Abnormal structural and functional network topological properties associated with left prefrontal, parietal, and occipital cortices significantly predict childhood TBI-related attention deficits: A semi-supervised deep learning study.

Introduction: Traumatic brain injury (TBI) is a major public health concern in children. Children with TBI have elevated risk in developing attention deficits. Existing studies have found that structural and functional alterations in multiple brain regions were linked to TBI-related attention deficits in children. Most of these existing studies have utilized conventional parametric models for group comparisons, which have limited capacity in dealing with large-scale and high dimensional neuroimaging measures that have unknown nonlinear relationships. Nevertheless, none of these existing findings have been successfully implemented to clinical practice for guiding diagnoses and interventions of TBI-related attention problems. Machine learning techniques, especially deep learning techniques, are able to handle the multi-dimensional and nonlinear information to generate more robust predictions. Therefore, the current research proposed to construct a deep learning model, semi-supervised autoencoder, to investigate the topological alterations in both structural and functional brain networks in children with TBI and their predictive power for post-TBI attention deficits. Methods: Functional magnetic resonance imaging data during sustained attention processing task and diffusion tensor imaging data from 110 subjects (55 children with TBI and 55 group-matched controls) were used to construct the functional and structural brain networks, respectively. A total of 60 topological properties were selected as brain features for building the model. Results: The model was able to differentiate children with TBI and controls with an average accuracy of 82.86%. Functional and structural nodal topological properties associated with left frontal, inferior temporal, postcentral, and medial occipitotemporal regions served as the most important brain features for accurate classification of the two subject groups. Post hoc regression-based machine learning analyses in the whole study sample showed that among these most important neuroimaging features, those associated with left postcentral area, superior frontal region, and medial occipitotemporal regions had significant value for predicting the elevated inattentive and hyperactive/impulsive symptoms. Discussion: Findings of this study suggested that deep learning techniques may have the potential to help identifying robust neurobiological markers for post-TBI attention deficits; and the left superior frontal, postcentral, and medial occipitotemporal regions may serve as reliable targets for diagnosis and interventions of TBI-related attention problems in children.

Distinct Thalamic and Frontal Neuroanatomical Substrates in Children with Familial vs. Non-Familial Attention-Deficit/Hyperactivity Disorder (ADHD).

Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent, inheritable, and heterogeneous neurodevelopmental disorder. Children with a family history of ADHD are at elevated risk of having ADHD and persisting its symptoms into adulthood. The objective of this study was to investigate the influence of having or not having positive family risk factor in the neuroanatomy of the brain in children with ADHD. Cortical thickness-, surface area-, and volume-based measures were extracted and compared in a total of 606 participants, including 132, 165, and 309 in groups of familial ADHD (ADHD-F), non-familial ADHD (ADHD-NF), and typically developed children, respectively. Compared to controls, ADHD probands showed significantly reduced gray matter surface area in the left cuneus. Among the ADHD subgroups, ADHD-F showed significantly increased gray matter volume in the right thalamus and significantly thinner cortical thickness in the right pars orbitalis. Among ADHD-F, an increased volume of the right thalamus was significantly correlated with a reduced DSM-oriented t-score for ADHD problems. The findings of this study may suggest that a positive family history of ADHD is associated with the structural abnormalities in the thalamus and inferior frontal gyrus; these anatomical abnormalities may significantly contribute to the emergence of ADHD symptoms.

Abnormal Functional Network Topology and Its Dynamics during Sustained Attention Processing Significantly Implicate Post-TBI Attention Deficits in Children.

Traumatic brain injury (TBI) is highly prevalent in children. Attention deficits are among the most common and persistent post-TBI cognitive and behavioral sequalae that can contribute to adverse outcomes. This study investigated the topological properties of the functional brain network for sustained attention processing and their dynamics in 42 children with severe post-TBI attention deficits (TBI-A) and 47 matched healthy controls. Functional MRI data during a block-designed sustained attention task was collected for each subject, with each full task block further divided into the pre-, early, late-, and post-stimulation stages. The task-related functional brain network was constructed using the graph theoretic technique. Then, the sliding-window-based method was utilized to assess the dynamics of the topological properties in each stimulation stage. Relative to the controls, the TBI-A group had significantly reduced nodal efficiency and/or degree of left postcentral, inferior parietal, inferior temporal, and fusiform gyri and their decreased stability during the early and late-stimulation stages. The left postcentral inferior parietal network anomalies were found to be significantly associated with elevated inattentive symptoms in children with TBI-A. These results suggest that abnormal functional network characteristics and their dynamics associated with the left parietal lobe may significantly link to the onset of the severe post-TBI attention deficits in children.