Validation of accelerometer cut points in toddlers with and without cerebral palsy

Software and equations using segmental measures to estimate height in children and adolescents with cerebral palsy considering the level of gross motor function

BACKGROUND & AIMS

In children with Cerebral palsy (CP) bone deformities create a difficulty in the collection of height measures by direct methods. Body segments are an alternative to study for anthropometric evaluation in children with CP. Motor compromise affects growth in these children. To our knowledge, no equations have been developed to estimate height that consider the level of involvement of children with CP. The aim was to develop equations to estimate height using segmental measures for children with cerebral palsy (CP).

METHODS

This was a cross-sectional study. The sample consisted of children and adolescents with CP of both sexes from 2 to 19 years old from five cities in Argentina. Children whose height and knee-heel height (KH) could be measured were included. Height, KH, and clinical covariables were collected. Linear regression models with height as the dependent variable and KH as predictors adjusted for significant covariates were developed and compared for R2, adjusted R2, and the root mean square of the error.

RESULTS

242 children and adolescents (mean age 9 ± 4 years) with a confirmed diagnosis of CP were included. The interaction between height and other variables such KH, sex, GMFCS, and age was analyzed. Two equations were developed to estimate height according to GMFCS level (GMFCS Level I-III: H = 1.5 × KH(cm) + 2.28 × age(years) + 51; GMFCS Level IV-V: H = 2.13 × KH (cm)+ 0.91 × age(years) + 37). The concordance correlation coefficient between estimated and observed height was 0.95 (95%CI [0.94; 0.96]).

CONCLUSION

Height in children and adolescents with CP can be predicted using KH, GMFCS, and age. The equations and software can estimate height when this cannot be obtained directly.

Measuring Physical Activity in Children Who Do Not Walk and With Cerebral Palsy: An Exploratory Case Series

AIM

The purpose of this case series was to describe physical activity (PA) amount and intensity in the home and school environment. Accelerometers and heart rate (HR) monitors are reliable and valid measures of PA in children with cerebral palsy (CP) who can walk. There is limited research on PA measures in children with CP who cannot walk.

METHODS

Three 9-year-old boys with CP, Gross Motor Function Classification System levels IV and V, participated in a 1-week measurement period wearing waist- and wrist-worn triaxial accelerometers to measure PA counts and a wrist-worn HR monitor to measure PA intensity. PA intensity was calculated using an estimated HR max. Accelerometer counts were reported. Parents and school staff completed activity and eating journals.

RESULTS

Six days of PA and HR data were analyzed. Two participants spent more time in moderate/vigorous PA intensity during school compared to at home. Activities with greatest PA intensity included oral eating, communication, and social engagement. Higher activity counts were recorded from the wrist compared with the waist accelerometers. PA and eating journal adherence were high in both settings.

CONCLUSION

The findings provide preliminary data to evaluate PA amount and intensity in children with CP who have little walking ability. It is important to consider PA levels in daily activity for youth with CP when designing plans of care.

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.

Physical activity in young children across developmental and health states: the ActiveCHILD study

Background

Evidence about physical activity of young children across developmental and health states is very limited. Using data from an inclusive UK cohort, ActiveCHILD, we investigated relationships between objectively measured physical activity, child development, social context, and health-related quality of life (HRQoL).

Methods

Children (12-36 months), purposively sampled across health pathways, developmental abilities, and sociodemographic factors, were recruited through thirteen National Health Service organisations in England. Data were collected from 07/2017 to 08/2019 on: weekly physical activity (3-7 days) using waist-worn accelerometer (ActiGraph 3GTX); sociodemographics, parent actions, child HRQoL, and child development using questionnaires; and child health conditions using clinical records. A data-driven, unsupervised method, called hidden semi-Markov model (HSMM) segmented the accelerometery data and provided estimates of the total time spent active (any intensity) and very active (greater intensity) for each child. Relationships with the explanatory factors were investigated using multiple linear regression.

Findings

Physical activity data were obtained for 282 children (56% females, mean age 21 months, 37.5% with a health condition) covering all index of multiple deprivation deciles. The patterns of physical activity consisted of two daily peaks, children spending 6.44 (SD = 1.39) hours active (any intensity), of which 2.78 (SD = 1.38) hours very active, 91% meeting WHO guidelines. The model for total time active (any intensity) explained 24% of variance, with mobility capacity the strongest predictor (β = 0.41). The model for time spent very active explained 59% of variance, with mobility capacity again the strongest predictor (β = 0.76). There was no evidence of physical activity explaining HRQoL.

Interpretation

The findings provide new evidence that young children across developmental states regularly achieve mainstream recommended physical activity levels and challenges the belief that children with development problems need lower expectations for daily physical activity compared to peers. Advancing the rights of all children to participate in physical activity requires inclusive, equally ambitious, expectations for all.

