Gait Characterization in Duchenne Muscular Dystrophy (DMD) Using a Single-Sensor Accelerometer: Classical Machine Learning and Deep Learning Approaches.

Differences in gait patterns of children with Duchenne muscular dystrophy (DMD) and typically developing (TD) peers are visible to the eye, but quantifications of those differences outside of the gait laboratory have been elusive. In this work, we measured vertical, mediolateral, and anteroposterior acceleration using a waist-worn iPhone accelerometer during ambulation across a typical range of velocities. Fifteen TD and fifteen DMD children from 3 to 16 years of age underwent eight walking/running activities, including five 25 m walk/run speed-calibration tests at a slow walk to running speeds (SC-L1 to SC-L5), a 6-min walk test (6MWT), a 100 m fast walk/jog/run (100MRW), and a free walk (FW). For clinical anchoring purposes, participants completed a Northstar Ambulatory Assessment (NSAA). We extracted temporospatial gait clinical features (CFs) and applied multiple machine learning (ML) approaches to differentiate between DMD and TD children using extracted temporospatial gait CFs and raw data. Extracted temporospatial gait CFs showed reduced step length and a greater mediolateral component of total power (TP) consistent with shorter strides and Trendelenberg-like gait commonly observed in DMD. ML approaches using temporospatial gait CFs and raw data varied in effectiveness at differentiating between DMD and TD controls at different speeds, with an accuracy of up to 100%. We demonstrate that by using ML with accelerometer data from a consumer-grade smartphone, we can capture DMD-associated gait characteristics in toddlers to teens.

Gait Event Detection and Travel Distance Using Waist-Worn Accelerometers across a Range of Speeds: Automated Approach.

Estimation of temporospatial clinical features of gait (CFs), such as step count and length, step duration, step frequency, gait speed, and distance traveled, is an important component of community-based mobility evaluation using wearable accelerometers. However, accurate unsupervised computerized measurement of CFs of individuals with Duchenne muscular dystrophy (DMD) who have progressive loss of ambulatory mobility is difficult due to differences in patterns and magnitudes of acceleration across their range of attainable gait velocities. This paper proposes a novel calibration method. It aims to detect steps, estimate stride lengths, and determine travel distance. The approach involves a combination of clinical observation, machine-learning-based step detection, and regression-based stride length prediction. The method demonstrates high accuracy in children with DMD and typically developing controls (TDs) regardless of the participant's level of ability. Fifteen children with DMD and fifteen TDs underwent supervised clinical testing across a range of gait speeds using 10 m or 25 m run/walk (10 MRW, 25 MRW), 100 m run/walk (100 MRW), 6-min walk (6 MWT), and free-walk (FW) evaluations while wearing a mobile-phone-based accelerometer at the waist near the body's center of mass. Following calibration by a trained clinical evaluator, CFs were extracted from the accelerometer data using a multi-step machine-learning-based process and the results were compared to ground-truth observation data. Model predictions vs. observed values for step counts, distance traveled, and step length showed a strong correlation (Pearson's r = -0.9929 to 0.9986, p < 0.0001). The estimates demonstrated a mean (SD) percentage error of 1.49% (7.04%) for step counts, 1.18% (9.91%) for distance traveled, and 0.37% (7.52%) for step length compared to ground-truth observations for the combined 6 MWT, 100 MRW, and FW tasks. Our study findings indicate that a single waist-worn accelerometer calibrated to an individual's stride characteristics using our methods accurately measures CFs and estimates travel distances across a common range of gait speeds in both DMD and TD peers.

(-)-Epicatechin induces mitochondrial biogenesis and markers of muscle regeneration in adults with Becker muscular dystrophy.

