Distinct Cerebellar Contributions to Intrinsic Connectivity Networks

The Bilateral Precuneus as a Potential Neuroimaging Biomarker for Right Temporal Lobe Epilepsy: A Support Vector Machine Analysis

Background and Objective

While evidence has demonstrated that the default-mode network (DMN) plays a key role in the broad-scale cognitive problems that occur in right temporal lobe epilepsy (rTLE), little is known about alterations in the network homogeneity (NH) of the DMN in TLE. In this study, we used the NH method to investigate the NH of the DMN in TLE at rest, and an support vector machine (SVM) method for the diagnosis of rTLE.

Methods

A total of 43 rTLE cases and 42 healthy controls (HCs) underwent resting-state functional magnetic resonance imaging (rs-fMRI). Imaging data were analyzed with the NH and SVM methods.

Results

rTLE patients have a decreased NH in the right inferior temporal gyrus (ITG) and left middle temporal gyrus (MTG), but increased NH in the bilateral precuneus (PCu) and right inferior parietal lobe (IPL), compared with HCs. We found that rTLE had a longer performance reaction time (RT). No significant correlation was found between abnormal NH values and clinical variables of the patients. The SVM results showed that increased NH in the bilateral PCu as a diagnostic biomarker distinguished rTLE from HCs with an accuracy of 74.12% (63/85), a sensitivity 72.01% (31/43), and a specificity 72.81% (31/42).

Conclusion

These findings suggest that abnormal NH of the DMN exists in rTLE, and highlights the significance of the DMN in the pathophysiology of cognitive problems occurring in rTLE, and the bilateral PCu as a neuroimaging diagnostic biomarker for rTLE.

Assessment of functional and structural connectivity between motor cortex and thalamus using fMRI and DWI

Connectivity evaluations have been performed in a noninvasive manner by examining resting state fMRI alongside diffusion-weighted images (DWI). The spatial structures of coherent spontaneous BOLD fluctuations provided the most convincing preliminary evidence that the BOLD signal was predominantly of neuronal origin rather than non-neuronal, artifactual noise. In this study we have shown that in thalamocortical network, the results of functional connectivity analysis and DWI correspond well with each other, thereby providing cross-validation of the two techniques. We have used the resting state fMRI data of 3 subjects with 10 minute resting state functional images via a 3T Siemens scanner. we used cross correlation for functional analysis and reported thalamocortical results with p value=0.01 and cluster size=100, Then showed corresponding tracts connecting premotor cortex and thalamus. In addition, both techniques correspond well to histological delineation and invasive tract tracing, which provides a 'gold standard' validation of the two techniques. The degree of structural connectivity has been shown to correlate with the strength of functional connectivity, thereby providing a potentially straightforward structural explanation for many of the changes in functional connectivity in disease states.