Funding

Niina Kolehmainen, HEE/NIHR Integrated Clinical Academic Senior Clinical Lecturer, NIHR ICA-SCL-2015-01-00, was funded by the NIHR for this research project. Christopher Thornton, Olivia Craw, Laura Kudlek, and Laura Cutler were also funded from this award. Tim Rapley is a member of the NIHR Applied Research Collaboration North East and North Cumbria, with part of his time funded through the related award (NIHR200173). The views expressed in this publication are those of the author(s) and not necessarily those of the NIHR, NHS, or the UK Department of Health and Social Care. The work of Kianoush Nazarpour is supported by Engineering and Physical Sciences Research Council (EPSRC), under grant number EP/R004242/2.

An evaluation of data processing when using the ActiGraph GT3X accelerometer in non-ambulant children and adolescents with cerebral palsy

PURPOSE

To evaluate vertical acceleration, vector magnitude, non-wear time, valid day classifications, and valid period classifications in the data processing phase when using the ActiGraph GT3X accelerometer in non-ambulant children and adolescents with cerebral palsy (CP).

MATERIAL AND METHODS

Accelerometer data retrieved from 33 non-ambulant children and adolescents (4-17 years) with CP were analysed. Comparisons of (i) vertical acceleration versus vector magnitude, (ii) two different non-wear times, (iii) three different settings to classify a day as valid and (iv) two different settings to classify a period as valid were made.

RESULTS AND CONCLUSIONS

Vector magnitude and a non-wear time of at least 90 consecutive minutes statistically significantly increased minutes recorded per day, especially for sedentary time. There was a statistically significant difference in numbers of valid days depending on time criteria set to determine a valid day, whereas there was no statistically significant difference in valid periods using 3 compared to 4 days. This study suggests using the pre-settings in ActiLife; vector magnitude, non-wear time of 90 consecutive minutes, 500 min recorded per day with periods of at least 3 valid days when assessing physical activity objectively by the ActiGraph GT3X accelerometer in non-ambulant children and adolescents with CP.

Validation and Determination of Physical Activity Intensity GT3X+ Cut-Points in Children and Adolescents with Physical Disabilities: Preliminary Results in a Cerebral Palsy Population

Background: Children and adolescents with disabilities engage in low levels of moderate-to-vigorous intensity physical activity (MVPA), which may create the onset of a sedentary lifestyle. In light of this, MVPA levels must be quantified with a valid tool such as accelerometry. This study aimed to: (i) analyze the accuracy of Evenson cut-points by estimating MVPA and sedentary behavior (SB) in children and adolescents with disabilities; (ii) define new equations to estimate energy expenditure (EE) with the GT3X+ accelerometer in this population and particularly in those with cerebral palsy (CP); (iii) define specific GT3X+ cut-points to estimate MVPA in those with CP. Methods: A total of 23 children and adolescents with disabilities (10 ± 3 years; 44%females) participated in the study. GT3X+-counts and oxygen uptake (VO2) were measured in four laboratory walking conditions. Results: (i) Evenson cut-points were accurate; (ii) new equations were defined to effectively predict EE; (iii) specific GT3X+ cut-points (VM ≥ 702 counts·min−1; Y-Axis ≥ 360 counts·min−1) were defined for estimating MVPA levels in children and adolescents with CP. Conclusions: The use of specific cut-points for ActiGraph GT3X+ seems to be accurate to estimate MVPA levels in children and adolescents with disabilities and, particularly, in those with CP, at least in laboratory conditions.

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.

Operationalization, measurement, and health indicators of sedentary behavior in individuals with cerebral palsy: a scoping review

PURPOSE

To explore the operationalization and measurement of sedentary behavior (SB) in individuals with cerebral palsy (CP).

MATERIALS AND METHODS

We searched five databases from 2011 to 2020 for primary studies of experimental, qualitative, longitudinal, or observational designs measuring SB or postures typically characterized as sedentary (sitting, reclining, lying).

RESULTS

We screened 1112 citations and selected 47 studies. SB was operationalized through muscle activation, energy expenditure or oxygen consumption in typically sedentary postures (n = 9), and through thresholds and postures used by accelerometers, activity monitors, and a questionnaire to measure time spent in SB (n = 25). Seven out of the eight studies that measured energy expenditure found ≤1.5 metabolic equivalents of task (METs) for sitting and lying. While different accelerometer thresholds were used to measure SB, the behavior (SB) was consistently operationalized as sitting and lying. Little consistency existed in the subpopulation, instruments and cut-points for studies on validity or reliability of tools for measuring SB (n = 19).

CONCLUSIONS

Sitting and lying are considered sedentary postures, which is defined as ≤1.5 METs in individuals with CP. There is variability in the tools used to measure SB in individuals with CP. Therefore, consensus on the definition and reporting of SB is needed.Implications for rehabilitationAlthough sedentary behavior (SB) is increased in individuals with cerebral palsy (CP) compared to the typically developing population, there is no standard definition for SB for these individuals; this makes it difficult to synthesize data across studies.Sitting and lying are ≤1.5 METs in individuals with CP, suggesting we only need to measure posture to show change in SB.The commonly used accelerometer cut-point in the typically developing population of ≤100 counts per minute generally has excellent reliability across multiple devices in ambulatory children with CP.

