Identifying Refractory Epilepsy Without Structural Abnormalities by Fusing the Common Spatial Patterns of Functional and Effective EEG Networks

Yuhang Lin,Yan Cui,Dezhong Yao,Peishan Du,Liang Yu,Yang Xia,Yi Liang,Hongze Sun,Ke Chen,Daqing Guo,Zeru Wang

doi:10.1109/tnsre.2021.3071785

Abstract

Drug refractory epilepsy (RE) is believed to be associated with structural lesions, but some RE patients show no significant structural abnormalities (RE-no-SA) on conventional magnetic resonance imaging scans. Since most of the medically controlled epilepsy (MCE) patients also do not exhibit structural abnormalities, a reliable assessment needs to be developed to differentiate RE-no-SA patients and MCE patients to avoid misdiagnosis and inappropriate treatment. Using resting-state scalp electroencephalogram (EEG) datasets, we extracted the spatial pattern of network (SPN) features from the functional and effective EEG networks of both RE-no-SA patients and MCE patients. Compared to the performance of traditional resting-state EEG network properties, the SPN features exhibited remarkable superiority in classifying these two groups of epilepsy patients, and accuracy values of 90.00% and 80.00% were obtained for the SPN features of the functional and effective EEG networks, respectively. By further fusing the SPN features of functional and effective networks, we demonstrated that the highest accuracy value of 96.67% could be reached, with a sensitivity of 100% and specificity of 92.86%. Overall, these findings not only indicate that the fused functional and effective SPN features are promising as reliable measurements for distinguishing RE-no-SA patients and MCE patients but also may provide a new perspective to explore the complex neurophysiology of refractory epilepsy.

Highlights

D RUG refractory epilepsy (RE) patients, characterized by failed treatment with adequate trials of two appropriate and tolerated antiepileptic drugs [1], were believed to be associated with internal lesions in the brain [2]–[4]
The spatial pattern of network (SPN) features of the functional network in the beta band showed to be the most discriminative with 90.00% accuracy, whereas the best performance (80.00%) for the effective network was achieved in the full frequency band
For both functional and effective networks, the spatial features extracted by 4 pairs of SPN filters were most appropriate for classifying RE-no-SA patients and medically controlled epilepsy (MCE) patients in this study

Summary

Introduction

D RUG refractory epilepsy (RE) patients, characterized by failed treatment with adequate trials of two appropriate and tolerated antiepileptic drugs [1], were believed to be associated with internal lesions in the brain [2]–[4]. Structural abnormalities (RE-no-SA) on conventional magnetic resonance imaging (MRI) scans [5]–[7]. In the clinic, these RE-no-SA patients might be misdiagnosed as medically controlled epilepsy (MCE) patients because a majority of MCE patients are reported to have no significant structural abnormalities [1]. These RE-no-SA patients might be misdiagnosed as medically controlled epilepsy (MCE) patients because a majority of MCE patients are reported to have no significant structural abnormalities [1] Such a misdiagnosis could further cause inappropriate treatment and severe physical burden for patients. It is unrealistic to identify RE-no-SA patients with these high-cost multiple testing methods in the clinic, especially in developing countries

Methods

Results

Conclusion