利用獨立成份分析判讀老鼠主要運動區訊號

Using ICA to Distinguish the Signal of Primary Motor Cortex of Rats

藍志仁(Chih-Jen Lan)

指導教授 ： 駱榮欽

國立臺北科技大學/電資學院/電腦與通訊研究所/碩士(2007年)

摘要

獨立成份分析常被用於分析腦波，在臨床上的應用由於受到雜訊和干擾源影響使得我們無法直接解釋與分析腦波所代表的涵義。但是獨立成份分析技巧能讓我們取到獨立的訊號源並且做更進一步的訊號處理。本研究主要的方向也在於使用獨立成份分析判讀腦波訊號，與發展出一套演算法(動態增加權重)來對腦波訊號處理並且完成一套腦波擷取與分析系統。而整套系統採用個人電腦為平台自行研發一套低成本、低雜訊及同步影像擷取的分析系統，並用於活動中老鼠腦皮質層受外在刺激之訊號的擷取與分析。系統的建構使用Matlab和LabVIEW， 能執行 WT， FFT， JTFA， PCA ，ICA ，DIW和Cluster 之訊號分析。這個系統已完成一些以白色大鼠為對象的實驗，從這些實驗中分析並歸納出活動中老鼠的腦皮質訊號與行為之間的關連性與規則性。最後本研究提出一個演算法和一套分析流程用來判讀老鼠主要運動區之訊號以利大家做後續的研究。

關鍵字

主成份分析 ； 獨立成份分析 ； 動態增加權重 ； 訊號擷取 ； 訊號分析

並列摘要

Independent component analysis (ICA) technique is applied to the analysis of Electroencephalographic (EEG) signal. In practice, some artifacts problems limit the interpretation and analysis of clinical EEG signals since the rejected contaminated EEG segments results in an unacceptable data loss. In this contribution, ICA filters were trained based on the EEG data during these sessions were identified statistically independent source channels, which could then be further processed using other signal processing techniques. This paper discussed methods for independent source identification within multiple channels electroencephalographic (EEG) signals recordings. We develop the DIW method to classify EEG data of rat’s movement and have implemented a cortex bio-signal capture and analysis system. The system uses an IBM PC as platform to design a low cost, low noise bio-signal capture circuit and synchronally monitors the rat activity images. This system functions include Wavelet Transform (WT), Fast Fourier Transform (FFT), Joint Time-Frequency Analysis (JTFA), Correlation, Principal Component Analysis (PCA), Independent Component Analysis (ICA), Dynamic Increase Weight (DIW) and Cluster, developed by Matlab and LabVIEW to record and analyze signal. Using the system, several experiments of cortex bio-signal capture and analysis has been performed on the SD rats. From the results of the bio-signal analysis, we have found several relationship and regulation between the rat’s cortex bio-signal and behavior. Finally, this study offered a way of the identification capability of brain signal by cluster, and suggests avenues for future research.

並列關鍵字

PCA ； ICA ； DIW ； Signal Capture ； Signal Analysis ； EEG

參考文獻

[1] N. G. Hatsopoulos & Jignesh Joshi,” Multi-electrode recording from motor cortex in behaving primates.” IEEE EMBS Cancun, Mexico September 17-21, 2003

[2] R. Vig´ario, “Extraction of ocular artifacts from EEG using independent component analysis.” Electroenceph. Clin. Neurophysiol, 103(3):395–404, 1997.

[8] Aapo Hyvärinen , Juha Karhunen and Erkki Oja, ”Independent component analysis”, Copyright 2001 by John Wiley & Sons, Inc. All rights reserved.

[9] Shuhei Kinulawa, Manabu Kotani , Seiichi Ozawa, “A subspace Method for Feature Extraction Using Independent Components”, SICE2002 Aug,5-7,2002

[10] Eugene M. Izhikevich, ” Which Model to Use for Cortical Spiking Neurons?” In IEEE Transactions on Neural Networks, VOL. 15, NO. 5, September 2004