Introduction: Attention deficit hyperactivity disorder (ADHD) is a heritable, chronic, neurobehavioral disorder among children and adolescents. It is estimated that this condition is diagnosed in 5% of American preschool and school-aged children. Children or adolescents with ADHD are at high risk of failing in academics and in building social relationships, which can result in financial hardships for families and create a tremendous burden on society. The diagnosis of ADHD remains challenging. To date, behavior-based tests are the standard clinical approach to diagnosing ADHD.
To aid the clinical diagnosis of ADHD, brain functional connectome derived from advanced neuroimaging data (e.g., resting-state functional MRI and diffusion tensor imaging) have been used distinguishing those with ADHD from control subjects. Brain functional connectome is a comprehensive mapping of brain functional network connections. The constructed connectome spans spatial scales ranging from individual voxels to an aggregation of voxels (i.e., parcels or parcellations of interests). Among recent developments is the consensus that although network analyses at multi-scale level may be redundant, they could also provide supplementary information for complementary understanding and depicting of networks within different brain regions as well as across the entire brain. However, the most studies on ADHD diagnosis using functional connectome lack the consideration that brain networks are fundamentally multiscale entities.
Recently, we have proposed a multichannel, deep neural network (mcDNN) model analyzing multi-scale functional connectome data. (Chen, Li et al. 2019) The proposed mcDNN, which fuses multiscale brain functional connectome data, improved performance compared with the use of a single scale in ADHD diagnosis. Using a feature-ranking strategy, we found that several decision-making and recognition regions, including middle frontal gyrus, inferior orbitofrontal cortex, and fusiform gyrus, significantly contributed to ADHD diagnosis.
Chen M., Li H., Wang J., Dillman J.R., Parikh N.A., He L., (2019) A multi-channel deep neural network model analyzing multiscale functional brain connectome data for ADHD detection. Radiology, Artificial Intelligence. 2019;2(1):e190012. PMC Journal