AI-powered autism detection using facial expressions and eye tracking features

Autism spectrum disorder (ASD) is a neurodevelopmental condition that affects communication, social interaction, and behavior. Early detection and intervention are crucial for improving outcomes, but traditional diagnostic methods often rely on subjective assessments and lengthy observations. This project introduces an AI-powered solution for autism detection, leveraging facial expressions and eye-tracking data. The system utilizes advanced machine learning algorithms to identify patterns and characteristics that are indicative of autism.

The project integrates facial expression recognition using deep learning models such as Convolutional Neural Networks (CNN), MobileNet, DenseNet, and ResNet. These models are trained on a dataset of facial expressions from children with autism and non-autistic children. Additionally, eye-tracking features, such as gaze points and fixation durations, are analyzed using machine learning techniques, including Support Vector Machines (SVM), RandomForestClassifier, CNN, and AdaBoostClassifier.

By using these two distinct modalities, the system aims to provide an efficient, accessible, and non-invasive method for autism detection. The web-based platform enables users to upload facial expressions and eye-tracking data, which are then classified to provide predictions about autism. This tool has the potential to assist professionals in identifying early signs of autism, providing timely support for children and their families. The system is built with a user-friendly interface, using HTML, CSS, and JavaScript for the front-end, and Python with Flask for the back-end.

Keywords: Autism, Early detection, Facial expressions, Eye-tracking, Machine learning, CNN, MobileNet, DenseNet, ResNet, SVM.