大腦磁振影像的自動腫瘤分割技術之研究

NCUTIR > Other > National CHIN-YI University of Technology > 勤益學報 > Item 987654321/1918

Please use this identifier to cite or link to this item: http://ir.lib.ncut.edu.tw/handle/987654321/1918

Title:	大腦磁振影像的自動腫瘤分割技術之研究 Automatic Tumor Segmentation of Brain in MRI
Authors:	王圳木;楊勝智;劉昭勇 Wang, Chuin-Mu;Yang, Sheng-Chih;Liu, Zaho-Yong
Contributors:	電子工程系 Department of Electronic Engineering
Keywords:	抑制能量最小化法(CEM);磁振影像(MRI);分類法;大腦影像;腫瘤 Constrained Energy Minimization(CEM);Magnetic resonance imaging;Classification;Brain images;Tumor
Date:	1999-11
Issue Date:	2008-12-05 10:25:40 (UTC+8)
Publisher:	勤益科技大學
Abstract:	在這篇論文中，我們提出了一個新方法，可使腫瘤從腦部的多頻譜磁振影像(Magnetic Resonance,MR)內給分類出來，這些磁振影像的組成是由三個磁振參數，質子密度(Proton Density,PD),Tl分量(Weighted)和T2分量影像，經過多頻諳分析所處理出來的。我們提出的這個方法，稱為抑制能量最小化法(Constrained Energy Minimization,CEM)，被研發於[12]內，認為在所知道的領域中，把期望的特徵給分類出來，這方法所使用的基礎，就是矩陣運算中的最小化變異數無雜訊響應(Minimum Variance Distortionless Response,MVDR),CEM將多頻譜MR影像視為一個矩陣處理的問題，在這問題中的每一個親察值都代表一個頻譜頻帶(Spectral Band)，再使用有限脈衝響應濾波器(Finite Impulse Response,FlR)將輸出必v做最小化處理，使期望的特徵被強制到一個特定的增益。此方法已經由數個實驗所得證，實驗結果頭示出大腦組織正確地被分割為四張影像，腦瘤(Tumor)、灰質(Gray Matter)、白質(White Matter)以及腦脊髓液(Cerebral Spinal Fluid,CSF)，這些都象徵了這個方法是非常有用的。另外，對於計算所花的時間而言，實驗結果也顯示出繁雜計算的改進處。 In this paper, we present a new method to classify tumor in multispectral magnetic resonance (MR) images of the human brain. The MRI's consist of three magnetic resonance parameters proton density, T1-weighted, and T2-weighted images, which are processed with multispectral analysis. The proposed approach, called Constrained Energy Minimization (CEM) was developed in [12] where only the knowledge of the desired signature to be classified as required. It was derived based on the Minimum Variance Distortionless Response (MVDR) in array processing. CEM considers an MR image classification problem as an array-processing problem where each sensor represent one spectral band. It uses a finite impulse response (FIR) filter to minimize the output power while the deslred signature is constrained to a specific gain. The method has been evaluated through several experiments. Results show that the cerebral tissue was segmented accurately into four images, tumor, gray matter, white matter and cerebral spinal fluid indicating the possible usefulness of this method. As far as computing saving is concerned, the experimental results also show computational complexity improvement.
Relation:	勤益學報 No.17 p.127-136
Appears in Collections:	[National CHIN-YI University of Technology] 勤益學報

Files in This Item:

File	Description	Size	Format
Automatic Tumor.pdf		2267Kb	Adobe PDF	2897	View/Open

Loading...