• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Australasian physical & engineering sciences in medicine >Hidden pattern discovery on epileptic EEG with 1-D local binary patterns and epileptic seizures detection by grey relational analysis
【24h】

Hidden pattern discovery on epileptic EEG with 1-D local binary patterns and epileptic seizures detection by grey relational analysis

机译：一维局部二进制模式的癫痫脑电图的隐藏模式发现和通过灰色关联分析检测癫痫发作

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This paper proposes a novel approach to detect epilepsy seizures by using Electroencephalography (EEG), which is one of the most common methods for the diagnosis of epilepsy, based on 1-Dimension Local Binary Pattern (1D-LBP) and grey relational analysis (GRA) methods. The main aim of this paper is to evaluate and validate a novel approach, which is a computer-based quantitative EEG analyzing method and based on grey systems, aimed to help decision-maker. In this study, 1D-LBP, which utilizes all data points, was employed for extracting features in raw EEG signals, Fisher score (FS) was employed to select the representative features, which can also be determined as hidden patterns. Additionally, GRA is performed to classify EEG signals through these Fisher scored features. The experimental results of the proposed approach, which was employed in a public dataset for validation, showed that it has a high accuracy in identifying epileptic EEG signals. For various combinations of epileptic EEG, such as A-E, B-E, C-E, D-E, and A-D clusters, 100, 96, 100, 99.00 and 100 % were achieved, respectively. Also, this work presents an attempt to develop a new general-purpose hidden pattern determination scheme, which can be utilized for different categories of time-varying signals.
机译：本文基于一维局部二值模式（1D-LBP）和灰色关联分析（GRA），提出了一种通过脑电图（EEG）检测癫痫发作的新方法，这是诊断癫痫的最常用方法之一。 ） 方法。本文的主要目的是评估和验证一种新颖的方法，该方法是一种基于计算机的定量脑电图分析方法，基于灰色系统，旨在帮助决策者。在这项研究中，利用所有数据点的一维LBP提取原始EEG信号中的特征，采用Fisher评分（FS）选择代表性特征，也可以将其确定为隐藏模式。此外，执行GRA通过这些Fisher评分功能对EEG信号进行分类。该建议方法的实验结果在公共数据集中进行了验证，结果表明该方法在识别癫痫性脑电信号方面具有很高的准确性。对于癫痫性脑电图的各种组合，例如A-E，B-E，C-E，D-E和A-D簇，分别达到100％，96％，100％，99.00％和100％。同样，这项工作提出了一种尝试开发一种新的通用隐藏模式确定方案，该方案可用于不同类别的时变信号。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号