节点文献
基于人工神经网络的癫痫棘波检测方法
A method of epileptic spike detection based artificial neural network
【Author】 QIU Tian-shuang, CHU Meng (Department of Electronic Engineering, Dalian University of Technology, Dalian 116024, China)
【机构】 大连理工大学电子与信息工程学院;
【摘要】 脑电癫痫特征波的自动提取对于患者的诊断以及减轻医生的繁重劳动都具有重要的意义。本文提出一种时频分析与Jensen函数相结合的方法进行棘波检测,然后提取出棘波的波形特征, 并通过人工神经网络进行进一步的判决,从而降低棘波检测的误检率。在对真实的癫痫脑电信号 (EEG)的仿真实验中,该方法取得了较好的结果。
【Abstract】 The automatic spike detection in EEG is significant in both diagnosing illness and alleviating the heavy labour force of the doctor. This paper proposes a new method that combines time-frequency analysis and Jensen function to complete spike detection, and then makes certain the detected spikes with an artificial neural network (ANN) trained with a set of characteristics extracted from EEG signals, so as to reduce the ratio of the false detection. Applying the method to real epileptic EEG, we get better results.
【Key words】 EEG; time-frequency analysis; artificial neural network; spike detection;
- 【会议录名称】 大连理工大学生物医学工程学术论文集(第2卷)
- 【会议名称】大连理工大学生物医学工程学会研讨会
- 【会议时间】2005-12
- 【会议地点】中国辽宁大连
- 【分类号】R318
- 【主办单位】大连理工大学