodological study quality.

PROSPERO REGISTRATION NUMBER

CRD42020184751.

A Segmentation Scheme for Knowledge Discovery in Human Activity Spotting

Spotting the target activity in a sequence of activities and transitions without applying a predetermined window size is a challenging task. The sliding window method, which is the typical approach in segmenting the continuous data stream, operates with optimal segment sizes chosen considering the type and duration of the activities. Nevertheless, activity type and duration are usually unknown to the detection unit in practice. In this study, we proposed the nonpredetermined size windowing (NSW) scheme to spot the target activity performed in a sequence of unseen activities. NSW is built on classifying progress-based features in multilayer training and prediction stages where the time-domain progress is expressed in terms of polynomials. Thus, it operates without incorporating the information regarding duration and type of the activities. We verified our method with a proof-of-concept use case where data are acquired by a single wrist-worn 3-D accelerometer. We compared our method against fixed size windowing performed with varying window sizes and feature extraction schemes; windowed energy, peak frequency, Shannon entropy, and wavelet entropy. Our method outperforms the compared schemes, reaching a median accuracy of 89% and a median true positive rate of 1 in both intrasubject and intersubject cases based on an independent test set where each test instance contains a sequence of nine different activities.

Comparing estimates of physical activity in children across different cut-points and the associations with weight status

This study aimed to compare sedentary time (SED) and intensity-specific physical activity (PA) estimates and the associations of SED and PA with body mass index (BMI) and waist circumference (WC) using three different sets of cut-points in preschool-aged children. A total of 751 children (4.7 ± 0.9 years, boys 52.7%) wore an ActiGraph GT3X+BT accelerometer on their hip for 7 days (24 h). Euclidean norm -1 G with negative values rounded to zero (ENMO) and activity counts from vertical axis (VACounts) and vector magnitude (VMCounts) were derived. Estimates of SED and light, moderate, vigorous, and moderate-to-vigorous PA (MVPA) were calculated for commonly used cut-points by Hildebrand et al., Butte et al., and Evenson et al. Furthermore, the prevalence of meeting the PA recommendation, 180 min/day of which at least 60 min/day being MVPA, were assessed for the cut-points. Multilevel mixed analysis was used to examine associations of SED and PA with BMI and WC. In accordance with the results, SED and PA intensity estimates differed largely across cut-points (i.e., SED = 22-341 min/day; light PA = 52-257 min/day; moderate PA = 5-18 min/day; vigorous PA = 7-17 min/day; MVPA = 13-35 min/day), and the prevalence of children meeting the PA recommendation varied from 4% to 70%. Associations of SED and PA with BMI or WC varied between the cut-points. Our results indicate that SED and PA estimates in preschool-aged children between studies using these cut-points are poorly comparable. Methods facilitating accelerometer-derived PA estimate comparison between studies are highly warranted.

Evaluation of Wrist Accelerometer Cut-Points for Classifying Physical Activity Intensity in Youth

Background

Wrist worn accelerometers are convenient to wear and provide greater compliance. However, methods to transform the resultant output into predictions of physical activity (PA) intensity have been slow to evolve, with most investigators continuing the practice of applying intensity-based thresholds or cut-points. The current study evaluated the classification accuracy of seven sets of previously published youth-specific cut-points for wrist worn ActiGraph accelerometer data.

Methods

Eighteen children and adolescents [mean age (± SD) 14.6 ± 2.4 years, 10 boys, 8 girls] completed 12 standardized activity trials. During each trial, participants wore an ActiGraph GT3X+ tri-axial accelerometer on the wrist and energy expenditure (Youth METs) was measured directly using the Oxycon Mobile portable calorimetry system. Seven previously published sets of ActiGraph cut-points were evaluated: Crouter regression vertical axis, Crouter regression vector magnitude, Crouter ROC curve vertical axis, Crouter ROC curve vector magnitude, Chandler ROC curve vertical axis, Chandler ROC curve vector magnitude, and Hildebrand ENMO. Classification accuracy was evaluated via weighted Kappa. Confusion matrices were generated to summarize classification accuracy and identify patterns of misclassification.

