Integrated Automatic Detection, Classification and Imaging of High Frequency Oscillations With Stereoelectroencephalography.

Baotian Zhao,Baotian Zhao,Yao Wang,Xiu Wang,Kai Zhang,Xiu Wang,Jianguo Zhang,Lin Sang,Jiajie Mo,Jianguo Zhang,Wenhan Hu,Chang Liu,Chang Liu,Wenhan Hu,Yao Wang,Jianguo Zhang,Jiajie Mo,Kai Zhang,Jianguo Zhang,Kai Zhang,Xiaoqiu Shao,Yanshan Ma,Wenhan Hu,Xiaoqiu Shao,Wenhan Hu,Xiaoli Yang,Chao Zhang,Kai Zhang,Chao Zhang

doi:10.3389/fnins.2020.00546

Abstract

ObjectiveDuring presurgical evaluation for focal epilepsy patients, the evidence supporting the use of high frequency oscillations (HFOs) for delineating the epileptogenic zone (EZ) increased over the past decade. This study aims to develop and validate an integrated automatic detection, classification and imaging pipeline of HFOs with stereoelectroencephalography (SEEG) to narrow the gap between HFOs quantitative analysis and clinical application.MethodsThe proposed pipeline includes stages of channel inclusion, candidate HFOs detection and automatic labeling with four trained convolutional neural network (CNN) classifiers and HFOs sorting based on occurrence rate and imaging. We first evaluated the initial detector using an open simulated dataset. After that, we validated our full algorithm in a 20-patient cohort against three assumptions based on previous studies. Classified HFOs results were compared with seizure onset zone (SOZ) channels for their concordance. The receiver operating characteristic (ROC) curve and the corresponding area under the curve (AUC) were calculated representing the prediction ability of the labeled HFOs outputs for SOZ.ResultsThe initial detector demonstrated satisfactory performance on the simulated dataset. The four CNN classifiers converged quickly during training, and the accuracies on the validation dataset were above 95%. The localization value of HFOs was significantly improved by HFOs classification. The AUC values of the 20 testing patients increased after HFO classification, indicating a satisfactory prediction value of the proposed algorithm for EZ identification.ConclusionOur detector can provide robust HFOs analysis results revealing EZ at the individual level, which may ultimately push forward the transitioning of HFOs analysis into a meaningful part of the presurgical evaluation and surgical planning.

Highlights

A majority of seizures can be well controlled by antiepileptic drugs, approximately 30% of patients suffer from uncontrolled seizures despite pharmacotherapy, who are potential candidates for presurgical evaluation and subsequent surgery interventions (Devinsky, 1999; Jobst and Cascino, 2015)
Channel Selection Preimplantation 1 mm isotropic T1 weighted MRI images and the coordinates for each depth electrode contact in individual space, which were determined by coregistering the post implantation computerized tomography image with the preimplantation MRI, and raw SEEG data were collected
Brain extraction was performed by ROBEX (Iglesias et al, 2011) to make a binary brain mask to identify outside brain contacts

Summary

Introduction

A majority of seizures can be well controlled by antiepileptic drugs, approximately 30% of patients suffer from uncontrolled seizures despite pharmacotherapy, who are potential candidates for presurgical evaluation and subsequent surgery interventions (Devinsky, 1999; Jobst and Cascino, 2015). Epileptologists mainly focus on ictal SEEG to reveal SOZ; interictal HFOs have increased in popularity as a promising biomarker for the EZ over the past decade (Bragin et al, 1999; Jacobs et al, 2008). It has been well illustrated and replicated that the rates of HFOs were higher within the SOZ than outside (Worrell et al, 2004; Urrestarazu et al, 2007; Jacobs et al, 2008). A recent meta-analysis indicated a significant relationship between the removal of tissue with high HFOs rates and surgical outcomes (Holler et al, 2015)

Methods

Results

Conclusion