Recurrence of knee flexion contracture after surgical correction in children with cerebral palsy

PURPOSE

Knee flexion contracture (FC) and crouch gait are challenging to treat in children with cerebral palsy (CCP), and recurrent knee FC after surgery is a severe complication. The aim was to identify factors associated with recurrent knee FC after surgery.

METHODS

The records of 62 CCP (age 10.6±2.6 years) who underwent surgery and were followed for > six months were reviewed. Knee FC was treated by hamstring lengthening, posterior knee capsulotomy, and femoral shortening/extension osteotomy until full extension was obtained. Kaplan-Meier analysis was used to estimate the probability of the correction being maintained. Cox proportional hazard modeling was used to compare parameters between patients with and without recurrent knee FC, with the time to recurrence as the endpoint. Potential confounding factors were included in the multivariate analysis.

RESULTS

In total, the procedure was performed on 122 knees. The average weight-for-age z score was -1.3±1.2. The average follow-up period was 5.4±4.2 years. Knee FC recurrence was observed in 37 knees (30%). The average recurrence-free time was 4.9 years, with most cases of recurrence (33 knees) occurring within six years after surgery. The factors associated with recurrence were age (HR, 1.19) and a required posterior knee capsulotomy procedure (HR, 4.53).

CONCLUSION

Recurrent knee FC after correction is associated with CCP who are older at the time of operation and when posterior knee capsulotomy is performed. The age at operation should be optimized. The sequence of procedures should be performed only as necessary to minimize the chance of post-operative muscle weakness and recurrence.

Sleep Assessments for Children With Severe Cerebral Palsy: A Scoping Review

Objectives

To identify the sleep-based instruments in postural care intervention research and examine whether the instruments are suitable as postural care outcome measures specifically for children with severe cerebral palsy.

Data Sources

Investigators searched the electronic databases from 2 university library systems, including OVID Medline, CINAHL, OT Search, Cochrane Database of Systematic Reviews, and Health and Psychosocial Instruments for articles published between 2000 and October 2019.

Study Selection

The initial search yielded 1928 abstracts. Two independent investigators identified 8 English-language peer-reviewed articles that published postural care intervention study results.

Data Extraction

Investigators screened the 8 articles and found that 6 included sleep as a primary or secondary intervention outcome. The principal investigator then fully reviewed these 6 publications, recorded their sleep-related instruments, and applied Coster’s published guidelines (2013) to analyze the sleep-based instruments' suitability as outcome measures.

Data Synthesis:

Collectively, the 6 studies used 8 distinct measures, 6 of which (actigraphy, Chailey Sleep Questionnaire, Pediatric Sleep Questionnaire, polysomnography, sleep diary, and Sleep Disturbance Scale for Children) underwent analysis. As stand-alone instruments, none completely met criteria for suitability as outcome measures for those with severe cerebral palsy.

Conclusions

Combined use of the Sleep Disturbance Scale for Children and actigraphy may be favorable for assessing the sleep-related domains relevant to children with severe cerebral palsy. However, rehabilitation professionals should test sensitivity and specificity to understand the instruments’ ability for capturing changes in sleep from postural care intervention.

Sedentary Behavior in Children With Cerebral Palsy Between 1.5 and 12 Years: A Longitudinal Study

PURPOSE

To determine longitudinal change in sedentary behavior in children with cerebral palsy (CP) from 1.5 to 12 years.

METHODS

Ninety-one children, Gross Motor Function Classification System (GMFCS) levels I to III, who participated in a large longitudinal cohort study were participants. Longitudinal change was analyzed in objectively measured sedentary behavior and associations with sex, body mass index Z score, and socioeconomic status. Moderate-vigorous intensity physical activity (MVPA) was estimated at 8 to 12 years.

RESULTS

Average sedentary minutes/day peaked at 4 years in children at GMFCS I and 5 years in children at GMFCS II to III, then plateaued. Male sex was associated with increased sedentary behavior. At 8 to 12 years, children at GMFCS I, II, and III accumulated on average 54, 47, and 14 minutes/day, respectively, of MVPA.

CONCLUSIONS

When measured to 12 years, sedentary behavior peaks by 5 years for children with CP who are walking with differences in trajectory according to GMFCS.

