Neurocognitive Assessment Models for Early Diagnosis of Autism Spectrum Disorders in Schools

Abstract

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that is typified by impairment of social communication, behavioral patterns and cognitive functioning. Early diagnosis is important to effective intervention, but the traditional methods of diagnosis are usually subjective, time-consuming and can hardly be scaled to a school setting. This paper offers a new neurocognitive assessment model to identify ASD in early stages and uses multimodal measures of data, such as electroencephalography (EEG) signals, behavioral, and cognitive performance variables. The architecture uses the high-order machine learning and deep learning models, including Support Vector Machines (SVM), Random Forest, Convolutional Neural Networks (CNN), and ensemble models, to solve the complicated neurocognition patterns and formulate precise forecasts. The system proposed will add the data preprocessing, feature extraction, dimensionality reduction, and classification modules to maintain a robust and efficient performance. The accuracy, precision, recall and F1-score of the model are shown to be high and experimental results indicate that it is more accurate and precise than the traditional diagnostic methods and current machine learning models. The combination of the multimodal data provides a great boost in the reliability of the diagnostic process and allows detecting the at-risk students at an early stage. Altogether, the framework suggested has the potential to become a great solution to the problem of the early diagnosis of ASD as it helps to fill the gap between clinical evaluation and practical application in the classroom.