Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent and heterogeneous neurodevelopmental disorder, which is diagnosed using subjective symptom reports. Machine learning classifiers have been utilized to assist in the development of neuroimaging-based biomarkers for objective diagnosis of ADHD. However, existing basic model-based studies in ADHD report suboptimal classification performances and inconclusive results, mainly due to the limited flexibility for each type of basic classifier to appropriately handle multi-dimensional source features with varying properties. This study applied ensemble learning techniques (ELTs), a meta-algorithm that combine several basic machine learning models into one predictive model in order to decrease variance, bias, or improve predictions, in multimodal neuroimaging data collected from 72 young adults, including 36 probands (18 remitters and 18 persisters of childhood ADHD) and 36 group-matched controls. All currently available optimization strategies for ELTs (i.e., voting, bagging, boosting and stacking techniques) were tested in a pool of semifinal classification results generated by seven basic classifiers. The high-dimensional neuroimaging features for classification included regional cortical gray matter (GM) thickness and surface area, GM volume of subcortical structures, volume and fractional anisotropy of major white matter fiber tracts, pair-wise regional connectivity and global/nodal topological properties of the functional brain network for cue-evoked attention process. As a result, the bagging-based ELT with the base model of support vector machine achieved the best results, with significant improvement of the area under the receiver of operating characteristic curve (0.89 for ADHD vs. controls and 0.9 for ADHD persisters vs. remitters). Features of nodal efficiency in right inferior frontal gyrus, right middle frontal (MFG)-inferior parietal (IPL) functional connectivity, and right amygdala volume significantly contributed to accurate discrimination between ADHD probands and controls; higher nodal efficiency of right MFG greatly contributed to inattentive and hyperactive/impulsive symptom remission, while higher right MFG-IPL functional connectivity strongly linked to symptom persistence in adults with childhood ADHD. Considering their improved robustness than the commonly implemented basic classifiers, findings suggest that ELTs may have the potential to identify more reliable neurobiological markers for neurodevelopmental disorders.
17 June 2014 in Frontiers in Integrative Neuroscience
Convergence insufficiency (CI) is a prevalent binocular vision disorder with symptoms that include double/blurred vision, eyestrain, and headaches when engaged in reading or other near work. Randomized clinical trials support that Office-Based Vergence and Accommodative Therapy with home reinforcement leads to a sustained reduction in patient symptoms. However, the underlying neurophysiological basis for treatment is unknown. Functional activity and vergence eye movements were quantified from seven binocularly normal controls (BNC) and four CI patients before and after 18 hours of vergence training. An fMRI conventional block design of sustained fixation versus vergence eye movements stimulated activity in the frontal eye fields (FEF), the posterior parietal cortex (PPC) and the cerebellar vermis (CV). Comparing the CI patients’ baseline measurements to the post vergence training data sets with a paired t-test revealed the following: 1) the percent change in the BOLD signal in the FEF, PPC and CV significantly increased (p<0.02), 2) the peak velocity from 4° symmetrical convergence step responses increased (p<0.01) and 3) patient symptoms assessed using the CI Symptom Survey (CISS) improved (p<0.05). CI patient measurements after vergence training were more similar to levels observed within BNC. A regression analysis revealed the peak velocity from BNC and CI subjects before and after vergence training was significantly correlated to the percent BOLD signal change within the FEF, PPC and CV (r=0.6;p<0.05). Results have clinical implications for understanding the behavioral and neurophysiological changes after vergence training in patients with CI, which may lead to the sustained reduction in visual symptoms.