Clinical gait analysis and physical examination don't correlate with physical activity of children with cerebral palsy. Cross-sectional study

Gait analysis and physical clinical measures are usually performed in children with cerebral palsy to help the surgeons make therapeutic decision. However, the level of physical activity in daily life is not systematically assessed. The aim of this cross sectional study was to examine the correlations between: three-dimensional gait analysis kinematic and spatiotemporal parameters, clinical measures and physical activity. Participants were 30 children with cerebral palsy (10–18 y), with GMFCS I–III. Daily physical activity was measured with an Actigraph GT3X accelerometer in free living environment during seven consecutive days. The percent of time spent in sedentary, in moderate to vigorous physical activity and the number of steps per day were computed from the accelerometer data. Kinematics parameters did not correlate with physical activity. Moderate correlations were found between spatio-temporal parameters and physical activity, for instance timing of toe-off (r = −0.40, p = 0.03). Few physical examination parameters were correlated with physical activity, such as the hip flexors selective motor control (r = 0.69 with moderate to vigorous activity and r = 0.70 with steps per day, p < 0.05). The physical activity profile cannot be sufficiently determined by a combination of clinical measures.

Machine learning to quantify habitual physical activity in children with cerebral palsy

AIM

To investigate whether activity-monitors and machine learning models could provide accurate information about physical activity performed by children and adolescents with cerebral palsy (CP) who use mobility aids for ambulation.

METHOD

Eleven participants (mean age 11y [SD 3y]; six females, five males) classified in Gross Motor Function Classification System (GMFCS) levels III and IV, completed six physical activity trials wearing a tri-axial accelerometer on the wrist, hip, and thigh. Trials included supine rest, upper-limb task, walking, wheelchair propulsion, and cycling. Three supervised learning algorithms (decision tree, support vector machine [SVM], random forest) were trained on features in the raw-acceleration signal. Model-performance was evaluated using leave-one-subject-out cross-validation accuracy.

RESULTS

Cross-validation accuracy for the single-placement models ranged from 59% to 79%, with the best performance achieved by the random forest wrist model (79%). Combining features from two or more accelerometer placements significantly improved classification accuracy. The random forest wrist and hip model achieved an overall accuracy of 92%, while the SVM wrist, hip, and thigh model achieved an overall accuracy of 90%.

INTERPRETATION

Models trained on features in the raw-acceleration signal may provide accurate recognition of clinically relevant physical activity behaviours in children and adolescents with CP who use mobility aids for ambulation in a controlled setting.

WHAT THIS PAPER ADDS

Machine learning may assist clinicians in evaluating the efficacy of surgical and therapy-based interventions. Machine learning may help researchers better understand the short- and long-term benefits of physical activity for children with more severe motor impairments.

Assessing Motor Function in Congenital Muscular Dystrophy Patients Using Accelerometry

BACKGROUND

When tested in a controlled clinic environment, individuals with neuromuscular-related symptoms may complete motor tasks within normal predicted ranges. However, measuring activity at home may better reflect typical motor performance. The accuracy of accelerometry measurements in individuals with congenital muscular dystrophy (CMD) is unknown. We aimed to compare accelerometry and manual step counts and assess free-living physical activity intensity in individuals with CMD using accelerometry.

METHODS

Ambulatory pediatric CMD participants (n = 9) performed the 6-minute walk test in clinic while wearing ActiGraph GT3X accelerometer devices. During the test, manual step counting was conducted to assess concurrent validity of the ActiGraph step count in this population using Bland-Altman analysis. In addition, activity intensity of 6 pediatric CMD participants was monitored at home with accelerometer devices for an average of 7 days. Cut-point values previously validated for neuromuscular disorders were used for data analysis.

RESULTS

Bland-Altman and intraclass correlation analyses showed no concurrent validity between manual and ActiGraph-recorded step counts. Fewer steps were recorded by ActiGraph step counts compared with manual step counts (411 ± 74 vs 699 ± 43, respectively; P = .004). Although improved, results were in the same direction with the application of low-frequency extension filters (587 ± 40 vs 699 ± 43, P = .03). ActiGraph step-count data did not correlate with manual step count (Spearman ρ = 0.32, P = .41; with low-frequency extension: Spearman ρ = 0.45, P = .22). Seven-day physical activity monitoring showed that participants spent more than 80% of their time in the sedentary activity level.

CONCLUSIONS

In a controlled clinic setting, step count was significantly lower by ActiGraph GT3X than by manual step counting, possibly because of the abnormal gait in this population. Additional studies using triaxial assessment are needed to validate accelerometry measurement of activity intensity in individuals with CMD. Accelerometry outcomes may provide valuable measures and complement the 6-minute walk test in the assessment of treatment efficacy in CMD.

Walking Performance, Physical Activity, and Validity of the Early Activity Scale for Endurance in Young Children with Cerebral Palsy

AIMS

To describe the walking performance and physical activity of children with cerebral palsy (CP) by Gross Motor Function Classification System (GMFCS) level, age, sex, and geographical location; and, to examine the concurrent validity of the 4-item Early Activity Scale for Endurance (EASE) to walking performance and physical activity scores.

METHODS

Seventy-nine children with CP participated. Parents completed the 4-item EASE. All children wore an Actigraph monitor (n = 79), and children in GMFCS levels I - III also wore a StepWatch monitor (n = 50), for seven days.

