Autism Spectrum Disorder (ASD) is a complicated neurodevelopmental disorder with abnormal brain organization and neural connectivity. Behavioral examination is still used to diagnose it. This study examines how multimodal neuroimaging biomarkers can improve objective, early ASD diagnosis, in line with advances in neurological diagnostics. Anatomical changes and functional brain connection patterns associated with the condition were captured using structural and resting-state functional MRI. Our results show that structural and functional imaging improve diagnostic accuracy compared to unimodal techniques, emphasizing the need to combine biomarkers for effective identification. In particular, brain network organization showed altered connection strength, lower global efficiency, and atypical regional interactions, supporting physiological ideas of poor neuronal integration in ASD. Machine-learning-based analytical methods helped identify the most discriminative biomarkers for these complex brain patterns.

The findings suggest that graph-based representations of brain connections can help diagnose neurological disorders. This study highlights the growing relevance of advanced neuroimaging and computational analysis in neurological illnesses, which aid earlier diagnosis, risk stratification, and individualized intervention. These findings provide guidance on using imaging biomarkers in clinical practice.

Preeti Chouhan is an M.Tech student in medical imaging & informatics at the Indian Institute of Technology Kharagpur, specializing in AI-driven neuroimaging analysis. She has hands-on experience building multimodal machine learning pipelines using MRI and fMRI data to predict neurological disorders. Her work spans deep learning, computer vision, and generative AI, with a focus on healthcare applications and interpretability.