INTRODUCTION: We conducted an open-label study to examine the effects of the flavonoid (-)-epicatechin in seven ambulatory adult patients with Becker muscular dystrophy (BMD). METHODS: Seven participants received (-)-epicatechin 50 mg twice per day for 8 weeks. Pre- and postprocedures included biceps brachii biopsy to assess muscle structure and growth-relevant endpoints by western blotting, mitochondria volume measurement, and cristae abundance by electron microscopy, graded exercise testing, and muscle strength and function tests. RESULTS: Western blotting showed significantly increased levels of enzymes modulating cellular bioenergetics (liver kinase B1 and 5'-adenosine monophosphate-activated protein kinase). Peroxisome proliferator-activated receptor gamma coactivator-1alpha, a transcriptional coactivator of genes involved in mitochondrial biogenesis and cristae-associated mitofilin levels, increased as did cristae abundance. Muscle and plasma follistatin increased significantly while myostatin decreased. Markers of skeletal muscle regeneration myogenin, myogenic regulatory factor-5, myoblast determination protein 1, myocyte enhancer factor-2, and structure-associated proteins, including dysferlin, utrophin, and intracellular creatine kinase, also increased. Exercise testing demonstrated decreased heart rate, maximal oxygen consumption per kilogram, and plasma lactate levels at defined workloads. Tissue saturation index improved in resting and postexercise states. DISCUSSION: (-)-Epicatechin, an exercise mimetic, appears to have short-term positive effects on tissue biomarkers indicative of mitochondrial biogenesis and muscle regeneration, and produced improvements in graded exercise testing parameters in patients with BMD.

Twice-weekly glucocorticosteroids in infants and young boys with Duchenne muscular dystrophy.

INTRODUCTION: Glucocorticosteroids (GC) are effective in slowing weakness in boys with Duchenne muscular dystrophy (DMD). METHODS: This is a multisite, 1-year, open-label trial of twice-weekly prednisolone (5 mg/kg/dose) in infants/young boys (0.4-2.4 years) with DMD. We compared changes in Bayley III Scales of Infant Development (Bayley-III) with untreated boys followed for 1 year (historical control cohort [HCC]). Twenty-three of 25 participants completed the study. RESULTS: Treated boys gained an average of 0.5 points on the Bayley-III gross motor scaled score (GMSS) compared with the HCC who, on average, declined 1.3 points (P = 0.03). All boys maintained linear growth, and none developed Cushingoid features. Excessive weight gain occurred in 13 of 23 (56%) boys. DISCUSSION: This study provides evidence that twice-weekly GC is well tolerated in infants and young boys with DMD and improves GMSS. Excessive weight gain is a potential risk. Longer follow-up is required to determine whether early GC initiation is feasible in most infants/boys with DMD. Muscle Nerve 59:650-657, 2019.

Upper extremity 3-dimensional reachable workspace assessment in amyotrophic lateral sclerosis by Kinect sensor.

INTRODUCTION: Reachable workspace is a measure that provides clinically meaningful information regarding arm function. In this study, a Kinect sensor was used to determine the spectrum of 3-dimensional reachable workspace encountered in a cross-sectional cohort of individuals with amyotrophic lateral sclerosis (ALS). METHODS: Bilateral 3D reachable workspace was recorded from 10 subjects with ALS and 17 healthy controls. The data were normalized by each individual's arm length to obtain a reachable workspace relative surface area (RSA). Concurrent validity was assessed by correlation with scoring on the ALS Functional Rating Score-revised (ALSFRSr). RESULTS: The Kinect-measured reachable workspace RSA differed significantly between the ALS and control subjects (0.579 ± 0.226 vs. 0.786 ± 0.069; P < 0.001). The RSA demonstrated correlation with ALSFRSr upper extremity items (Spearman correlation ρ = 0.569; P = 0.009). With worsening upper extremity function, as categorized by the ALSFRSr, the reachable workspace also decreased progressively. CONCLUSIONS: This study demonstrates the feasibility and potential of using a novel Kinect-based reachable workspace outcome measure in ALS.

Reachable workspace and performance of upper limb (PUL) in duchenne muscular dystrophy.

INTRODUCTION: The Kinect-based reachable workspace relative surface area (RSA) is compared with the performance of upper limb (PUL) assessment in Duchenne muscular dystrophy (DMD). METHODS: 29 individuals with DMD (ages: 7-23; Brooke: 1-5) underwent both Kinect-based reachable workspace RSA and PUL assessments. RSAs were also collected from 24 age-matched controls. Total and quadrant RSAs were compared with the PUL total, shoulder-, middle-, and distal-dimension scores. RESULTS: The total reachable workspace RSA correlated well with the total PUL score (Spearman ρ = -0.602; P < 0.001), and with each of the PUL dimensional scores: shoulder (ρ = -0.624; P < 0.001), middle (ρ = -0.564; P = 0.001), and distal (ρ = -0.630; P < 0.001). With quadrant RSA, reachability in a particular quadrant was closely associated with respective PUL dimensional-level function (lateral-upper quadrant for shoulder-, lateral-upper/lower quadrants for middle-, and lateral-lower quadrant for distal-level function). CONCLUSIONS: This study demonstrates concurrent validity of the reachable workspace outcome measure (RSA) with the DMD-specific upper extremity outcome measure (PUL).