Results

The cut-points exhibited only moderate agreement with directly measured PA intensity, with Kappa ranging from 0.45 to 0.58. Although the cut-points classified sedentary behavior accurately (> 95%), classification accuracy for the light (3-51%), moderate (12-45%), and vigorous-intensity trials (30-88%) was generally poor. All cut-points underestimated the true intensity of the walking trials, with error rates ranging from 35 to 100%, while the intensity of activity trials requiring significant upper body and/or arm movements was consistently overestimated. The Hildebrand cut-points which serve as the default option in the popular GGIR software package misclassified 30% of the light intensity trials as sedentary and underestimated the intensity of moderate and vigorous intensity trials 75% of the time.

Conclusion

Published ActiGraph cut-points for the wrist, developed specifically for school-aged youth, do not provide acceptable classification accuracy for estimating daily time spent in light, moderate, and vigorous intensity physical activity. The development and deployment of more robust accelerometer data reduction methods such as functional data analysis and machine learning approaches continues to be a research priority.

Device-based measurement of physical activity in pre-schoolers: Comparison of machine learning and cut point methods

INTRODUCTION

Machine learning (ML) accelerometer data processing methods have potential to improve the accuracy of device-based assessments of physical activity (PA) in young children. Yet the uptake of ML methods by health researchers has been minimal and the use of cut-points (CP) continues to be the norm, despite evidence of significant misclassification error. The lack of studies demonstrating a relative advantage for ML approaches over CP methods maybe a key contributing factor.

PURPOSE

The current study compared the accuracy of PA intensity predictions provided by ML classification models and previously published CPs for preschool-aged children.

METHODS

In a free-living study, 31 preschool-aged children (mean age = 4.0 ± 0.9 y) wore wrist and hip ActiGraph GT3X+ accelerometers while completing a video recorded 20-minute free play session. Ground truth PA intensity was coded continuously using the Children's Activity Rating Scale (CARS). Accelerometer data was classified as sedentary (SED), light intensity (LPA), or moderate-to-vigorous intensity (MVPA) using ML random forest PA classifiers and published CPs for preschool-aged children. Performance differences were evaluated in a hold-out sample by comparing weighted kappa statistics, classification accuracy for each intensity band, and equivalence testing.

RESULTS

ML classification models (hip: κ = 0.76; wrist: κ = 0.72) exhibited significantly higher agreement with ground truth PA intensity than CP methods (hip: κ = 0.38-0.49; wrist: κ = 0.31-0.44). For the ML models, classification accuracy for SED and LPA ranged from 83% - 88%, while classification accuracy for MVPA ranged from 68% - 78%. For the CP's, classification accuracy ranged from 50% - 94% for SED, 19% - 75% for LPA, and 44% - 76.1% for MVPA. ML classification models showed equivalence (within ± 0.5 SD) with directly observed time in SED, LPA, and MVPA. None of the CP's exhibited evidence of equivalence.

CONCLUSIONS

Under free living conditions, ML classification models for hip or wrist accelerometer data provide more accurate assessments of PA intensity in young children than CP methods. The results demonstrate the relative advantage of ML methods over threshold-based approaches and adds to a growing evidence base supporting the feasibility and accuracy of ML accelerometer data processing methods.

Association between preschooler movement behaviours, family dog ownership, dog play and dog walking: Findings from the PLAYCE study

Physical inactivity in childhood is a major public health issue. Dog ownership has been widely reported to lead to greater physical activity in adults and school-aged children. We examined if dog ownership and dog-facilitated physical activity were associated with higher physical activity in preschoolers. Secondary analysis of the 'Play Spaces & Environments for Children's Physical Activity' (PLAYCE, 2015-2018) study involving 1366, 2-5-year-olds from 122 long day-care centres in Perth, Australia was conducted. Socio-demographics and movement behaviours (physical activity, screen time, sleep) were examined by dog ownership, dog play and dog walking. Dog-owning preschoolers did physical activity 8 times/week more but 6 min/day less park play than non-dog owners (all p < 0.05). Dog-owning preschoolers who played with their dog ≥ 3 times/week did more physical activity, outdoor play and had 16 min/day more sleep (all p < 0.05). For dog-owners, family dog walking ≥ 3 times/week was positively associated with preschooler physical activity, outdoor play and negatively associated with screen time (all p < 0.05). Our findings suggest that the physical activity-related benefits from having a family dog may be realised when preschoolers spend time playing and walking their dog. Dog walking and play, not dog ownership alone, may be an important source of physical activity for preschoolers.

