Automated Source Estimation of Scalp EEG Epileptic Activity Using eLORETA Kurtosis Analysis
Automated Source Estimation of Scalp EEG Epileptic Activity Using eLORETA Kurtosis Analysis
复制标题
DOI：
10.1159/000495522
复制
发表时间：
2018-01-01
影响因子：
3.2
通讯作者：
Kinoshita, Toshihiko
中科院分区：
文献类型：
--
作者：
Ikeda, Shunichiro;Ishii, Ryouhei;Kinoshita, Toshihiko
目的：eLORETA（精确低分辨率脑电磁断层扫描）是由Pascual-Marqui等人[Int J Psychophysiol. 1994 Oct; 18（1）：49-65]，用于通过脑电图（EEG）数据对大脑中的电流源密度的三维表示。峰度分析允许识别大脑中的尖峰活动。在本研究中，我们重点评价了eLORETA峰度分析的可靠性。为此，对3例部分性发作的癫痫患者进行了eLORETA峰态源定位和eLORETA电流源密度（CSD）分析。方法：采用19通道数字化脑电系统进行脑电测量。我们将带通滤波器设置为传统的频带设置（1-4、4-8、8-15、15-30和30-60 Hz）以及5-10和20-70 Hz，并进行eLORETA峰度分析，以比较局灶性癫痫患者阵发性活动的源定位与EEG数据的视觉解释和eLORETA的CSD分析。结果如下：通过从20 Hz到70 Hz的带通滤波和传统频带设置预处理的EEG数据的eLORETA峰度分析未显示与eLORETA的CSD分析结果兼容的任何离散阵发性源活动。在所有3例病例中，在5-10 Hz下过滤的eLORETA峰度分析均显示θ波段的阵发性活动，均与目视检查结果和CSD分析结果一致。讨论内容：我们的研究结果表明，eLORETA峰度分析的EEG数据可能是有用的，为癫痫患者的棘状阵发性活动源的识别。由于EEG被广泛应用于癫痫的临床实践，eLORETA峰度分析是一种很有前途的方法，可以应用于癫痫活动映射。(c)2019 S. Karger AG，巴塞尔
Objectives: eLORETA (exact low-resolution brain electromagnetic tomography) is a technique created by Pascual-Marqui et al. [Int J Psychophysiol. 1994 Oct; 18(1): 49-65] for the 3-dimensional representation of current source density in the brain by electroencephalography (EEG) data. Kurtosis analysis allows for the identification of spiky activity in the brain. In this study, we focused on the evaluation of the reliability of eLORETA kurtosis analysis. For this purpose, the results of eLORETA kurtosis source localization of paroxysmal activity in EEG were compared with those of eLORETA current source density (CSD) analysis of EEG data in 3 epilepsy patients with partial seizures. Methods: EEG was measured using a digital EEG system with 19 channels. We set the bandpass filter at traditional frequency band settings (1-4, 4-8, 8-15, 15-30, and 30-60 Hz) and 5-10 and 20-70 Hz and performed eLORETA kurtosis to compare the source localization of paroxysmal activity with that of visual interpretation of EEG data and CSD analysis of eLORETA in focal epilepsy patients. Results: The eLORETA kurtosis analysis of EEG data preprocessed by bandpass filtering from 20 to 70 Hz and traditional frequency band settings did not show any discrete paroxysmal source activity compatible with the results of CSD analysis of eLORETA. In all 3 cases, eLORETA kurtosis analysis filtered at 5-10 Hz showed paroxysmal activities in the theta band, which were all consistent with the visual inspection results and the CSD analysis results. Discussion: Our findings suggested that eLORETA kurtosis analysis of EEG data might be useful for the identification of spiky paroxysmal activity sources in epilepsy patients. Since EEG is widely used in the clinical practice of epilepsy, eLORETA kurtosis analysis is a promising method that can be applied to epileptic activity mapping. (c) 2019 S. Karger AG, Basel