臺灣博碩士論文加值系統

近年來由於肺癌的高罹患率，越來越多的學者專家投入使用電腦斷層掃描(CT)診斷肺部異常的領域。胸部斷層掃描影像是很方便的工具，現在也被廣泛地使用在許多的相關研究中。然而，對專科醫生來說，病人大量的斷層掃描影像一直是一大挑戰之一，所以許多文獻相繼提出使用電腦輔助自動診斷來幫助人為的檢視工作。但是隨著許多方法的提出，一些問題也浮現出來，自動診斷其實也是非常耗時的工作，因為斷層掃描的切片是非常大量的。就算某個病人確定他患有肺結核，那並不代表說他的每一張影像切片裡面都存在著肺結核。於是此研究提出了一個混合的演算法，目的在於先對肺部影像進行分類工作，將肺部影像分類成3個種類：有肺結核、無肺結核、肺部發炎。在醫生或自動診斷系統開始偵測前，可以先使用這個分類機制，如此一來便能確保輸入的斷層掃描影像存在著肺結核，才需要偵測出肺結核實際的位置、形狀或其他資訊。在初步的實驗結果中顯示，以奇異值分解(SVD)為基礎的小波包轉換(WP-SVD)並以粗集(RS)做分類方法，其正確率高於其他的分類方法，這證明了本研究方法的可用性。

Recently, high prevalence of lung cancer, more and more researcher concern about diagnosing pulmonary lesions in chest CT images. Chest computed tomography (CT) is a well-established tool, which is widely applied to related detection work. However, the great amount of CT scans of each patient still is a challenge for specialists. Hence, many literatures have proposed methods for automatic diagnosis by computer-aided detection (CAD) to assist artificial inspection. But there are some problems brought with those proposed methods, such as time-consuming work of automatically diagnosing by CAD for the large number of CT slices. Even though a patient has trouble with pulmonary nodule positively, that not means there are nodules in each chest slice of him. Therefore this study proposed a hybrid method to initially classify lungs images into 3 classes: nodule, non-nodule, inflammation. The proposed method can be executed before doctor diagnosis or computer-aided system, which can be sure that input CT image need to be detected out the actual positions, shapes or other information of nodules. The results display a higher accuracy in RS based on WP-SVD than other classification methods, which verifies that proposed method can reduce time and cost of lung nodule diagnosis.

中文摘要 i
英文摘要 ii
誌 謝 iii
Contents iv
List of Tables v
List of Figures vi
1. Introduction 1
2. Materials 4
2.1. Datasets 4
2.1.1. LIDC 4
2.1.2. Chest CT images from regional teaching Hospital (RTH) 4
2.2. Related works 5
2.2.1. Singular value decomposition (SVD) 5
2.2.2. Discrete wavelet packets transform (DWPT) 7
2.2.3. Rough sets theory 10
3. Proposed method 13
3.1. The procedure of proposed method 13
3.2. Proposed three stages automated pulmonary segmentation algorithm 16
3.2.1. Segmenting roughly chest 16
3.2.2. Filling the background 19
3.2.3. Extracting the bounding box of lungs 21
3.3. Proposed a novel image process algorithm 23
Step 1: Image adjust contrast 23
Step 1-1: LIDC dataset 24
Step 1-2: RTH dataset 26
Step 2: Lung Extraction 27
Step 3: SVD image reconstruction 29
Step 4: DWPT compute feature extraction phase 30
Step 5: Compute feature value and reduce feature 33
Step 6: Classify lungs images 35
4. Experimental Results 36
4.1. Experiments 36
4.2. Results 38
5. Discussion 39
Appendix A. Region growing algorithm. 40
Reference 41