Cross-validation of cut-points in preschool children using different accelerometer placements and data axes

The present study cross-validated various cut-points to assess physical activity and sedentary behaviour in preschoolers, using hip- and wrist-worn accelerometers and both vertical axis and vector magnitude data. Secondly, we examined the influence of epoch length on time estimates of physical activity and sedentary behaviour. Sixty-four preschoolers (34 girls) wore two accelerometers, on their right hip and dominant wrist, during 1 hour of free play. Preschoolers' activities were observed by two trained researchers. Area under the curve (AUC) was calculated for the receiving operating characteristic (ROC) curves as a measure of precision. AUC ranges were 0.603-0.723 for sedentary behaviour, 0.472-0.545 for light physical activity and 0.503-0.661 for moderate-to-vigorous physical activity (MVPA), indicating poor to fair precision. Percentage of time classified as sedentary behaviour, light or MVPA according to observation and accelerometer data varied largely between cut-points, accelerometer placements and axes. The influence of epoch length on time estimates was minimal across cut-points, except for one hip-based vector magnitude cut-point. Across all accelerometer placements and data axes, no set of cut-points demonstrated adequate precision for sedentary behaviour, light physical activity and MVPA. The highly variable and omnidirectional activity pattern of preschoolers may explain the lack of adequate cut-points.

HARTH: A Human Activity Recognition Dataset for Machine Learning

Existing accelerometer-based human activity recognition (HAR) benchmark datasets that were recorded during free living suffer from non-fixed sensor placement, the usage of only one sensor, and unreliable annotations. We make two contributions in this work. First, we present the publicly available Human Activity Recognition Trondheim dataset (HARTH). Twenty-two participants were recorded for 90 to 120 min during their regular working hours using two three-axial accelerometers, attached to the thigh and lower back, and a chest-mounted camera. Experts annotated the data independently using the camera’s video signal and achieved high inter-rater agreement (Fleiss’ Kappa =0.96). They labeled twelve activities. The second contribution of this paper is the training of seven different baseline machine learning models for HAR on our dataset. We used a support vector machine, k-nearest neighbor, random forest, extreme gradient boost, convolutional neural network, bidirectional long short-term memory, and convolutional neural network with multi-resolution blocks. The support vector machine achieved the best results with an F1-score of 0.81 (standard deviation: ±0.18), recall of 0.85±0.13, and precision of 0.79±0.22 in a leave-one-subject-out cross-validation. Our highly professional recordings and annotations provide a promising benchmark dataset for researchers to develop innovative machine learning approaches for precise HAR in free living.

A Machine Learning Classification Model for Monitoring the Daily Physical Behaviour of Lower-Limb Amputees

There are currently limited data on how prosthetic devices are used to support lower-limb prosthesis users in their free-living environment. Possessing the ability to monitor a patient’s physical behaviour while using these devices would enhance our understanding of the impact of different prosthetic products. The current approaches for monitoring human physical behaviour use a single thigh or wrist-worn accelerometer, but in a lower-limb amputee population, we have the unique opportunity to embed a device within the prosthesis, eliminating compliance issues. This study aimed to develop a model capable of accurately classifying postures (sitting, standing, stepping, and lying) by using data from a single shank-worn accelerometer. Free-living posture data were collected from 14 anatomically intact participants and one amputee over three days. A thigh worn activity monitor collected labelled posture data, while a shank worn accelerometer collected 3-axis acceleration data. Postures and the corresponding shank accelerations were extracted in window lengths of 5–180 s and used to train several machine learning classifiers which were assessed by using stratified cross-validation. A random forest classifier with a 15 s window length provided the highest classification accuracy of 93% weighted average F-score and between 88 and 98% classification accuracy across all four posture classes, which is the best performance achieved to date with a shank-worn device. The results of this study show that data from a single shank-worn accelerometer with a machine learning classification model can be used to accurately identify postures that make up an individual’s daily physical behaviour. This opens up the possibility of embedding an accelerometer-based activity monitor into the shank component of a prosthesis to capture physical behaviour information in both above and below-knee amputees. The models and software used in this study have been made open source in order to overcome the current restrictions of applying activity monitoring methods to lower-limb prosthesis users.

