Towards automated video-based assessment of dystonia in dyskinetic cerebral palsy: A novel approach using markerless motion tracking and machine learning.

Introduction: Video-based clinical rating plays an important role in assessing dystonia and monitoring the effect of treatment in dyskinetic cerebral palsy (CP). However, evaluation by clinicians is time-consuming, and the quality of rating is dependent on experience. The aim of the current study is to provide a proof-of-concept for a machine learning approach to automatically assess scoring of dystonia using 2D stick figures extracted from videos. Model performance was compared to human performance. Methods: A total of 187 video sequences of 34 individuals with dyskinetic CP (8-23 years, all non-ambulatory) were filmed at rest during lying and supported sitting. Videos were scored by three raters according to the Dyskinesia Impairment Scale (DIS) for arm and leg dystonia (normalized scores ranging from 0-1). Coordinates in pixels of the left and right wrist, elbow, shoulder, hip, knee and ankle were extracted using DeepLabCut, an open source toolbox that builds on a pose estimation algorithm. Within a subset, tracking accuracy was assessed for a pretrained human model and for models trained with an increasing number of manually labeled frames. The mean absolute error (MAE) between DeepLabCut's prediction of the position of body points and manual labels was calculated. Subsequently, movement and position features were calculated from extracted body point coordinates. These features were fed into a Random Forest Regressor to train a model to predict the clinical scores. The model performance trained with data from one rater evaluated by MAEs (model-rater) was compared to inter-rater accuracy. Results: A tracking accuracy of 4.5 pixels (approximately 1.5 cm) could be achieved by adding 15-20 manually labeled frames per video. The MAEs for the trained models ranged from 0.21 ± 0.15 for arm dystonia to 0.14 ± 0.10 for leg dystonia (normalized DIS scores). The inter-rater MAEs were 0.21 ± 0.22 and 0.16 ± 0.20, respectively. Conclusion: This proof-of-concept study shows the potential of using stick figures extracted from common videos in a machine learning approach to automatically assess dystonia. Sufficient tracking accuracy can be reached by manually adding labels within 15-20 frames per video. With a relatively small data set, it is possible to train a model that can automatically assess dystonia with a performance comparable to human scoring.

Ready for handwriting? A reference data study on handwriting readiness assessments.

INTRODUCTION: Early evaluation of writing readiness is essential to predict and prevent handwriting difficulties and its negative influences on school occupations. An occupation-based measurement for kindergarten children has been previously developed: Writing Readiness Inventory Tool In Context (WRITIC). In addition, to assess fine motor coordination two tests are frequently used in children with handwriting difficulties: the modified Timed Test of In-Hand Manipulation (Timed TIHM) and the Nine-Hole Peg Test (9-HPT). However, no Dutch reference data are available. AIM: To provide reference data for (1) WRITIC, (2) Timed-TIHM and (3) 9-HPT for handwriting readiness assessment in kindergarten children. METHODS: Three hundred and seventy-four children from Dutch kindergartens in the age of 5 to 6.5 years (5.6±0.4 years, 190 boys/184 girls) participated in the study. Children were recruited at Dutch kindergartens. Full classes of the last year were tested, children were excluded if there was a medical diagnosis such as a visual, auditory, motor or intellectual impairment that hinder handwriting performance. Descriptive statistics and percentiles scores were calculated. The score of the WRITIC (possible score 0-48 points) and the performance time on the Timed-TIHM and 9-HPT are classified as percentile scores lower than the 15th percentile to distinguish low performance from adequate performance. The percentile scores can be used to identify children that are possibly at risk developing handwriting difficulties in first grade. RESULTS: WRITIC scores ranged from 23 to 48 (41±4.4), Timed-TIHM ranged from 17.9 to 64.5 seconds (31.4± 7.4 seconds) and 9-HPT ranged from 18.2 to 48.3 seconds (28.4± 5.4). A WRITIC score between 0-36, a performance time of more than 39.6 seconds on the Timed-TIHM and more than 33.8 seconds on the 9-HPT were classified as low performance. CONCLUSION: The reference data of the WRITIC allow to assess which children are possibly at risk developing handwriting difficulties.