[1]Arivazhagan, S., & Ganesan, L. (2003). Texture classification using wavelet transform. Pattern Recognition Letters, 24, 1513-1521.
[2]Armato, S.G., McLennan, G., McNitt-Gray, M.F., Meyer, C.R., Yankelevitz, D., & Aberle, D.R., …Clarke, L.P. (2004). Lung image database consortium developing a resource for the medical imaging research community. Radiology, 232, 739-748.
[3]Avci, E. (2007). An expert system based on Wavelet Neural Network-Adaptive Norm Entropy for scale invariant texture classification. Expert Systems with Applications, 32, 919-926.
[4]Avci, E. (2008). Comparison of wavelet families for texture classification by using wavelet packet entropy adaptive network based fuzzy inference system. Applied Soft Computing, 8, 225-231.
[5]Avci, E., Turkoglu, I., & Poyraz, M. (2005a). A new approach based on scalogram for automatic target recognition with X-band Doppler radar. Asian Journal of Information Technology, 4, 133-140.
[6]Avci, E., Turkoglu, I., & Poyraz, M. (2005b). Intelligent target recognition based on wavelet packet neural network. Expert Systems With Applications, 29, 175-182.
[7]Bae, K.T., Kim, J.S., Na, Y.H., Kim, K.G., & Kim, J.H. (2005). Pulmonary nodules at chest CT: effect of computer-aided diagnosis on radiologists detection performance. Radiology, 236, 286-293
[8]Chung, K.L., Yang W.N., Huang, Y.H., Wu, S.T., & Hsu, Y.C. (2007). On SVD-based watermarking algorithm. Applied Mathematics and Computation, 188, 54-57.
[9]Dehmeshki, J., Ye, X., Lin, X.Y., Valdivieso, M., & Amin, H. (2007). Automated detection of lung nodules in CT images using shape-based genetic algorithm. Computerized Medical Imaging and Graphics, 31, 408-417.
[10]Gonzalez, R.C., & Woods, R.E. Digital Image Processing [Second Edition]. Englewood Cliffs, NJ: Prentice-Hall.
[11]Grzymala-Busse, J.W. (1997). A new version of the rule induction system LERS. Fundamenta Informaticae, 31, 27-39.
[12]Helen, H., Jeongjin, L., & Yeny, Y. (2008). Automatic lung nodule matching on sequential CT images. Computers in Biology and Medicine, 38, 623-634.
[13]Huang, K., & Aviyente, S. (2006). Information-theoretic wavelet packet subband selection for texture classification. Signal Processing, 86, 1410-1420.
[14]Kubo, M., Yamamoto, T., Kawata, Y., Niki, N., Eguchi, K., Ohmatsu, H., …Nishiyama, H. (2001). CAD system for the assistance of comparative reading for lung cancer using retrospective helical CT images, 4322, 1788-1795.
[15]Lee, S.L.A., Kouzani, A.Z., & Hu, E.J. (2010). Random forest based lung nodule classification aided by clustering. Computerized Medical Imaging and Graphics, 34, 535-542.
[16]Mallat, S. (1999). A Wavelet Tour of Signal Processing, Academic Press. New York: Academic Press.
[17]Mallat, S., & Zhong, S. (1992). Characterization of signals from multiscale edges. IEEE Transactions Pattern Analysis and Machine Intelligence, 14, 710-732.
[18]Messay, T., Hardie, R.C., & Rogers S.K. (2010). A new computationally efficient CAD system for pulmonary nodule detection in CT imagery. Medical Image Analysis, 14, 390-406.
[19]Mullaly, W., Betke, M., Hong, H., Wang, J., Mann, K., & Ko, J.P. (2002). Multi-criterion 3D segmentation and registration of pulmonary nodules on CT: a preliminary investigation. Proceedings of the International Conference on Diagnostic Imaging and Analysis (ICDIA), 176-181.
[20]Muneeswaran, K., Ganesan, L., Arumugam, S., & Soundar, K.R. (2005). Texture classification with combined rotation and scale invariant wavelet features. Pattern Recognition, 38, 1495-1506.
[21]Pawlak, Z. (1982). Rough sets. International Journal of Computational Information Science, 341-356.
[22]Pawlak, Z. (1991). Rough sets: Theoretical aspects of reasoning about data. Dordrecht: Kluwer.
[23]Pawlak, Z., & Skowron, A. (2007). Rudiments of rough sets. Information Sciences, 177, 3-27.
[24]Sousa, J.R., Silva, A.C., de Paiva, A.C., & Nunes, R.A. (2010). Methodology for automatic detection of lung nodules in computerized tomography images. Computer Methods and Programs in Biomedicine, 98, 1-14.
[25]Takizawa, H., Yamamoto, S., & Shiina, T. (2007). Accuracy improvement of pulmonary nodule detection based on spatial statistical analysis of thoracic CT scans. IEICE transactions on Information and Systems, E90-D, 1168-1174.
[26]Vozalis, M.G., & Margaritis, K.G. (2007). Using SVD and demographic data for the enhancement of generalized Collaborative Filtering. Information sciences, 177, 3017-3037.
[27]Yeny, Y., & Helen, H. (2008). Correction of segmented lung boundary for inclusion of pleural nodules and pulmonary vessels in chest CT images. Computers in Biology and Medicine, 38, 845-857.
[28]http://www.mathworks.com/matlabcentral/fileexchange/19084