Hexa-Net Framework: A Fresh ADHD-Specific Model for Identifying ADHD Based on Integrating Brain Atlases

Attention Deficit Hyperactivity Disorder (ADHD) is a frequent neurodevelopmental disorder affecting children and adults, which is routinely diagnosed based on subjective observations and behavioural assessments. Recent advancements in neuroimaging, particularly in resting-state functional magnetic resonance imaging (rs-fMRI), have provided a better understanding of the functional brain network impairments linked to ADHD. The human brain naturally consists of resting-state networks (RSNs) that are spatially distinct and functionally homogenous. Therefore, identifying ADHD biomarkers using the human brain’s RSNs is a promising approach. In order to make accurate statistical inferences in brain science, it is necessary to utilize brain atlases for localizing network-of-interest (NoIs). However, locating the spatial components of these RSNs using human brain functional atlases poses challenges due to a lack of disease-specific atlases and atlases concordance issues. This research (1) conducts a study and addresses six RSNs that are frequently referenced in ADHD literature: (Auditory-, Cognitive Control-, Dorsal Attention-, Default Mode-, Sensorimotor-, and Ventral Attention-) Networks (2) Introduces a framework that attempts to enhance the generation of ADHD-specific brain reference, named “Hexa-Net”; This comprehensive approach may improve the reliability and applicability of ADHD studies to a fresh level via segregating and integrating the brain into (NoIs) by evaluating predetermined brain atlases. We hypothesize that the Hexa-Net Model can offer a more precise and unbiased method for identifying ADHD-related impairments. As a result, this framework serves as a practical guide for analyzing biomarkers from rs-fMRI scans to aid in diagnosing ADHD.