跳到主要內容

臺灣博碩士論文加值系統

(18.97.14.82) 您好!臺灣時間:2025/02/18 23:52
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:楊妤婷
研究生(外文):Yu-Ting Yang
論文名稱:形態學切割為基礎的腦部疾病MRI之分類
論文名稱(外文):A hybrid classification approach based on morphological segmentation for Classifying MRI brain diseases.
指導教授:鄭景俗鄭景俗引用關係
指導教授(外文):Ching-Hsue Cheng
學位類別:碩士
校院名稱:國立雲林科技大學
系所名稱:資訊管理系碩士班
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2013
畢業學年度:102
語文別:英文
論文頁數:39
中文關鍵詞:切割CART腦疾病小波包核磁共振成像SVM
外文關鍵詞:SVMwavelet packet transformMRIbrain diseases.CARTsegmentation
相關次數:
  • 被引用被引用:0
  • 點閱點閱:412
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
二十一世紀以來全球人口老化的問題日趨嚴重,大眾開始關心腦部退化的相關疾病議題。核磁共振成像是目前最先進的醫療成像方法之一,其特性適用於腦部。各種自動化與半自動化的腦部核磁共振影像切割已廣泛應用於醫療影像處理來輔助醫療人員診斷與分類腦部疾病。從過去的研究發現,雖然自動化切割省時省力但是只限定於處理一些特定型態的影像,對於不同輪廓、形狀的影像處理能力有限。此外混合式的方法結合不同方法的優點並改善原有缺點,比起單一切割方法更能達到良好的切割績效。因此,本研究提出一個混合式半自動切割方法演算法,基於提出演算法為基礎提出一個三階段的腦部影像處理方法來提升分類的正確率。第一階段,利用切割演算法切割出感興趣的腦部區域並使用形狀過濾器篩選影像,第二階段利用小波轉換解構及計算其特徵值,第三階段利用機器學習之分類演算法將影像分為兩類:正常與不正常腦影像。最後實驗結果顯示有經過形狀過濾器篩選資料集的正確率(77.32(0.72)-80.13(1.95)%)遠高於未經篩選的實驗(56.44(1.48) - 58.37(1.13)%)。此結果可以作為相關分類方法參考與醫療人員在MR影像診斷上參考使用。
Since 21 Century, the global population aging problem is becoming more and more serious, people began to concern topics related to brain degeneration disease. Magnetic resonance imaging is one of the most advanced medical imaging method, its characteristics for the brain. Brain MRimages of all kinds of automatic and semi-automatic segmentation has been commonly used in medical image processing to medical auxiliary diagnosis and classification of brain diseases. Findings from past studies, although the automatic segmentation time-saving but only limited to the treatment of some certain types of images, for different contour, shape the image processing ability is limited. In addition, the advantages of hybrid method combining different methods and to improve the shortcomings, compared with single segmentation method can achieve good segmentation performance. Therefore, this study proposes a hybrid semi-automatic segmentation method algorithm, based on the performance of the proposed algorithm based on a three stage of brain image processing method to improve the rate of correct classification. The first part, the segmentation algorithm segmentout the brain regions of interest.The second part,deconstruction and calculating the feature by using wavelet transform.The third part, use machine learning classification algorithm to classifyimagesinto two kind: normal and abnormal brain images. Finally, experimental results show accuracy through shape filter select datasets (77.32 (0.72) - 80.13 (1.95)%) is evidenthigher than without shape filter (56.44 (1.48) - 58.37 (1.13)%). This result can be used as a reference on related classification method and medical personnel diagnose MR images.
Contents
摘要 i
Abstract ii
List of Tables iv
List of Figures v
1. Introduction 1
1-1 Background and Motivation 1
1-2 Research Objectives 3
1-3 Organization of the thesis 4
2. Literature Review 5
2.1 Magnetic Resonance Imaging and brain diseases 5
2.2 Segmentation techniques 5
2.2.1 Morphological operation 7
2.2.2 Binarization operation 7
2.3 Wavelet packet transform 8
2.4 Classification algorithm 10
This section will reviewthe classification algorithm used in this study :Support vector machine, Sequential Minimal Optimization (SMO) and Classification and regression trees. 10
2.4.1 Support vector machine 10
2.4.2 Classification and regression trees 12
3. Proposed method 15
3.1 Concept 15
3.2 The proposed algorithm 20
3.2.1 Datasets 20
3.2.2 Proposed algorithm 20
4. Experiment and Results 25
4.1Evaluation and Comparison 25
4.1.1 Whole Brain Atlasdataset experiment results 25
4.1.2 Region hospital dataset experiment results 26
4.3Findings 29
5. Conclusion 31
Reference 32
Avci, E. (2008). Comparison of wavelet families for texture classification by using wavelet packet entropy. Applied Soft Computing, 8(1), pp. 225-231.