RESULTS

Only GMFCS level yielded significant differences in average strides taken per day, in strides per day taken faster than 30 strides per minute, in average physical activity counts per minute, and in minutes per day spent in moderate to vigorous physical activity. The 4-item EASE findings were moderately correlated with average physical activity counts per minute (.61, p< .001) and minutes per day spent in moderate to vigorous physical activity (.62, p < .001).

CONCLUSIONS

GMFCS level is predictive of both walking performance and physical activity in children with CP. The 4-item EASE may provide a quick and valid way to monitor physical activity in children with CP.

Hip and Wrist-Worn Accelerometer Data Analysis for Toddler Activities

Although accelerometry data are widely utilized to estimate physical activity and sedentary behavior among children age 3 years or older, for toddlers age 1 and 2 year(s), accelerometry data recorded during such behaviors have been far less examined. In particular, toddler’s unique behaviors, such as riding in a stroller or being carried by an adult, have not yet been examined. The objective of this study was to describe accelerometry signal outputs recorded during participation in nine types of behaviors (i.e., running, walking, climbing up/down, crawling, riding a ride-on toy, standing, sitting, riding in a stroller/wagon, and being carried by an adult) among toddlers. Twenty-four toddlers aged 13 to 35 months (50% girls) performed various prescribed behaviors during free play in a commercial indoor playroom while wearing ActiGraph wGT3X-BT accelerometers on a hip and a wrist. Participants’ performances were video-recorded. Based on the video data, accelerometer data were annotated with behavior labels to examine accelerometry signal outputs while performing the nine types of behaviors. Accelerometer data collected during 664 behavior assessments from the 21 participants were used for analysis. Hip vertical axis counts for walking were low (median = 49 counts/5 s). They were significantly lower than those recorded while a toddler was “carried” by an adult (median = 144 counts/5 s; p < 0.01). While standing, sitting, and riding in a stroller, very low hip vertical axis counts were registered (median ≤ 5 counts/5 s). Although wrist vertical axis and vector magnitude counts for “carried” were not higher than those for walking, they were higher than the cut-points for sedentary behaviors. Using various accelerometry signal features, machine learning techniques showed 89% accuracy to differentiate the “carried” behavior from ambulatory movements such as running, walking, crawling, and climbing. In conclusion, hip vertical axis counts alone may be unable to capture walking as physical activity and “carried” as sedentary behavior among toddlers. Machine learning techniques that utilize additional accelerometry signal features could help to recognize behavior types, especially to differentiate being “carried” from ambulatory movements.

Physical activity in non-ambulatory toddlers with cerebral palsy

BACKGROUND

Children with cerebral palsy are less likely to be physically active than their peers, however there is limited evidence regarding self-initiated physical activity in toddlers who are not able, or who may never be able, to walk.

AIMS

The aim of this study was to measure self-initiated physical activity and its relationship to gross motor function and participation in non-ambulatory toddlers with cerebral palsy.

METHODS AND PROCEDURES

Participants were between the ages of 1-3 years. Physical activity during independent floor-play at home was recorded using a wearable tri-axial accelerometer worn on the child's thigh. The Gross Motor Function Measure-66 and the Child Engagement in Daily Life, a parent-reported questionnaire of participation, were administered.

OUTCOMES AND RESULTS

Data were analyzed from the twenty participants who recorded at least 90 min of floor-play (mean: 229 min), resulting in 4598 total floor-play minutes. The relationship between physical activity and gross motor function was not statistically significant (r = 0.20; p = 0.39), nor were the relationships between physical activity and participation (r = 0.05-0.09; p = 0.71-0.84).

CONCLUSIONS AND IMPLICATIONS

The results suggest physical activity during floor-play is not related to gross motor function or participation in non-ambulatory toddlers with cerebral palsy. Clinicians and researchers should independently measure physical activity, gross motor function, and participation.

Longitudinal trajectories of physical activity and walking performance by gross motor function classification system level for children with cerebral palsy

Purpose: This study aims to present developmental trajectories for physical activity (PA) and walking performance for children with cerebral palsy (CP).Materials and methods: Seventy-nine children with CP, 39 (49%) female, Gross Motor Functional Classification System levels I-V, and mean age 91.3 months (+/-27.7 SD) participated. Participants in levels I-V wore the Actigraph to capture PA and children in levels I-II also wore a StepWatch (SW) (n = 43) to measure walking performance. Trajectories for average PA counts/minute and number of minutes of moderate to vigorous PA were generated for levels I, II, and III/IV/V (aggregate). Single leg strides/day and average strides faster than 30 strides/min trajectories were generated for levels I-II.Results: Participants did not display plateaus in PA or walking performance based on functional level. Children in all levels showed a decrease in amount and intensity of PA from 3.0 to 12 years old, with participants in level I demonstrating the steepest decline. Children in level I decreased slightly, and level II increased slightly in both walking performance measures from 3.0 to 12 years old.Conclusions: Longitudinal curves demonstrate variations in PA and walking performance by functional level and provide prognostic information as to what changes may be anticipated for children with CP.Implications for rehabilitationLongitudinal developmental trajectories for physical activity and walking performance for children with cerebral palsy across functional levels are documented.Trajectories have potential to support collaborative intervention planning between therapists and families relative to physical activity and walking performance.

