Background: Neurodevelopmental disorders (NDs) are characterized by heterogeneity, complexity, and interactions among multiple domains with long-lasting effects in adulthood. Identifying and assessing children at risk for NDs is crucial. However, many children remain misdiagnosed, missing out on opportunities for effective interventions. Digital tools can help clinicians assist and identify NDs. The concept of using serious games to enhance healthcare has gained attention among a growing group of scientists, entrepreneurs, and clinicians.Objective: This study aims to define the main principles of automated mobile NDs detection.
Methods:In this study, 229 children aged 4 to 12 participated after responding to an open call. Children with neurodevelopmental disorders, other conditions, or those taking medication were not included in the study. Parents provided consent for their children to participate. Children interacted face-to-face with a mobile serious game called 'Apsou'. The game was designed to measure 18 primary domains including speech, language, psychomotor, cognitive, psychoemotional, and hearing abilities. The measurements were based on the children's performance in specific tasks such as gameplay and verbal responses. The data collected was analyzed using descriptive statistics and Principal Component Analysis (PCA).Results: A sample of 229 typically developing preschoolers and early school-aged children played the Apsou mobile serious game for automated detection of NDs. Performing a PCA, the findings identified five main components accounting for about 80% of the data variability that potentially have significant prognostic implications for a safe diagnosis of neurodevelopmental disorders. Varimax rotation explained 61.44% of the total variance. The results underscore key theoretical principles crucial for the automated detection of Neurodevelopmental Disorders. These principles encompass communication skills, speech and language development, vocal processing, cognitive and sensory functions, and visual-spatial skills. The components identified in this study align with the theoretical principles of typically developmental domains described in other studies, further validating the robustness of our findings.
Conclusions:The findings of this study underscore the core developmental domains, crucial for a comprehensive model leading to highly accurate predictions and classifications for automated screening, diagnosis, prognosis, or intervention planning in NDs. Importantly, these findings offer valuable insights for creating machine learning applications to support clinical decision-making. Clinical Trial: 18435-15.05.2020 approved by the