Autism Spectrum Disorder (ASD) is a complex neurological developmental disability that appears during early childhood. Conventional ASD diagnostic techniques rely on behavioural observations, characteristics, and clinical interviews. To overcome these limitations, numerous machine learning (ML) and Deep Learning (DL) techniques have been used to assist physicians. For the past three decades, biomedical images have been employed to diagnose neurodevelopmental disorders. The functional Magnetic Resonance (MR) images used in this study. This paper proposes a novel machine learning framework to classify ASD control from healthy controls. The proposed framework consists of two stages. In the first stage, an enhanced Convolutional Neural Network (CNN) is proposed to extract features. In the second stage, the extracted features are given to the machine learning classifiers. The proposed method is tested on the 1112 fMRI images. A total of 539 ASD participants and 573 healthy controls are included in this study. A total of 17 datasets from the ABIDE website are used. These datasets are collected from various international medical laboratories. The proposed framework outperforms the existing methods. The proposed algorithm achieved 92.45% across the entire ABIDE dataset and 98.61% on the individual dataset.
Yugander et al. (Thu,) studied this question.