Ensembler_fMRI: An Intelligent Approach for the Early Prediction of Autism Disorder

Autism Disorder(AD) During ) is a serious developmental abnormality that affects the behavior and communication of an individual. Under the category of Autism disorder, falls a group of complex and heterogenous disorders i.e., Autism spectrum disorder (ASD) and Attention Deficit Hyperactivity Disorder (ADHD). Due to their significant comorbidity and the mutually reinforcing nature of their symptoms, many disorders can be difficult to diagnose. Traditional diagnosis of AD follows a behavioral dependent approach and often fail to detect overlap or comorbidity between the two disorders, leading to misdiagnosis and an inappropriate treatment. In the existing literature, various automated models based on artificial intelligence(AI) techniques and quantitative measures such as fMRI, eye, and electroencephalography (EEG) have been developed for differentiating ASD from healthy or ADHD from healthy. Among these measures, fMRI has emerged as a potential tool to serve as a biomarker for AD. However, despite having significant overlap and misdiagnosis, very few research have attempted to distinguish ADHD from ASD. Keeping this essence, in this paper, an Ensembler_fMRI framework has been proposed that combines the multiple AI techniques to analyze fMRI data and extract relevant features that can distinguish ADHD from ASD. The proposed framework aims to leverage the available ASD and ADHD databases (ABIDE and ADHD-200) to improve the differential diagnosis between the disorders. The framework achieved an impressive accuracy of 98%, demonstrating its potential as a valuable tool in clinical practice.