Hearables: Making Sense from Motion Artefacts in Ear-EEG for Real-Life Human Activity Classification

Ear-worn devices are rapidly gaining popularity as they provide the means for measuring vital signals in an unobtrusive, 24/7 wearable and discrete fashion. Naturally, these devices are prone to motion artefacts when used in out-of-lab environments, various movements and activities cause relative movement between user's skin and the electrodes. Historically, these artefacts are seen as nuisance resulting in discarding the segments of signal wherever such artefacts are present. In this work, we make use of such artefacts to classify different daily activities that include sitting, speaking aloud, chewing and walking. To this end, multiple classification techniques are employed to identify these activities using 8 features calculated from the electrode and microphone signal embedded in a generic multimodal in-ear sensor. The results show an overall training accuracy of 93% and 90% and a testing accuracy of 85% and 80% when using a KNN and a 2-layer neural network respectively, thus providing a much needed, simple and reliable framework for real-life human activity classification.

Study protocol for Healthy Conversations @ Playgroup: a multi-site cluster randomized controlled trial of an intervention to promote healthy lifestyle behaviours in young children attending community playgroups

BACKGROUND

Early childhood is a critical window for preventing obesity and chronic disease. Yet, 1 in 4 Australian children aged 5 years and under are affected by overweight or obesity; and significant proportions of children under 5 years fail to meet guidelines for diet quality, physical activity (PA), screen time, and sleep. Consequently, effective interventions to promote healthy lifestyle behaviors and prevent obesity during early childhood are needed. Community playgroups provide an opportunity for parents, carers, and children to meet in a safe and relaxed environment to play and share information. The structure, low cost and reach of playgroups provide a unique platform to engage parents in a scalable program to promote healthful lifestyle behaviors and prevent childhood obesity. However, the evidence base for the effectiveness of health promotion programs delivered in community playgroup settings is limited and lacking credible evidence from rigorously conducted randomized controlled trials.

METHODS

The Healthy Conversations @ Playgroup randomized controlled trial (RCT) aims to address the underlying behavioral risk factors for obesity by helping parents take effective steps to improve their child's dietary, PA, screen time, and sleep behaviors. The intervention program comprises 10 "healthy conversations" led by a trained peer facilitator, designed to increase parents' behavioral capability and self-efficacy to implement autonomy-supportive parenting practices. The program will be delivered biweekly during regularly scheduled playgroup sessions over 10-weeks. Effectiveness will be tested in a 2-arm cluster RCT involving 60 community playgroups in three states across Australia. After baseline assessments, participating playgroups will be randomly allocated to either intervention or wait-list control conditions. Primary outcomes (vegetable intake, discretionary foods, daily PA, screen time, sleep duration, and body mass index [BMI] z-score) will be assessed at baseline, immediately post-intervention (10-weeks; T2) and 6-months post-intervention (T3). Outcomes will be assessed for differential change at T2 and T3.

DISCUSSION

The Healthy Conversations @ Playgroup trial will rigorously evaluate a novel peer-led intervention program to promote healthful lifestyle behaviors and prevent obesity in children and families attending community playgroups. If effective, the program could be immediately scaled-up and delivered in community playgroups across Australia.

TRIAL REGISTRATION

Trial registered 22nd January 2021 with the Australian and New Zealand Clinical Trials Registry ( ACTRN12621000055808 ).

