A Virtual Reality Based System for the Screening and Classification of Autism.

Area of Science:

• Neurodevelopmental disorders
• Artificial intelligence in healthcare
• Virtual reality applications

Background:

• Autism Spectrum Disorders (ASD) are neurodevelopmental conditions affecting communication and social interaction.
• Delayed diagnosis of ASD can impede timely intervention and support, leading to challenges like depression and unemployment.
• Current diagnostic processes often involve long waiting times, highlighting a need for more efficient screening methods.

Purpose of the Study:

• To develop and validate a novel system for Autism Spectrum Disorder screening.
• To utilize virtual reality (VR) social interaction for objective behavioral data collection.
• To employ machine learning for accurate classification of individuals with and without ASD.

Main Methods:

• A VR-based shopping scenario was designed to simulate a common social interaction.
• Behavioral responses of participants during the VR experience were tracked and recorded.
• Machine learning algorithms were applied to analyze recorded behaviors for ASD classification.

Main Results:

• The proposed VR system achieved high accuracy in distinguishing between autistic and typically developing individuals.
• The study demonstrated the feasibility of using VR social interactions for ASD screening.
• The findings suggest a potential for reducing diagnostic delays through this technology.

Conclusions:

• Virtual reality social interaction combined with machine learning offers a promising tool for early ASD screening.
• This approach can contribute to more objective criteria and categorization in mental disorder assessment.
• The developed system has the potential to significantly impact the assessment and support of individuals with ASD.