27904
A New Ipad Game for Ecological Motor Assessment of Children: Bespoke Wearable and Smart Tablet Engineering for Autism-Friendly Assessment.
Increasing evidence suggests motor control disruption in individuals with autism spectrum disorder (ASD) may be a cardinal symptom. Thus, recording sub-second kinematics may lead to the identification and objective quantification of motor characteristics specific to autism, potentially allowing a novel instrument to aid early screening and diagnosis. Kinematic analysis has traditionally involved expensive and laborious optical motion tracking systems. With the recent advent of wearable inertial movement units sensors (IMUs), new possibilities are afforded for accessible serious game platforms for the measurement of goal-directed hand movements to a target, with many repetitions for high-grade statistical analysis.
Objectives:
This study: To engineer a serious game paradigm for fun, fast, and ecologically valid computational kinematic assessment of possible autism-specific goal-directed arm motor patterns in children.
Methods:
Equipment. An integrated platform of (i) a bespoke engineered wearable IMU with tri-axial accelerometer and gyroscope, (ii) smart tablet device (iPad mini), and (iii) a bespoke engineered forceplate (fPad).
Wearable sensor was engineered with an Arduino component protected within a 3D-printed acrylic container strapped to the child’s wrist (60g). The fPad contained 4 strain gauges on metal legs bolted to the underside of an acrylic plate allowing measurement of applied forces during gameplay. The iPad was situated upon the fPad and pre-loaded with a browser-based game. The iPad and fPad were encased within a handcrafted balsa wood case with sanded edged, painted matt-black for maximum screen contrast.
Data were sampled at 80Hz, sent to an on-site server over Wi-Fi and synchronised.
The game, Bubble-Pop, encouraged regular arcing motions to hit targets in corners of the screen. Targets were bubbles 2.7 cm in diameter, with centres distanced at 11.2cm horizontally or 7.2cm vertically from each other. Level 1 has 4 bubbles, one in each corner, requiring 40 pops with choice of popping order. Level 2 has 2 bubbles, in the top corners, requiring 30 pops with 1 target at a time. Levels are played 3 times to generate 210 windows of movement over 7 minutes.
Results:
Motor kinematic profiles generated from the wearable, iPad, and fPad include: hand acceleration and jerk, duration, impact force to target acceleration, target accuracy and precision, target contact duration, touch gesture variables, and variability of these features. Motor profiles are compared within and between groups for improved characterisation of children’s motor pattern.
Conclusions:
Sub-second motor kinematic data is achievable using a wearable, iPad, and fPad serious game solution. Children enjoy the game and repeat goal-directed movements many times per trial (>210 movements in ca. 7 mins) providing high-quality data for robust statistical analyses of motor kinematics. In sum, we have designed a new, fun bespoke technological tool to facilitate motor assessment of children with or without ASD. It may ultimately lead to improved diagnosis of ASD in early childhood with cheaper, simpler, faster and more accessible assessment of movements in children than currently available. In future work we will employ this technology to identify motor patterns unique to ASD in a large cohort (n>100) of children 2-5 years old.