基于卡尔曼滤波器的癫痫棘波检测方法。

Epileptic spike detection using a Kalman filter based approach.

作者信息

Tzallas Alexandros T, Oikonomou Vaggelis P, Fotiadis Dimitrios I

机构信息

Dept. of Medical Physics, University of Ioannina, Greece.

出版信息

Conf Proc IEEE Eng Med Biol Soc. 2006;2006:501-4. doi: 10.1109/IEMBS.2006.260780.

DOI:10.1109/IEMBS.2006.260780

PMID:17945981

Abstract

The electroencephalogram (EEG) consists of an underlying background process with superimposed transient nonstationarities such as epileptic spikes (ESs). The detection of ESs in the EEG is of particular importance in the diagnosis of epilepsy. In this paper a new approach for detecting ESs in EEG recordings is presented. It is based on a time-varying autoregressive model (TVAR) that makes use of the nonstationarities of the EEG signal. The autoregressive (AR) parameters are estimated via Kalman filtering (KF). In our method, the EEG signal is first preprocessed to accentuate ESs and attenuate background activity, and then passed through a thresholding function to determine ES locations. The proposed method is evaluated using simulated signals as well as real inter-ictal EEGs.

摘要

脑电图（EEG）由一个潜在的背景过程以及叠加的瞬态非平稳性组成，如癫痫棘波（ESs）。在脑电图中检测癫痫棘波对于癫痫的诊断尤为重要。本文提出了一种在脑电图记录中检测癫痫棘波的新方法。它基于一个时变自回归模型（TVAR），该模型利用了脑电图信号的非平稳性。自回归（AR）参数通过卡尔曼滤波（KF）进行估计。在我们的方法中，首先对脑电图信号进行预处理，以突出癫痫棘波并减弱背景活动，然后通过一个阈值函数来确定癫痫棘波的位置。使用模拟信号以及实际发作间期脑电图对所提出的方法进行了评估。