Machine learning algorithms for activity recognition in ambulant children and adolescents with cerebral palsy

BACKGROUND

Cerebral palsy (CP) is the most common physical disability among children (2.5 to 3.6 cases per 1000 live births). Inadequate physical activity (PA) is a major problem effecting the health and well-being of children with CP. Practical, yet accurate measures of PA are needed to evaluate the effectiveness of surgical and therapy-based interventions to increase PA. Accelerometer-based motion sensors have become the standard for objectively measuring PA in children and adolescents; however, current methods for estimating physical activity intensity in children with CP are associated with significant error and may dramatically underestimate HPA in children with more severe mobility limitations. Machine learning (ML) models that first classify the PA type and then predict PA intensity or energy expenditure using activity specific regression equations may be more accurate than standalone regression models. However, the feasibility and validity of ML methods has not been explored in youth with CP. Therefore, the purpose of this study was to develop and test ML models for the automatic identification of PA type in ambulant children with CP.

METHODS

Twenty two children and adolescents (mean age: 12.8 ± 2.9 y) with CP classified at GMFCS Levels I to III completed 7 activity trials while wearing an ActiGraph GT3X+ accelerometer on the hip and wrist. Trials were categorised as sedentary (SED), standing utilitarian movements (SUM), comfortable walking (CW), and brisk walking (BW). Random forest (RF), support vector machine (SVM), and binary decision tree (BDT) classifiers were trained with features extracted from the vector magnitude (VM) of the raw acceleration signal using 10 s non-overlapping windows. Performance was evaluated using leave-one-subject out cross validation.

RESULTS

SVM (82.0-89.0%) and RF (82.6-88.8%) provided significantly better classification accuracy than BDT (76.1-86.2%). Hip (82.7-85.5%) and wrist (76.1-82.6%) classifiers exhibited comparable prediction accuracy, while the combined hip and wrist (86.2-89.0%) classifiers achieved the best overall performance. For all classifiers, recognition accuracy was excellent for SED (94.1-97.9%), good to excellent for SUM (74.0-96.6%) and brisk walking (71.5-86.0%), and modest for comfortable walking (47.6-70.4%). When comfortable and brisk walking were combined into a single walking class, recognition accuracy ranged from 90.3 to 96.5%.

CONCLUSIONS

ML methods provided acceptable classification accuracy for detection of a range of activities commonly performed by ambulatory children with CP. The resultant models can help clinicians more effectively monitor bouts of brisk walking in the community. The results indicate that 2-step models that first classify PA type and then predict energy expenditure using activity specific regression equations are worthy of exploration in this patient group.

Assessment of upper limb use in children with typical development and neurodevelopmental disorders by inertial sensors: a systematic review

Understanding development of bimanual upper limb (UL) activities in both typical and atypical conditions in children is important for: i) tailoring rehabilitation programs, ii) monitoring progress, iii) determining outcomes and iv) evaluating effectiveness of treatment/rehabilitation. Recent technological advances, such as wearable sensors, offer possibilities to perform standard medical monitoring. Body-worn motion sensors, mainly accelerometers, have shown very promising results but, so far, these studies have mainly focused on adults. The main aim of this review was to report the evidence of UL activity of both typically developing (TD) children and children with neurodevelopmental disorders (NDDs) that are reliably reported and comparable, using a combination of multiple wearable inertial sensors, both in laboratory and natural settings. Articles were selected from three research databases (PubMed, Web of Science and EBSCO). Included studies reported data on children aged 0-20 years old simultaneously wearing at least two inertial sensors on upper extremities. The collected and reported data were relevant in order to describe the amount of physical activity performed by the two ULs separately. A total of 21 articles were selected: 11 including TD, and 10 regarding NDDs. For each article, a review of both clinical and technical data was performed. We considered inertial sensors used for following aims: (i) to establish activity intensity cut-points; (ii) to investigate validity and reliability of specified markers, placement and/or number of inertial sensors; (iii) to evaluate duration and intensity of natural UL movements, defined motor tasks and tremor; and (iv) to assess efficacy of certain rehabilitation protocols. Our conclusions were that inertial sensors are able to detect differences in use between both hands and that all reviewed studies support use of accelerometers as an objective outcome measure, appropriate in assessing UL activity in young children with NDDs and determining intervention effectiveness. Further research on responsiveness to interventions and consistency with use in real-world settings is needed. This information could be useful in planning UL rehabilitation strategies.

Quality of life and habitual physical activity in children with cerebral palsy aged 5 years: A cross-sectional study

OBJECTIVE

To compare quality of life (QOL) according to ambulatory status and to investigate association with habitual physical activity (HPA) in children with cerebral palsy (CP) aged 5 years.

