RESUMO
Gray/white matter contrast (GWC) decreases with aging and has been found to be a useful MRI biomarker in Alzheimer's disease (AD), but its utility in Parkinson's disease (PD) patients has not been investigated. The aims of the study were to test whether GWC is sensitive to aging changes in PD patients, if PD patients differ from healthy controls (HCs) in GWC, and whether the use of GWC data would improve the sensitivity of cortical thickness analyses to differentiate PD patients from controls. Using T1-weighted structural images, we obtained individual cortical thickness and GWC values from a sample of 90 PD patients and 27 controls. Images were processed with the automated FreeSurfer stream. GWC was computed by dividing the white matter (WM) by the gray matter (GM) values and projecting the ratios onto a common surface. The sample characteristics were: 52 patients and 14 controls were males; mean age of 64.4 ± 10.6 years in PD and 64.7 ± 8.6 years in controls; 8.0 ± 5.6 years of disease evolution; 15.6 ± 9.8 UPDRS; and a range of 1.5-3 in Hoehn and Yahr (H&Y) stage. In both PD and controls we observed significant correlations between GWC and age involving almost the entire cortex. When applying a stringent cluster-forming threshold of p < 0.0001, the correlation between GWC and age also involved the entire cortex in the PD group; in the control group, the correlation was found in the parahippocampal gyrus and widespread frontal and parietal areas. The GWC of PD patients did not differ from controls', whereas cortical thickness analyses showed thinning in temporal and parietal cortices in the PD group. Cortical thinning remained unchanged after adjusting for GWC. GWC is a very sensitive measure for detecting aging effects, but did not provide additional information over other parameters of atrophy in PD.
RESUMO
There is growing interest in the potential of neuroimaging to help develop non-invasive biomarkers in neurodegenerative diseases. In this study, connection-wise patterns of functional connectivity were used to distinguish Parkinson's disease patients according to cognitive status using machine learning. Two independent subject samples were assessed with resting-state fMRI. The first (training) sample comprised 38 healthy controls and 70 Parkinson's disease patients (27 with mild cognitive impairment). The second (validation) sample included 25 patients (8 with mild cognitive impairment). The Brainnetome atlas was used to reconstruct the functional connectomes. Using a support vector machine trained on features selected through randomized logistic regression with leave-one-out cross-validation, a mean accuracy of 82.6% (p < 0.002) was achieved in separating patients with mild cognitive impairment from those without it in the training sample. The model trained on the whole training sample achieved an accuracy of 80.0% when used to classify the validation sample (p = 0.006). Correlation analyses showed that the connectivity level in the edges most consistently selected as features was associated with memory and executive function performance in the patient group. Our results demonstrate that connection-wise patterns of functional connectivity may be useful for discriminating Parkinson's disease patients according to the presence of cognitive deficits.
