Review of Progress in Diagnostic Studies of Autism Spectrum Disorder Using Neuroimaging.

This review looked at brain imaging studies for autism diagnosis over the past 20 years. Researchers are using brain scans (MRI) combined with computer programs to find biological signs of autism in the brain. Currently, autism is diagnosed through behavioral observations and interviews. The goal is to use brain imaging to help diagnose autism earlier and more accurately. The studies looked at brain thickness, size, and how different brain areas connect with each other.

This review examined two decades of neuroimaging research for autism spectrum disorder (ASD) diagnosis. The authors analyzed studies using structural and functional MRI combined with machine learning approaches to identify biological markers for ASD. They categorized neuroimaging studies into three types: brain thickness, volume, and functional connectivity studies. The review highlights how advanced computational methods are being applied to neuroimaging data to potentially improve early ASD diagnosis beyond current behavioral assessment tools like ADOS and ADI-R.

The authors also developed a lightweight CNN model achieving high classification accuracy, though this represents a single computational approach rather than validated clinical practice.

While neuroimaging combined with machine learning shows promise for identifying biological markers of autism, these approaches remain in early research stages. Current clinical practice should continue relying on established diagnostic tools like ADOS and ADI-R. Future research may lead to complementary biological diagnostic methods.

This is a review article without reported sample sizes or systematic methodology details. The high CNN accuracy represents a single model's performance rather than validated clinical practice. No information provided about study selection criteria, quality assessment, or generalizability of findings across populations.

This review article summarizes the recent advances in the diagnostic studies of autism spectrum disorders (ASDs) considering some of the most influential research articles from the last two decades. ASD is a heterogeneous neurodevelopmental disorder characterized by abnormalities in social interaction, communication, and behavioral patterns as well as some unique strengths and differences. The current diagnosis systems are based on autism diagnostic observation schedule (ADOS) or autism diagnostic interview-revised (ADI-R), but biological markers are also important for an effective diagnosis of ASDs. The amalgamation of neuroimaging techniques, such as structural and functional magnetic resonance imaging (sMRI and fMRI), with machine-learning and deep-learning approaches helps throw new light on typical biological markers of ASDs at the early stage of life.

To assess the performance of a deep neural network, we develop a light-weighted CNN model for ASD classification. The overall accuracy, precision, and F1-score of the proposed model are 99.92%, 99.93% and 99.92%, respectively. All the neuroimaging studies we have reviewed can be divided into 3 categories, viz. thickness, volume and functional connectivity-based studies. We conclude with a discussion of the major findings of considered studies and promising directions for future research in this field.