Clinical trial readiness in non-ambulatory boys and men with duchenne muscular dystrophy: MDA-DMD network follow-up.

INTRODUCTION: Outcomes sensitive to change over time in non-ambulatory boys/men with Duchenne muscular dystrophy (DMD) are not well-established. METHODS: Subjects (n = 91; 16.8 ± 4.5 years old) were assessed at baseline and 6-month intervals for 2 years. We analyzed all subjects using an intent-to-treat model and a subset of stronger subjects with Brooke Scale score ≤4, using repeated measures. RESULTS: Eight patients (12-33 years old) died during the study. Sixty-six completed 12-month follow-up, and 51 completed 24-month follow-up. Those taking corticosteroids performed better at baseline, but rates of decline were similar. Forced vital capacity percent predicted (FVC% predicted) declined significantly only after 2 years. However, Brooke and Egen Klassifikation (EK) Scale scores, elbow flexion, and grip strength declined significantly over both 1 and 2 years. CONCLUSION: Brooke and EK Scale scores, elbow flexion, and grip strength were outcomes most responsive to change. FVC% predicted was responsive to change over 2 years. Corticosteroids benefited non-ambulatory DMD subjects but did not affect decline rates of measures tested here. Muscle Nerve 54: 681-689, 2016.

Reachable workspace reflects dynamometer-measured upper extremity strength in facioscapulohumeral muscular dystrophy.

INTRODUCTION: It is not known whether a reduction in reachable workspace closely reflects loss of upper extremity strength in facioscapulohumeral muscular dystrophy (FSHD). In this study we aimed to determine the relationship between reachable workspace and quantitative upper extremity strength measures. METHODS: Maximal voluntary isometric contraction (MVIC) testing of bilateral elbow flexion and shoulder abduction by hand-held dynamometry was performed on 26 FSHD and 27 control subjects. In addition, Kinect sensor-based 3D reachable workspace relative surface areas (RSAs) were obtained. Loading (500-g weight) effects on reachable workspace were also evaluated. RESULTS: Quantitative upper extremity strength (MVIC of elbow flexion and shoulder abduction) correlated with Kinect-acquired reachable workspace RSA (R = 0.477 for FSHD, P = 0.0003; R = 0.675 for the combined study cohort, P < 0.0001). Progressive reduction in RSA reflected worsening MVIC measures. Loading impacted the moderately weak individuals the most with additional reductions in RSA. CONCLUSIONS: Reachable workspace outcome measure is reflective of upper extremity strength impairment in FSHD.

Upper extremity 3-dimensional reachable workspace analysis in dystrophinopathy using Kinect.

INTRODUCTION: An innovative upper extremity 3-dimensional (3D) reachable workspace outcome measure acquired using the Kinect sensor is applied toward Duchenne/Becker muscular dystrophy (DMD/BMD). The validity, sensitivity, and clinical meaningfulness of this novel outcome measure are examined. METHODS: Upper extremity function assessment (Brooke scale and NeuroQOL questionnaire) and Kinect-based reachable workspace analyses were conducted in 43 individuals with dystrophinopathy (30 DMD and 13 BMD, aged 7-60 years) and 46 controls (aged 6-68 years). RESULTS: The reachable workspace measure reliably captured a wide range of upper extremity impairments encountered in both pediatric and adult, as well as ambulatory and non-ambulatory individuals with dystrophinopathy. Reduced reachable workspaces were noted for the dystrophinopathy cohort compared with controls, and they correlated with Brooke grades. In addition, progressive reduction in reachable workspace correlated directly with worsening ability to perform activities of daily living, as self-reported on the NeuroQOL. CONCLUSION: This study demonstrates the utility and potential of the novel sensor-acquired reachable workspace outcome measure in dystrophinopathy.

Reachable workspace in facioscapulohumeral muscular dystrophy (FSHD) by Kinect.

