Detecting Autism Spectrum Disorder through Eye-Tracking: A CNN LSTM Model

In this study, Specific visual attention patterns in children with Autism Spectrum Disorder (ASD), including reduced interest in faces and increased interest in objects. These patterns can have a high effect on their capacity to understand social cues needed for communication and cognitive growth. Recognizing this important because it helps us create better early interventions. Eye-tracking data analysis through deep learning algorithms has become a useful method for studying such minor eye movement variations which might not be detected by traditional techniques. The research concentrated only on kids aged between two to six years and it found out that those who had ASD consistently spent lesser periods looking at other people’s faces than their counterparts who were typically developing. This lack of focus on facial features could hinder one’s ability to identify emotions, make eye contact or even participate in social interactions all of which are basic requirements for relationships establishment hence deep learning models can be used by researchers as they enable early diagnosis recognition with improved accuracy thereby providing more appropriate interventions.The outcomes of this research could be beneficial to practitioners in the medical field as well as educators since they will be able to come up with individualized therapeutic approaches geared towards improving both learning and socio skills among young individuals living with autism spectrum disorder (ASD).