Quantitative assessment of trunk movements in functional reaching in children and adolescents with dyskinetic cerebral palsy.

BACKGROUND: Trunk control and upper limb function are often disturbed in people with dyskinetic cerebral palsy. While trunk control is fundamental in upper limb activities, insights in trunk control in dyskinetic cerebral palsy are missing. This study aimed to determine trunk movement characteristics in individuals with dyskinetic cerebral palsy during reaching. METHODS: Twenty individuals with dyskinetic cerebral palsy (MACS level I-III (16y6m)) and 20 typical developing peers (17y2m) were included. Participants performed three tasks: reach forward, reach sideways, and reach and grasp vertically, using a cross-sectional study design. Movements were analyzed using 3D motion capture and a sensor on the trunk. Trunk range of motion, joint angle at point of task achievement, peak and range of angular velocity and linear acceleration were compared between groups using Mann-Whitney U and independent t-tests. FINDINGS: Participants with dyskinetic cerebral palsy showed higher trunk range of motion in all planes during reach forward and reach and grasp vertically, and in rotation and lateral flexion during reach sideways. During reach and grasp vertically, the joint angle at point of task achievement differed in the transversal plane. Ranges of angular velocity and linear acceleration were higher for all tasks and planes for participants with dyskinetic cerebral palsy, and for peak values in nearly all planes. INTERPRETATION: Current results provide insights in trunk control at population level. This is a first step towards a better and individualized evaluation and treatment for trunk control, being an important factor in improving functional activities for individuals with dyskinetic cerebral palsy.

Assessment of movement disorders using wearable sensors during upper limb tasks: A scoping review.

Background: Studies aiming to objectively quantify movement disorders during upper limb tasks using wearable sensors have recently increased, but there is a wide variety in described measurement and analyzing methods, hampering standardization of methods in research and clinics. Therefore, the primary objective of this review was to provide an overview of sensor set-up and type, included tasks, sensor features and methods used to quantify movement disorders during upper limb tasks in multiple pathological populations. The secondary objective was to identify the most sensitive sensor features for the detection and quantification of movement disorders on the one hand and to describe the clinical application of the proposed methods on the other hand. Methods: A literature search using Scopus, Web of Science, and PubMed was performed. Articles needed to meet following criteria: 1) participants were adults/children with a neurological disease, 2) (at least) one sensor was placed on the upper limb for evaluation of movement disorders during upper limb tasks, 3) comparisons between: groups with/without movement disorders, sensor features before/after intervention, or sensor features with a clinical scale for assessment of the movement disorder. 4) Outcome measures included sensor features from acceleration/angular velocity signals. Results: A total of 101 articles were included, of which 56 researched Parkinson's Disease. Wrist(s), hand(s) and index finger(s) were the most popular sensor locations. Most frequent tasks were: finger tapping, wrist pro/supination, keeping the arms extended in front of the body and finger-to-nose. Most frequently calculated sensor features were mean, standard deviation, root-mean-square, ranges, skewness, kurtosis/entropy of acceleration and/or angular velocity, in combination with dominant frequencies/power of acceleration signals. Examples of clinical applications were automatization of a clinical scale or discrimination between a patient/control group or different patient groups. Conclusion: Current overview can support clinicians and researchers in selecting the most sensitive pathology-dependent sensor features and methodologies for detection and quantification of upper limb movement disorders and objective evaluations of treatment effects. Insights from Parkinson's Disease studies can accelerate the development of wearable sensors protocols in the remaining pathologies, provided that there is sufficient attention for the standardisation of protocols, tasks, feasibility and data analysis methods.

Instrumented assessment of motor function in dyskinetic cerebral palsy: a systematic review.

