Autism Spectrum Disorder (ASD) Classification based on Facial Expressions using Pre-trained CNN Models and Support Vector Machine (SVM)
Autism Spectrum Disorder (ASD), usually referred to as autism, is a disorder that affects the development of humans characterized by deficits in verbal and non-verbal communication, and restricted and repetitive patterns of behavior. Facial Expression and Morphological Features has been used as a way to detect and classify ASD. The classification of ASD could be automated using Convolutional Neural Network. This research aims to contribute by making it easier for individuals to be diagnosed with ASD and getting early treatment. By creating a CNN-based classifier models for ASD classification, it provides a more efficient and accurate approach to identifying individuals with ASD based on their facial expressions. Convolutional layers of several powerful pre-trained CNN models are used as feature extractors, and the fully connected layers of the corresponding CNN models were replaced by SVM as a classifier.