Binning, M., Liu, J., Kestle, J., Brockmeyer, D., &; Walker, M. (2007, 11). Optic pathway gliomas: a review. Neurosurgical Focus, 23(5), pp. p1-8.
Boser, E. B., Guyon, M. I., &; Vapnik, V. (1992). A training algorithm for optimal margin classifiers. ACM, pp. p144-152.
Breiman, L., Friedman, J. H., Olshen, R. A., &; Stone, C. I. (1984). Classification and regression trees. Belmont, Calif: Wadsworth.
Burrus, C., Gopinath, R., &; Guo, H. (1988). Introduction to Wavelet and Wavelet Transforms. New Jersey: Prentice-Hall.
Chen, Y. (2011). A robust fully automatic scheme for general image segmentation. Digital Signal Processing, 21, pp. 87-99.
Chu, C., &; Aggarwal, K. J. (1993). The integration of image segmentation maps using region and edge information. IEEE Transaction on Pattern Analysis and Machine Intelligence, 15(12), pp. p1241-1252.
Chuang, C.-L. (2013, 7). Application of hybrid case-based reasoning for enhanced performance in bankruptcy prediction. Information Sciences, 236, pp. pp. 174-185.
Cortes, C., &; Vapnik, V. (1995, 9). Support-vector networks. Machine Learning, 20(3), pp. 273-297.
Devasahayam, S. (2000). Signals and Systems in Biomedical Engineering. Kluwer Academic Publishers.
Fan, J., Yau, D. Y., Elmagarmid, K. A., &; Aref, W. (2001). Automatic Image Segmentation by Integrating Color-Edge Extraction and Seeded Region Growing. IEEE Transactions on Image Processing, 10(10), pp. p1454-1466.
Fan, X. S., Man, Z., &; Samur, R. (2004). Edge based region growing: a new image segmentation method. ACM.
Fleisher, A. S., Chen, K., Quiroz, Y. T., Jakimovich, L. J., Gomez, M. G., Langois, C. M., . . . Reiman, E. M. (2012, 11 5). Florbetapir PET analysis of amyloid-β deposition in the presenilin 1 E280A autosomal dominant Alzheimer’s disease kindred: a cross-sectional study. The Lancet Neurology.
Gregg C., F., Kirkwood F., A., William T., A., &; ADHERE Investigators. (2003, 10). Risk stratification for in-hospital mortality in heart failure using classification and regression tree (CART) methodology: analysis of 33,046 patients in the ADHERE registry. Journal of Cardiac Failure, 9(5), p. pp. s79.
Guler, I., Demirhan, A., &; Karak, R. (2009, 7). Interpretation of MR images using self-organizing maps and knowledge-based expert systems. Digital Signal Processing, 19(4), pp. p688-677.
Haar. (1910). Theorie der orthogonalen funktionen -systeme. Mathematische Annalen, 69, pp. p331-371.
Kohonen, T. (1990, 9). The self-organization map. Proceedings of the IEEE, 78(9), pp. 1464-1480.
Li, Y. (2006, 10). Predicting materials properties and behavior using classification and regression trees. Materials Science and Engineering, 433(1-2), pp. pp.261-268.
Mallat, S. (1988). A Theory for Multiresolution Signal Decomposition, The Wave Representation. Comm. Pure Appl. Math, 41, pp. p674-693.
McEvoy, L. K., Holland, D., Hagler, D. J., Fennema-Notestine, C., Brewer, B. J., &; Dale, M. A. (2011). Mild Cognitive Impairment: Baseline and Longitudinal Structural MR Imaging Measures Improve Predictive Prognosis. Radiology, 259(3), pp. p834-843.
Otsu, N. (1979). A Threshold Selection Method from Gray-Level Histograms. IEEE Transactions on Systems, Man, and Cybernetics, 9(1), pp. 62-66.
Pavlidis, T., &; Liow, Y. (1990). Integrating region growing and edge detection. IEEE Transaction on Pattern Analysis and Machine Intelligence, 12(3), pp. p225-233.
Toulson, D., &; Boyce, J. (1992). Segmentation of MR image using neural nets. Image and Vision Computing, 10, pp. p324-328.
Turkoglu, I., Arslan, A., &; Ilkay, E. (2003). An intelligent system for diagnosis of the heart valve diseases with wavelet packet neural Networks. Computers in Biology and Medicine, 33(4), pp. p319-331.
Weizman, L., Ben Sira, L., Joskowicz, L., Constantini, S., Precel, R., Shofty, B., &; Ben Bashat, D. (2012). Automatic segmentation, internal classification, and follow-up of optic. Medical Image Analysis, 16, pp. 177-188.
Yu, X., Yla-Jasski, J., Huttunen, O., Vehkomaki, T., Sipila, O., &; Katila, T. (1992). Image segmentation combining region growing and edge detection. In Proc. of the 11th IAPR International Conf. on Pattern Recognition, 3, pp. p481-484.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