Co-Design of a Virtual Reality Multiplayer Adventure Game for Adolescents With Autism Spectrum Disorder: Mixed Methods Study.

BACKGROUND: Virtual reality (VR) adventure games can offer ideal technological solutions for training social skills in adolescents with autism spectrum disorder (ASD), leveraging their support for multisensory and multiplayer interactions over distance, which may lower barriers to training access and increase user motivation. However, the design of VR-based game environments for social skills training is still understudied and deserves the deployment of an inclusive design approach to ensure its acceptability by target users. OBJECTIVE: We aimed to present the inclusive design process that we had followed to develop the Zentastic VR adventure game to foster social skills training in adolescents with ASD and to investigate its feasibility as a training environment for adolescents. METHODS: The VR game supports multiplayer training sessions involving small groups of adolescents and their therapists, who act as facilitators. Adolescents with ASD and their therapists were involved in the design and in an explorative acceptability study of an initial prototype of the gaming environment, as well as in a later feasibility multisession evaluation of the VR game final release. RESULTS: The feasibility study demonstrated good acceptability of the VR game by adolescents and an enhancement of their social skills from baseline to posttraining. CONCLUSIONS: The findings provide preliminary evidence of the benefits that VR-based games can bring to the training of adolescents with ASD and, potentially, other neurodevelopmental disorders.

EYE-C: Eye-Contact Robust Detection and Analysis during Unconstrained Child-Therapist Interactions in the Clinical Setting of Autism Spectrum Disorders.

The high level of heterogeneity in Autism Spectrum Disorder (ASD) and the lack of systematic measurements complicate predicting outcomes of early intervention and the identification of better-tailored treatment programs. Computational phenotyping may assist therapists in monitoring child behavior through quantitative measures and personalizing the intervention based on individual characteristics; still, real-world behavioral analysis is an ongoing challenge. For this purpose, we designed EYE-C, a system based on OpenPose and Gaze360 for fine-grained analysis of eye-contact episodes in unconstrained therapist-child interactions via a single video camera. The model was validated on video data varying in resolution and setting, achieving promising performance. We further tested EYE-C on a clinical sample of 62 preschoolers with ASD for spectrum stratification based on eye-contact features and age. By unsupervised clustering, three distinct sub-groups were identified, differentiated by eye-contact dynamics and a specific clinical phenotype. Overall, this study highlights the potential of Artificial Intelligence in categorizing atypical behavior and providing translational solutions that might assist clinical practice.

Is Smiling the Key? Machine Learning Analytics Detect Subtle Patterns in Micro-Expressions of Infants with ASD.

Time is a key factor to consider in Autism Spectrum Disorder. Detecting the condition as early as possible is crucial in terms of treatment success. Despite advances in the literature, it is still difficult to identify early markers able to effectively forecast the manifestation of symptoms. Artificial intelligence (AI) provides effective alternatives for behavior screening. To this end, we investigated facial expressions in 18 autistic and 15 typical infants during their first ecological interactions, between 6 and 12 months of age. We employed Openface, an AI-based software designed to systematically analyze facial micro-movements in images in order to extract the subtle dynamics of Social Smiles in unconstrained Home Videos. Reduced frequency and activation intensity of Social Smiles was computed for children with autism. Machine Learning models enabled us to map facial behavior consistently, exposing early differences hardly detectable by non-expert naked eye. This outcome contributes to enhancing the potential of AI as a supportive tool for the clinical framework.