Psychometric properties of instruments to measure parenting practices and children's movement behaviors in low-income families from Brazil

Background

Childhood obesity has increased remarkably in low and middle-income (LMIC) countries. Movement behaviors (physical activity, screen time, and sleep) are crucial in the development of overweight and obesity in young children. Yet, few studies have investigated the relationship between children’s movement behaviors and parenting practices because validated measures for use among families from LMIC are lacking. This study evaluated the psychometric properties of previously validated measures of young children’s physical activity, screen time, and sleep and parenting practices, translated and culturally adapted to Brazilian families.

Methods

A total of 78 parent-child dyads completed an interviewer-administered survey twice within 7 days. Child physical activity, sedentary time and sleep were concurrently measured using a wrist-worn accelerometer. Internal consistency and test-retest reliability was assessed using McDonald’s Omega and Intraclass Correlation Coefficients (ICC’s). Concurrent validity was evaluated by calculating Spearman correlations between parent reported child behaviors and accelerometer measured behaviors.

Results

Seventeen of the 19 parenting practices scales exhibited acceptable internal consistency reliability (Ω ≥ 0.70). Test-retest reliability ICC’s were acceptable and ranged from 0.82 - 0.99. Parent reported child physical activity was positively correlated with objectively measured total movement (rho= 0.29 - 0.46, p < .05) and energetic play (rho= 0.29 – 0.40, p < .05). Parent reported child screen time was positively correlated with objectively measured sedentary time; (rho = 0.26, p < .05), and inversely correlated with total movement (rho = - 0.39 – - 0.41, p < .05) and energetic play (rho = - 0.37 – - 0.41, p < .05). Parent reported night-time sleep duration was significantly correlated with accelerometer measured sleep duration on weekdays (rho = 0.29, p < .05), but not weekends.

Conclusions

Measurement tools to assess children’s movement behaviors and parenting practices, translated and culturally adapted for use in Brazilian families, exhibited acceptable evidence of concurrent validity, internal consistency, and test-retest reliability.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-021-01320-y.

Age Differences in Estimating Physical Activity by Wrist Accelerometry Using Machine Learning

Accelerometer-based fitness trackers and smartwatches are proliferating with incessant attention towards health tracking. Despite their growing popularity, accurately measuring hallmark measures of physical activities has yet to be accomplished in adults of all ages. In this work, we evaluated the performance of four machine learning models: decision tree, random forest, extreme gradient boosting (XGBoost) and least absolute shrinkage and selection operator (LASSO), to estimate the hallmark measures of physical activities in young (20-50 years), middle-aged (50-70 years], and older adults (70-89 years]. Our models were built to recognize physical activity types, recognize physical activity intensities, estimate energy expenditure (EE) and recognize individual physical activities using wrist-worn tri-axial accelerometer data (33 activities per participant) from a large sample of participants (n = 253, 62% women, aged 20-89 years old). Results showed that the machine learning models were quite accurate at recognizing physical activity type and intensity and estimating energy expenditure. However, models performed less optimally when recognizing individual physical activities. F1-Scores derived from XGBoost's models were high for sedentary (0.955-0.973), locomotion (0.942-0.964) and lifestyle (0.913-0.949) activity types with no apparent difference across age groups. Low (0.919-0.947), light (0.813-0.828) and moderate (0.846-0.875) physical activity intensities were also recognized accurately. The root mean square error range for EE was approximately 1 equivalent of resting EE [0.835-1.009 METs]. Generally, random forest and XGBoost models outperformed other models. In conclusion, machine learning models to label physical activity types, activity intensity and energy expenditure are accurate and there are minimal differences in their performance across young, middle-aged and older adults.

