Detecting autism spectrum disorder in children using eye tracking and machine learning

The number of people with autism has increased in recent years. The disorder is characterized by cognitive inexperience that affects the nervous system and children's growth and development. It appears frequently in the first few years of a child's life and has a negative effect on visual behavior. Machine learning (ML) has a lot of potential to classify this problem. In this paper, a program for diagnosing autism was built by eye in the Python language and using machine-learning technology to support vector machines with the use of a computer vision algorithm known as Viola-Jones. The algorithm has achieved an accuracy of 89%. The proposed algorithm supports vector machines to correctly predict the autism spectrum-using eye tracking technology. We used it to evaluate group images of children and used it to identify the eye. Differences that are difficult to recognize with the naked eye can be detected using machine learning methods, in which a computer system has been created that uses images as input and detects autism. The algorithm achieved a vector machine support accuracy of. We conclude that machine learning gives more accurate, rapid, and effective results in detecting autism compared to traditional methods.