Automatic Lateralization of Temporal Lobe Epilepsy Based on MEG Network Features Using Support Vector Machines

Figure 2

Flowchart showing the network parameter-based determination of TLE lateralization. First, based on the connectivity matrix shown in Figure 1, node-related network parameters (node degree, node efficiency, and node betweenness) were calculated (step ). Node degree is considered a basic and important measure of centrality and represents how strongly one node is interacting, structurally and functionally, with other nodes in the network. Node efficiency is one of the most common measures of integration and can be considered as the average inverse shortest path length. Node betweenness based on the number of shortest paths between nodes is a relatively sensitive measure of centrality. The parameters were then incorporated into feature vectors. Here, we conducted dimension reduction using PCA (step ). The optimal feature vectors were then input into an SVM for classification (step ). Finally, by training the SVM, unilateral TLE could be classified with a high degree of accuracy.