摘要
提出了一种新的睡眠梭形波 (sleep spindle)识别方法——离散 Gabor谱分解 ,在研究睡眠脑电波特征的基础上利用这一高分辨率的时频分析方法对睡眠脑电进行了分析处理。结果显示 ,离散 Gabor谱方法可有效地从睡眠脑电波中识别出梭形波 ,为睡眠自动分阶的实现提供了特征。该方法识别梭形波的准确率已接近神经内科专家目测的水平 ,识别准确率达到 93%以上。睡眠梭形波的自动识别为研究睡眠的神经内科专家解除了阅读睡眠脑电图谱的繁冗工作 。
In this paper, a new method for the auto detection sleep spindle is presented. The method is based on the Discrete Gabor Spectrogram (DGS), a high resolution time frequency analysis. Sleep EEG 5 were processed and the spindles were accurately detected. By this method the accuracy of auto detection almost approched the level of visulal detection. Auto detection of sleep spindles could release the expert from reading long term sleep EEG and provide useful information for sleep studies.
出处
《生物医学工程学杂志》
EI
CAS
CSCD
北大核心
2000年第1期50-55,共6页
Journal of Biomedical Engineering