BACKGROUND: In this systematic review we investigate which instrumented measurements are available to assess motor impairments, related activity limitations and participation restrictions in children and young adults with dyskinetic cerebral palsy. We aim to classify these instrumented measurements using the categories of the international classification of functioning, disability and health for children and youth (ICF-CY) and provide an overview of the outcome parameters. METHODS: A systematic literature search was performed in November 2019. We electronically searched Pubmed, Embase and Scopus databases. Search blocks included (a) cerebral palsy, (b) athetosis, dystonia and/or dyskinesia, (c) age 2-24 years and (d) instrumented measurements (using keywords such as biomechanics, sensors, smartphone, and robot). RESULTS: Our search yielded 4537 articles. After inspection of titles and abstracts, a full text of 245 of those articles were included and assessed for further eligibility. A total of 49 articles met our inclusion criteria. A broad spectrum of instruments and technologies are used to assess motor function in dyskinetic cerebral palsy, with the majority using 3D motion capture and surface electromyography. Only for a small number of instruments methodological quality was assessed, with only one study showing an adequate assessment of test-retest reliability. The majority of studies was at ICF-CY function and structure level and assessed control of voluntary movement (29 of 49) mainly in the upper extremity, followed by assessment of involuntary movements (15 of 49), muscle tone/motor reflex (6 of 49), gait pattern (5 of 49) and muscle power (2 of 49). At ICF-CY level of activities and participation hand and arm use (9 of 49), fine hand use (5 of 49), lifting and carrying objects (3 of 49), maintaining a body position (2 of 49), walking (1 of 49) and moving around using equipment (1 of 49) was assessed. Only a few methods are potentially suitable outside the clinical environment (e.g. inertial sensors, accelerometers). CONCLUSION: Although the current review shows the potential of several instrumented methods to be used as objective outcome measures in dyskinetic cerebral palsy, their methodological quality is still unknown. Future development should focus on evaluating clinimetrics, including validating against clinical meaningfulness. New technological developments should aim for measurements that can be applied outside the laboratory.

Assessment of net knee moment-angle characteristics by instrumented hand-held dynamometry in children with spastic cerebral palsy and typically developing children.

BACKGROUND: The limited range of motion during walking in children with spastic cerebral palsy (SCP) may be the result of altered mechanical characteristics of muscles and connective tissues around the knee joint. Measurement of static net knee moment-angle relation will provide insights into these alterations, for which instrumented hand-held dynamometry may be applied. The aims of this study were: (1) to test the measurement error of the estimated net knee moment-angle characteristics, (2) to determine the correlation between knee extension angle measurement at a standardized knee moment and popliteal angle from common physical examination and (3) to compare net knee moment-angle characteristics in SCP versus typically developing children. METHODS: With the child lying in sideward position, the knee was extended by moving the lower leg by a hand-held force transducer on a low friction cart. Force data were collected for a range of knee angles. Data were excluded when activity (EMG) levels of knee extensor and flexor muscles exceeded the EMG level during rest by more than two standard deviations. The net knee flexion moments were calculated from recorded force data and measured moment arm. Reliability for knee angles corresponding with 0.5, 1, 2, 3, and 4 Nm knee net flexion moments was assessed by standard error of measurements (SEM) and smallest detectable difference (SDD). RESULTS: For between day comparison, SEMs were about 5° and SDDs were below 14° for knee angles at 1-4 Nm net knee flexion moments. In SCP children, the knee angle measured at 4 Nm knee flexion moment was not related to the popliteal angle (r = 0.52). The slope at 4 Nm of the knee moment-angle curve in SCP children was significantly higher than that in typically developing children. CONCLUSIONS: The presented knee hand-held dynamometry allows assessment of net knee flexion moment-knee angle characteristics in typically developing and SCP children and can be used to identify clinically relevant changes as a result of treatment. Overall stiffness of structures that contribute to the net knee flexion moment at the knee (i.e. muscles, tendons, ligaments) is elevated in SCP children.