METHODS

Fifty-eight participants were classified using Gross Motor Function Classification System (GMFCS) as level I = 33, II = 8, III = 6, IV = 3 and V = 8 and assessed for motor function using 66-item Gross Motor Function Measure (GMFM-66). Participants wore an ActiGraph^® triaxial accelerometer for 3 days to measure HPA. Parents completed the parent proxy Cerebral Palsy Quality of Life questionnaire for Children (CP QOL-Child). Linear regression analyses were performed.

RESULTS

Ambulant children with CP (GMFCS I-III) had better parent-reported QOL than non-ambulant children (GMFCS IV-V) in domains of feelings about functioning (mean difference (MD) = 20.0; 95% confidence interval (CI) = 11.7, 28.2), participation and physical health (MD = 14.5; 95%CI = 4.7, 24.4), and emotional well-being and self-esteem (MD = 12.5; 95%CI = 4.8, 20.1). HPA was not associated with QOL domains after controlling for motor function. GMFM scores accounted for 39% of variation for feelings about functioning domain (MD = 0.4; 95%CI = 0.2, 0.6).

CONCLUSIONS

In children with CP aged 5 years, HPA was not associated with parent-reported QOL. Gross motor function contributed to QOL domains of feelings about functioning.

Longitudinal physical activity and sedentary behaviour in preschool-aged children with cerebral palsy across all functional levels

AIM

To investigate longitudinal changes of habitual physical activity (HPA) and sedentary time in children with cerebral palsy (CP) aged 1 year 6 months to 5 years across all functional abilities.

METHOD

At study entry, 95 children (62 males, 33 females) were classified using the Gross Motor Function Classification System (GMFCS) at levels I (50), II (9), III (16), IV (6), and V (14). Physical activity was recorded on a total of 159 occasions at four possible time points: 1 year 6 months to 2 years; 2 years 6 months to 3 years; 4 years; and 5 years using ActiGraph for 3 days. Mixed-effects regression models were used for analyses.

RESULTS

Participants classified at GMFCS levels I and II had stable HPA as they aged. HPA significantly decreased at 5 years in children classified at GMFCS levels III to V. Sedentary time significantly increased at 4 years and 5 years in all participants. Annual HPA significantly reduced in children classified at GMFCS levels III to V (-123 counts/min, 95% confidence interval [CI] -206 to -40) while annual sedentary time significantly increased in all participants (GMFCS levels I-II: 2.4%, 95% CI 0.7-4.1; GMFCS levels III-V: 6.9%, 95% CI 4.6-9.2).

INTERPRETATION

Children with CP at all GMFCS levels should be encouraged to be physically active from early childhood as HPA levels start to decline from 4 years. Breaks in sedentary time are required for all children with CP from the age of 3 years.

Body composition, diet, and physical activity: a longitudinal cohort study in preschoolers with cerebral palsy

BACKGROUND

Altered body composition in children with cerebral palsy (CP) could be due to differences in energy intake, habitual physical activity (HPA), and sedentary time.

OBJECTIVE

We investigated the longitudinal relation between the weight-for-age z score (WZ), fat-free mass (FFM), percentage of body fat (%BF), and modifiable lifestyle factors for all Gross Motor Function Classification System (GMFCS) levels (I-V).

DESIGN

The study was a longitudinal population-based cohort study of children with CP who were aged 18-60 mo (364 assessments in 161 children; boys: 61%; mean ± SD recruitment age: 2.8 ± 0.9 y; GMFCS: I, 48%; II, 11%; III, 15%; IV, 11%; and V, 15%). A deuterium dilution technique or bioelectrical impedance analysis was used to estimate FFM, and the %BF was calculated. Energy intake, HPA, and sedentary time were measured with the use of a 3-d weighed food diary and accelerometer wear. Data were analyzed with the use of a mixed-model analysis.

RESULTS

Children in GMFCS group I did not differ from age- and sex-specific reference children with typical development for weight. Children in GMFCS group IV were lighter-for-age, and children in GMFCS group V had a lower FFM-for-height than those in GMFCS group I. Children in GMFCS groups II-V had a higher %BF than that of children in GMFCS group I, with the exception of orally fed children in GMFCS group V. The mean %BF of children with CP classified them as overfat or obese. There was a positive association between energy intake and FFM and also between HPA level and FFM for children in GMFCS group I.

CONCLUSIONS

Altered body composition was evident in preschool-age children with CP across functional capacities. Gross motor function, feeding method, energy intake, and HPA level in GMFCS I individuals are the strongest predictors of body composition in children with CP between the ages of 18 and 60 mo.

Decision Trees for Detection of Activity Intensity in Youth with Cerebral Palsy

PURPOSE

To develop and test decision tree (DT) models to classify physical activity (PA) intensity from accelerometer output and Gross Motor Function Classification System (GMFCS) classification level in ambulatory youth with cerebral palsy (CP) and compare the classification accuracy of the new DT models to that achieved by previously published cut points for youth with CP.