Precision nutrition: A systematic literature review

Precision Nutrition research aims to use personal information about individuals or groups of individuals to deliver nutritional advice that, theoretically, would be more suitable than generic advice. Machine learning, a subbranch of Artificial Intelligence, has promise to aid in the development of predictive models that are suitable for Precision Nutrition. As such, recent research has applied machine learning algorithms, tools, and techniques in precision nutrition for different purposes. However, a systematic overview of the state-of-the-art on the use of machine learning in Precision Nutrition is lacking. Therefore, we carried out a Systematic Literature Review (SLR) to provide an overview of where and how machine learning has been used in Precision Nutrition from various aspects, what such machine learning models use as input features, what the availability status of the data used in the literature is, and how the models are evaluated. Nine research questions were defined in this study. We retrieved 4930 papers from electronic databases and 60 primary studies were selected to respond to the research questions. All of the selected primary studies were also briefly discussed in this article. Our results show that fifteen problems spread across seven domains of nutrition and health are present. Four machine learning tasks are seen in the form of regression, classification, recommendation and clustering, with most of these utilizing a supervised approach. In total, 30 algorithms were used, with 19 appearing more than once. Models were through the use of four groups of approaches and 23 evaluation metrics. Personalized approaches are promising to reduce the burden of these current problems in nutrition research, and the current review shows Machine Learning can be incorporated into Precision Nutrition research with high performance. Precision Nutrition researchers should consider incorporating Machine Learning into their methods to facilitate the integration of many complex features, allowing for the development of high-performance Precision Nutrition approaches.

Cross-Sectional Associations Between Wake-Time Movement Compositions and Mental Health in Preschool Children With and Without Motor Coordination Problems

Movement behaviors have been found to be important correlates of health for children and may be particularly important for children with Developmental Coordination Disorder (DCD) who often experience greater mental health problems. To date, however, little research has investigated the daily movement composition of preschool children with Developmental Coordination Disorder (DCD) and/or its association with mental health. The purpose of the current study was to: (1) examine whether differences in movement compositions (i.e., sedentary time, light physical activity, moderate-to-vigorous physical activity) exist between typically developing (TD) preschool-age children and those at risk for DCD (rDCD); and (2) investigate associations between movement compositions and mental health indicators. This cross-sectional study used the baseline cohort data from the Coordination and Activity Tracking in CHildren (CATCH) study. A total of 589 preschool-age children (Mage = 4.94 ± 0.59 years; 57.4% boys) were included in this analysis, of which 288 scored at or below the 16th percentile on the Movement Assessment Battery for Children-2 and were thus classified as rDCD. Wake time movement behaviors were measured using accelerometers and parents completed the Child Behavior Checklist to assess their child's mental health (i.e., internalizing and externalizing problems). Compositional data analysis techniques were used. After adjusting for potential confounders, the results demonstrated similar movement compositions between TD and rDCD children. Among the full sample, findings revealed a significant association between sedentary time and externalizing problems, however, each of the other associations did not reach statistical significance. These results are consistent with emerging evidence demonstrating similar patterns of physical activity and sedentary time among TD children and those classified as rDCD during the preschool years. Although movement behaviors explained little variance in mental health during this period, future research should investigate when movement compositions diverge, and how these changes may impact the mental health of TD children and those classified as rDCD later in childhood.

Empirical Mode Decomposition Based Multi-Modal Activity Recognition

This paper aims to develop an activity recognition algorithm to allow parents to monitor their children at home after school. A common method used to analyze electroencephalograms is to use infinite impulse response filters to decompose the electroencephalograms into various brain wave components. However, nonlinear phase distortions will be introduced by these filters. To address this issue, this paper applies empirical mode decomposition to decompose the electroencephalograms into various intrinsic mode functions and categorize them into four groups. In addition, common features used to analyze electroencephalograms are energy and entropy. However, because there are only two features, the available information is limited. To address this issue, this paper extracts 11 different physical quantities from each group of intrinsic mode functions, and these are employed as the features. Finally, this paper uses the random forest to perform activity recognition. It is worth noting that the conventional approach for performing activity recognition is based on a single type of signal, which limits the recognition performance. In this paper, a multi-modal system based on electroencephalograms, image sequences, and motion signals is used for activity recognition. The numerical simulation results show that the percentage accuracies based on three types of signal are higher than those based on two types of signal or the individual signals. This demonstrates the advantages of using the multi-modal approach for activity recognition. In addition, our proposed empirical mode decomposition-based method outperforms the conventional filtering-based method. This demonstrates the advantages of using the nonlinear and adaptive time frequency approach for activity recognition.