Growth assessment of school-age children using dual-task observation

Abstract

Conventional approaches to the assessment of growth among children involve manual evaluation and treat different aspects of growth status separately. In contrast, this study presents an automated method for assessing growth status that considers various aspects of growth simultaneously. We first applied the dual-task paradigm (where two tasks are performed simultaneously) to collect data on anthropometric, kinematic, and cognitive aspects of growth at the same time. With the collected data on a large number of typically developing individuals, we constructed a statistical model of growth features and ages and also estimated participants' ages using regression analysis. By comparing the value for a participant to the average level of performance, we were able to provide an initial judgment of a child's growth status. The experiment results demonstrated that, among children, the growth features developed with age and that the estimation of growth status using this model was feasible.