METHODS

Youth with CP (GMFCS levels I-III) (N = 51) completed seven activity trials with increasing PA intensity while wearing a portable metabolic system and ActiGraph GT3X accelerometers. DT models were used to identify vertical axis (VA) and vector magnitude (VM) count thresholds corresponding to sedentary (SED) (<1.5 METs), light-intensity PA (LPA) (≥1.5 and <3 METs) and moderate-to-vigorous PA (MVPA) (≥3 METs). Models were trained and cross-validated using the "rpart" and "caret" packages within R.

RESULTS

For the VA (VA_DT) and VM DT (VM_DT), a single threshold differentiated LPA from SED, whereas the threshold for differentiating MVPA from LPA decreased as the level of impairment increased. The average cross-validation accuracies for the VC_DT were 81.1%, 76.7%, and 82.9% for GMFCS levels I, II, and III. The corresponding cross-validation accuracies for the VM_DT were 80.5%, 75.6%, and 84.2%. Within each GMFCS level, the DT models achieved better PA intensity recognition than previously published cut points. The accuracy differential was greatest among GMFCS level III participants, in whom the previously published cut points misclassified 40% of the MVPA activity trials.

CONCLUSIONS

The GMFCS-specific cut points provide more accurate assessments of MVPA levels in youth with CP across the full spectrum of ambulatory ability.

Habitual Physical Activity in Children With Cerebral Palsy Aged 4 to 5 Years Across All Functional Abilities

PURPOSE

To compare ambulatory status in children with cerebral palsy aged 4 to 5 years with their habitual physical activity and time spent sedentary, and to compare their activity with physical activity guidelines.

METHODS

Sixty-seven participants-independently ambulant, marginally ambulant, and nonambulant-wore accelerometers for 3 days. Time spent sedentary as a percentage of wear time and activity counts were compared between groups.

RESULTS

There were significant differences in time spent sedentary and activity counts between groups. Children who were independently ambulant were more likely to meet physical activity guidelines.

CONCLUSION

Children with cerebral palsy spent more than half of their waking hours in sedentary time. Interventions to reduce sedentary behavior and increase habitual physical activity are needed in children with cerebral palsy at age 4 to 5 years.

Sedentary and Active Time in Toddlers with and without Cerebral Palsy

INTRODUCTION/PURPOSE

To evaluate differences in sedentary time and compare levels of physical activity and sedentary behavior to the Australian physical activity recommendations between toddlers with cerebral palsy (CP) according to functional capacity (Gross Motor Function Classification System [GMFCS]) and age-matched children with typical development (CTD).

METHODS

Children (2.4 ± 0.5 yr old) were split into CTD (n = 20), GMFCS I-II (n = 32), GMFCS III (n = 14), and GMFCS IV-V (n = 12) groups and wore a triaxial ActiGraph® for 3 d. Validated cut points were applied to identify sedentary and active time and the number and duration of sedentary bouts and breaks for each group. Analysis of variance (ANOVA) with post hoc testing, chi-square analysis, and the Fisher exact test were used to compare groups.

RESULTS

No difference between the CTD group (49%) and GMFCS I-II group (52%) was found for sedentary time as a percentage of wear time. The GMFCS III group was more sedentary than both these groups (62%, P < 0.05). The GMFCS IV-V group was more sedentary than all the other groups (74%, P < 0.05). The CTD group and GMFCS I-II group was more likely to spend 180 min or longer in active play on all 3 d than the GMFCS IV-V group (P < 0.05). The GMFCS IV-V group was more likely to have sedentary bouts ≥60 min or longer than all other groups (P < 0.05).

CONCLUSION

Differences in sedentary behavior between the CTD and mildly impaired children with CP (GMFCS I-II) are not evident in the toddler years. Children with moderate-to-severe functional impairment are progressively more sedentary and less likely to meet physical activity guidelines. Further research into the health implications of high levels of sedentary behavior in toddlers is required.

Validation of Accelerometer Cut-Points in Children With Cerebral Palsy Aged 4 to 5 Years

PURPOSE

To derive and validate triaxial accelerometer cut-points in children with cerebral palsy (CP) and compare these with previously established cut-points in children with typical development.

METHODS

Eighty-four children with CP aged 4 to 5 years wore the ActiGraph during a play-based gross motor function measure assessment that was video-taped for direct observation. Receiver operating characteristic and Bland-Altman plots were used for analyses.

RESULTS

The ActiGraph had good classification accuracy in Gross Motor Function Classification System (GMFCS) levels III and V and fair classification accuracy in GMFCS levels I, II, and IV. These results support the use of the previously established cut-points for sedentary time of 820 counts per minute in children with CP aged 4 to 5 years across all functional abilities.

CONCLUSIONS

The cut-point provides an objective measure of sedentary and active time in children with CP. The cut-point is applicable to group data but not for individual children.