INTRODUCTION: A depth-ranging sensor (Kinect) based upper extremity motion analysis system was applied to determine the spectrum of reachable workspace encountered in facioscapulohumeral muscular dystrophy (FSHD). METHODS: Reachable workspaces were obtained from 22 individuals with FSHD and 24 age- and height-matched healthy controls. To allow comparison, total and quadrant reachable workspace relative surface areas (RSAs) were obtained by normalizing the acquired reachable workspace by each individual's arm length. RESULTS: Significantly contracted reachable workspace and reduced RSAs were noted for the FSHD cohort compared with controls (0.473 ± 0.188 vs. 0.747 ± 0.082; P < 0.0001). With worsening upper extremity function as categorized by the FSHD evaluation subscale II + III, the upper quadrant RSAs decreased progressively, while the lower quadrant RSAs were relatively preserved. There were no side-to-side differences in reachable workspace based on hand-dominance. CONCLUSIONS: This study demonstrates the feasibility and potential of using an innovative Kinect-based reachable workspace outcome measure in FSHD.

Outcome reliability in non-ambulatory boys/men with Duchenne muscular dystrophy.

INTRODUCTION: Therapeutic trials in Duchenne muscular dystrophy (DMD) often exclude non-ambulatory individuals. Here we establish optimal and reliable assessments in a multicenter trial. METHODS: Non-ambulatory boys/men with DMD (N = 91; 16.7 ± 4.5 years of age) were assessed by trained clinical evaluators. Feasibility (percentage completing task) and reliability [intraclass correlation coefficients (ICCs) between morning and afternoon tests] were measured. RESULTS: Forced vital capacity (FVC), assessed in all subjects, showed a mean of 47.8 ± 22% predicted (ICC 0.98). Brooke Upper Extremity Functional Rating (Brooke) and Egen Klassifikation (EK) scales in 100% of subjects showed ICCs ranging from 0.93 to 0.99. Manual muscle testing, range of motion, 9-hole peg test, and Jebsen-Taylor Hand Function Test (JHFT) demonstrated varied feasibility (99% to 70%), with ICCs ranging from 0.99 to 0.64. We found beneficial effects of different forms of corticosteroids for the Brooke scale, percent predicted FVC, and hand and finger strength. CONCLUSIONS: Reliable assessment of non-ambulatory boys/men with DMD is possible. Clinical trials will have to consider corticosteroid use.

Upper extremity reachable workspace evaluation with Kinect.

We propose a novel low-cost method for quantitative assessment of upper extremity workspace envelope using Microsoft Kinect camera. In clinical environment there are currently no practical and cost-effective methods available to provide arm-function evaluation in three-dimensional space. In this paper we examine the accuracy of the proposed technique for workspace estimation using Kinect in comparison with a motion capture system. The experimental results show that the developed system is capable of capturing the workspace with sufficient accuracy and robustness.

mHealth application for upper extremity range of motion and reachable workspace.

We present mobile health (mHealth) applications utilizing embedded phone sensors as an angle-measuring device for upper-limb range of motion (ROM) and estimation of reachable workspace to assist in evaluation of upper limb functional capacity. Our results show that the phone can record accurate measurements, as well as provide additional functionalities for clinicians.

Functional outcome measures in young, steroid-naïve boys with Duchenne muscular dystrophy.

The purpose of this study was to quantitate motor performance in 196 genetically confirmed steroid-naïve boys with Duchenne muscular dystrophy (DMD), to evaluate the test-retest reliability of measures of motor performance in young DMD boys, and to assess correlations among the different functional outcomes including timed tests. Boys aged 4-7 years were recruited in the FOR-DMD study, a comparative effectiveness study of different steroid regimens in DMD. Eligible boys had to be able to rise from the floor independently and to perform pulmonary function testing consistently. The boys were evaluated with standardized assessments at the screening and baseline visits at 32 sites in 5 countries (US, UK, Canada, Italy, Germany). Assessments included timed rise from floor, timed 10 m walk/run, six-minute walk distance, North Star Ambulatory Assessment (NSAA) and forced vital capacity (FVC). Mean age at baseline was 5.9 years (range 4.1-8.1 years). Test-retest reliability was high for functional assessments, regardless of time lag between assessments (up to 90 days) and for the majority of age groups. Correlations were strong among the functional measures and timed tests, less so with FVC. Physiotherapy measures are reliable in a young, steroid-naïve population and rise from floor velocity appears to be a sensitive measure of strength in this population.

