Table 1.
Patient characteristics.
Table 2.
Differences between patients and healthy controls regarding theta band network characteristics.
Figure 1.
Theta band PLI and network characteristics for patients and healthy controls.
Parameters were averaged for each sensor on a group level and displayed on a helmet-shaped surface to show global patterns of differences between patient groups. Note that particularly in LGG patients, theta band clustering and participation coefficients show global alterations irrespective of local PLI values. Abbreviations: CTL = healthy controls; LGG = low-grade glioma patients; HGG = high-grade glioma patients; NGL = non-glioma patients; PLI = phase lag index; Cw,i* = nodal clustering coefficient; Lw,i* = nodal path length; = within-module degree z-score;
= participation coefficient. *In the analysis we use normalized average weighted clustering coefficient (Cw/Cws) and normalized average weighted shortest path length (Lw/Lws) instead of the unnormalized values for each vertex i, Cw,i and Lw,i which are visualized here. Cw/Cws and Lw/Lws are calculated by first averaging over nodes and then dividing Cw and Lw by a reference value Cws and Lws, in order to get normalized values. However, this normalization does not affect the spatial distribution of Cw,i and Lw,i, and therefore the original data is presented.
Figure 2.
Example of theta band connection differences between a LGG patient and a HGG patient, both suffering from a tumor located in the right frontal lobe.
The upper images show T2-weighted MRI images of the tumor. The lower images show theta band PLI levels (background colors; red colors represent high PLI levels, blue colors represent low PLI levels). Note that the tumor region seems to have the highest theta band PLI. The colored lines represent connections between sensors, each color representing another module. Connections are shown when their strength passes an arbitrary threshold chosen for optimal connection visualization. In HGG patients, only few connections exist above the threshold. Note that especially connections to the tumor region in LGG patients pass the threshold. However, two other modules are also clearly shown that are not found in the HGG patient, suggesting that the differences between LGG and HGG patients networks are not restricted to the tumor region.
Figure 3.
Theta band synchronizability and seizure frequency in low grade glioma patients.
Note that seizure frequency is plotted on a logarithmic scale. See tables S4 and S5 for seizure frequency and synchronizability values for each patient.
Figure 4.
Theta band synchronizability and attention as measured by Stroop tests.
Attention scores are presented as z-scores gained by comparison with healthy controls matched for age, gender and educational level. See table S4 for attention scores and synchronizability values for each patient.
Table 3.
Overview of MEG functional connectivity studies on lesional epilepsy patients.