JMIR Publications

ABSTRACT

Background:

Approximately 17% of children in the United States have been diagnosed with a developmental or neurological disorder that affects upper-limb (UL) movements needed for completing activities of daily living. Gold-standard laboratory assessments of the UL are objective and precise but may not be portable, while clinical assessments can be time-intensive. We developed MoEvGame, a mobile health (mHealth) gamification software system for the iPad, as a potential advanced technology to assess UL motor functions.

Objective:

This feasibility study examines whether MoEvGame can assess children’s whole-limb movement, fine motor skills, manual dexterity, and bimanual coordination. The specific aims were to i) design and develop a novel mHealth gamified software tools to examine theory-driven features of UL movement, ii) analyze spatiotemporal game data with new algorithms and statistical techniques to quantify movement performance as a parameter of speed, accuracy, and precision, and iii) validate assessment methods with healthy participants from school.

Methods:

Elementary school children (N=31, median age=9.0 years, and age range=4.0-14.0) participated by playing five games. The game tasks were focused on key features of skilled motor control: (i) whole limb reaching, (ii) fine motor control and manual dexterity, and (iii) bilateral coordination. Spatiotemporal game data were transferred and stored in a cloud-based data management server for further processing and analysis. We applied change point detection (i.e., Pruned Exact Linear Time method), signal processing techniques, and other algorithms to calculate movement speed and accuracy from spatiotemporal parameters. Different statistical methods (i.e., Pearson correlation, mean, standard deviation, p-value, 95% confidence interval) were used to compare speed-accuracy tradeoffs and evaluate the relationship between age and motor performance.

Results:

A negative correlation was identified between speed and accuracy in the whole limb movement (r= -0.30 to -0.42). Significant relationships between age and whole limb/manual dexterity performance were found: older participants exhibited lower errors with faster completion times compared to younger participants. Significant differences in bimanual coordination were found related to phase synchronization (In-phase Congruent, μ=28.85 σ=18.97 vs Anti-phase Congruent, μ=112.64 σ=25.82 and In-phase Mirrored, μ=23.78 σ=16.07 vs Anti-phase Mirrored μ=121.39 σ=28.19). Moreover, the average speed (RPS) and travel distance (m) of the in-phase mode were significantly higher than those of the anti-phase coordination.

Conclusions:

Results of this feasibility study show that spatiotemporal data captured from the mHealth application can quantify motor performance. Moving beyond traditional assessments, MoEvGame incorporates gamification into ubiquitous and accessible technology as a fast, flexible, and objective tool for UL motor assessment.