The 6-minute walk test as a new outcome measure in Duchenne muscular dystrophy.

Walking abnormalities are prominent in Duchenne muscular dystrophy (DMD). We modified the 6-minute walk test (6MWT) for use as an outcome measure in patients with DMD and evaluated its performance in 21 ambulatory boys with DMD and 34 healthy boys, ages 4 to 12 years. Boys with DMD were tested twice, approximately 1 week apart; controls were tested once. The groups had similar age, height, and weight. All tests were completed. Boys who fell recovered rapidly from falls without injury. Mean +/- SD [range] 6-minute walk distance (6MWD) was lower in boys with DMD than in controls (366 +/- 83 [125-481] m vs. 621 +/- 68 [479-754] m; P < 0.0001; unpaired t-test). Test-retest correlation for boys with DMD was high (r = 0.91). Stride length (R(2) = 0.89; P < 0.0001) was the major determinant of 6MWD for both boys with DMD and controls. A modified 6MWT is feasible and safe, documents disease-related limitations on ambulation, is reproducible, and offers a new outcome measure for DMD natural history and therapeutic trials.

The 6-minute walk test in Duchenne/Becker muscular dystrophy: longitudinal observations.

In this study we used the 6-minute walk distance (6MWD) to characterize ambulation over time in Duchenne/Becker muscular dystrophy (DBMD). The 6MWD was assessed in 18 boys with DBMD and 22 healthy boys, ages 4-12 years, over mean [range] intervals of 58 [39-87] and 69 [52-113] weeks, respectively. Height and weight increased similarly in both groups. At 52 weeks, 6MWD decreased in 12 of 18 (67%) DBMD subjects (overall mean [range]: 357 [125-481] to 300 [0-510] meters; Δ -57 meters, -15.9%), but increased in 14 of 22 (64%) healthy subjects (overall mean [range]: 623 [479-754] to 636 [547-717] meters; Δ +13 meters, +2.1%). Two DBMD subjects lost ambulation. Changes in 6MWD depended on stride length and age; improvements usually occurred by 7-8 years of age; older DBMD subjects worsened, whereas older healthy subjects were stable. The 6MWD changes at 1 year confirm the validity of this endpoint and emphasize that preserving ambulation must remain a major goal of DBMD therapy.

Longitudinal study of upper extremity reachable workspace in fascioscapulohumeral muscular dystrophy.

Facioscapulohumeral Dystrophy (FSHD) results in slowly progressive strength impairment, especially the upper extremities. Recent discoveries regarding pathophysiology have led to exciting novel therapeutic strategies. To further facilitate drug development, improved FSHD outcome measures that are functionally-relevant and sensitive to longitudinal change will be critical. Recently, a motion sensor (Kinect)-based upper extremity outcome called 'reachable workspace' that provides a quantitative reconstruction of an individual's reachability was developed. In this study, changes in reachable workspace were tracked upwards for five-years in 18 FSHD subjects. Results show -1.63 %/year decline in total reachable workspace (p = 0.144); with most notable decline in the above-the-shoulder level quadrants (upper-lateral Q3: -9.5 %/year, p < 0.001 and upper-medial Q1: -6.8 %/ year, p = 0.063) with no significant changes in the lower quadrants (Q2, Q4). Reachable workspace declined more significantly if the subjects were challenged with 500 g wrist weights: total reachable workspace: -1.82 %/year, p = 0.039; Q1: -7.20 %/year, p = 0.041; Q3: -8.09 %/year, p = 0.001. Importantly, reachable workspace outcome was also able to distinguish subgroups in FSHD: mildly- and severely-affected with essentially unchanging reachability over years, and moderately-affected who demonstrate the most detectable changes longitudinally. The study demonstrates utility for measuring declines in upper quadrant reachability, and provides enrichment/stratification of FSHD populations most likely to show treatment effects in clinical trials.

Motor Function Test Reliability During the NeuroNEXT Spinal Muscular Atrophy Infant Biomarker Study.

