標題: 禪坐腦電波頻域-空間特性分析
Spatio – Spectral Characterization of Meditation EEG Signal
作者: 王富滄
Fu- Tsang Wang
羅佩禎
Pei-Chen Lo
電控工程研究所
關鍵字: 腦電波;EEG
公開日期: 2004
摘要: 本篇論文的目的是利用訊號處理技術分析多通道腦電波訊號(electroencephalograph, EEG)之空間-頻率關聯性。在本論文中,我們分成兩部分:第一部分使用一致分析法(Coherence),來分析與萃取受測者多通道腦電波中α波能量的空間關聯性;第二部分觀察受測者α波能量在禪坐(休息)前、中、後的變化。 一致分析法可用以分析腦電波訊號(本研究以禪坐腦電波為主)兩兩通道間之局部空間性,比較前後腦與左右腦兩兩通道間α波功率頻譜函數(Power Spectrum)的關聯性,實驗組與控制組功率頻譜函數變化的差異,或是禪修者的腦電波之變化。並標準化至 0 ∼ 1呈現,更利於觀察腦電波信號Coherence之權重分布。 另一方面,觀察受測者α波能量在禪坐(休息)前、中、後權重空間分布,以頭殼圖表現空間分布,並將其以平均權重範圍為0 ∼ 1表示,可以比較出實驗組與控制組在實驗過程中30通道間α波能量的差異性,並分析其間實驗組與控制組α波能量的增減情形,再對應腦電波之局部區域的相關分析,可較明顯看出其中的變化。
The aim of this thesis is to employ signal processing techniques in the analysis of the spatio-spectral correlation of the multi-channel EEG (electroencephalograph) signals. This thesis is divided into two parts. In the first part, we evaluate the spatial coherence of α rhythmic power in multi-channel EEGs. In the second part, we observe the variation of α rhythmic power between different sessions, that is, before, during, and after meditation (rest). Coherence analysis provides an approach for estimating the degree of local spatial correlation between two channels. On one hand, this study focused on the meditation EEG signals. The results were normalized to 0 ∼ 1 for objective comparison. The display method of topographic mapping allows us to easily visualize the weighting distribution of coherence parameters. On the other hand, we observed the variation of α rhythmic power of each subject between different sessions, that is, before, during, and after meditation (rest). We compared the results of α rhythm power among 30 channels between experimental and control groups, and analyzed the increment and decrement of α rhythmic power during that time. In addition, local EEG correlation was investigated for thoroughly understanding the spatial characteristics of EEG rhythms under Zen meditation.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009212599
http://hdl.handle.net/11536/68934
Appears in Collections:Thesis


Files in This Item:

  1. 259901.pdf
  2. 259902.pdf
  3. 259903.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.