EEG-driven RNN Classification for Prognosis of Neurodegeneration in At-Risk Patients
EEG-driven RNN Classification for Prognosis of Neurodegeneration in At-Risk Patients
复制标题
EEG 驱动的 RNN 分类对高危患者神经退行性疾病的预后
DOI：
10.1007/978-3-319-44778-0_36
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
A. Soria
中科院分区：
文献类型：
--
作者：
G. Ruffini;D. Ibáñez;M. Castellano;S. Dunne;A. Soria
快速眼动行为障碍（RBD）是帕金森病（PD）等神经退行性疾病的重要危险因素。我们在此描述了一个循环神经网络（RNN），用于分类从RBD患者和健康对照（HC）收集的EEG数据，形成118名受试者的平衡队列，其中50%的RBD患者最终发展为PD或路易体痴呆（LBD）。在早期的工作[1,2]中，我们使用EEG平均谱特征实现支持向量机分类器（svm）来预测双HC与PD问题的病程，准确率为85%。尽管在很大程度上取得了成功，但这种方法并没有试图利用脑电图信号的非线性动态特征，而这些特征被认为包含有用的信息。本文描述了一种回声状态网络（ESN）分类器，该分类器能够处理脑电功率在不同频段的动态特征。分类器的输入是在几个选定的频率和通道上1秒平均脑电功率的时间序列。回声状态网络的性能在HC与PD问题中达到85%的测试集精度，使用我们在之前使用支持向量机解决该问题的工作中选择的相同的信道和频带子集。
REM Behavior Disorder (RBD) is a serious risk factor for neurodegenerative diseases such as Parkinson’s disease (PD). We describe here a recurrent neural network (RNN) for classification of EEG data collected from RBD patients and healthy controls (HC) forming a balanced cohort of 118 subjects in which 50 % of the RBD patients eventually developed either PD or Lewy Body Dementia (LBD). In earlier work [1, 2], we implemented support vector machine classifiers (SVMs) using EEG mean spectral features to predict the course of disease in the dual HC vs. PD problem with an accuracy of 85 %. Although largely successful, this approach did not attempt to exploit the non-linear dynamic characteristics of EEG signals, which are believed to contain useful information. Here we describe an Echo State Network (ESN) classifier capable of processing the dynamic features of EEG power at different spectral bands. The inputs to the classifier are the time series of 1 second-averaged EEG power at several selected frequencies and channels. The performance of the ESN reaches 85 % test-set accuracy in the HC vs. PD problem using the same subset of channels and bands we selected in our prior work on this problem using SVMs.