BACKGROUND: The NeuroNEXT SMA Infant Biomarker Study, a two year, longitudinal, multi-center study of infants with SMA type 1 and healthy infants, presented a unique opportunity to assess multi-site rater reliability on three infant motor function tests (MFTs) commonly used to assess infants with SMA type 1. OBJECTIVE: To determine the effect of prospective MFT rater training and the effect of rater experience on inter-rater and intra-rater reliability for the Test of Infant Motor Performance Screening Items (TIMPSI), the Children's Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP-INTEND) and the Alberta Infant Motor Scale (AIMS). METHODS: Training was conducted utilizing a novel set of motor function test (MFT) videos to optimize accurate MFT administration and reliability for the study duration. Inter- and intra-rater reliability of scoring for the TIMPSI and inter-rater reliability of scoring for the CHOP INTEND and the AIMS was assessed using intraclass correlation coefficients (ICC). Effect of rater experience on reliability was examined using ICC. Agreement with 'expert' consensus scores was examined using Pearson's correlation coefficients. RESULTS: Inter-rater reliability on all MFTs was good to excellent. Intra-rater reliability for the primary MFT, the TIMPSI, was excellent for the study duration. Agreement with 'expert' consensus was within predetermined limits (≥85%) after training. Evaluator experience with SMA and MFTs did not affect reliability. CONCLUSIONS: Reliability of scores across evaluators was demonstrated for all three study MFTs and scores were reproducible on repeated administration. Evaluator experience had no effect on reliability.

Longitudinal evaluation of upper extremity reachable workspace in ALS by Kinect sensor.

Our objective was to evaluate longitudinal changes in Microsoft Kinect measured upper extremity reachable workspace relative surface area (RSA) versus the revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R), ALSFRS-R upper extremity sub-scale and Forced Vital Capacity (FVC) in a cohort of patients diagnosed with amyotrophic lateral sclerosis (ALS). Ten patients diagnosed with ALS (ages 52-76 years, ALSFRS-R: 8-41 at entry) were tested using single 3D depth sensor, Microsoft Kinect, to measure reachable workspace RSA across five visits spanning one year. Changes in RSA, ALSFRS-R, ALSFRS-R upper extremity sub-scale, and FVC were assessed using a linear mixed model. Results showed that upper lateral quadrant RSA declined significantly in one year by approximately 19% (p <0.01) while all other quadrants and total RSA did not change significantly in this time-period. Simultaneously, ALSFRS-R upper extremity sub-scale worsened significantly by 25% (p <0.01). In conclusion, upper extremity reachable workspace RSA as a novel ALS outcome measure is capable of objectively quantifying declines in upper extremity ability over time in patients with ALS with more granularity than other common outcome measures. RSA may serve as a clinical endpoint for the evaluation of upper extremity targeted therapeutics.

Machine Learning to Improve Energy Expenditure Estimation in Children With Disabilities: A Pilot Study in Duchenne Muscular Dystrophy.

BACKGROUND: Children with physical impairments are at a greater risk for obesity and decreased physical activity. A better understanding of physical activity pattern and energy expenditure (EE) would lead to a more targeted approach to intervention. OBJECTIVE: This study focuses on studying the use of machine-learning algorithms for EE estimation in children with disabilities. A pilot study was conducted on children with Duchenne muscular dystrophy (DMD) to identify important factors for determining EE and develop a novel algorithm to accurately estimate EE from wearable sensor-collected data. METHODS: There were 7 boys with DMD, 6 healthy control boys, and 22 control adults recruited. Data were collected using smartphone accelerometer and chest-worn heart rate sensors. The gold standard EE values were obtained from the COSMED K4b2 portable cardiopulmonary metabolic unit worn by boys (aged 6-10 years) with DMD and controls. Data from this sensor setup were collected simultaneously during a series of concurrent activities. Linear regression and nonlinear machine-learning-based approaches were used to analyze the relationship between accelerometer and heart rate readings and COSMED values. RESULTS: Existing calorimetry equations using linear regression and nonlinear machine-learning-based models, developed for healthy adults and young children, give low correlation to actual EE values in children with disabilities (14%-40%). The proposed model for boys with DMD uses ensemble machine learning techniques and gives a 91% correlation with actual measured EE values (root mean square error of 0.017). CONCLUSIONS: Our results confirm that the methods developed to determine EE using accelerometer and heart rate sensor values in normal adults are not appropriate for children with disabilities and should not be used. A much more accurate model is obtained using machine-learning-based nonlinear regression specifically developed for this target population.