臺灣博碩士論文加值系統

根據衛生署統計肺癌則已連續二十多年居於男性惡性腫瘤死因之第二位與女性惡性腫瘤死因之第一位。由於肺癌的末期治癒率相當的低，故如何在早期就診斷出肺癌是相當重要的議題。
低劑量電腦斷層掃描已有能力偵測到直徑1-2 mm的肺部腫瘤，但是肺部電腦斷層掃描的判讀卻不是一件容易的事。例如，對於分辨圓形的腫瘤與管狀的血管結構，醫師必須逐張切片影像仔細檢查，這是非常耗費時間且極為容易造成判斷錯誤。尤其是在電腦斷層影像中尋找微小肺腫瘤更是容易造成失誤。如果醫師僅以肉眼尋找，其準確程度有其極限。
而電腦輔助偵測不僅在研究上不斷被證明可以協助醫師提高電腦斷層掃描中肺腫瘤的偵測率，在醫療儀器市場上所逐漸受到重視。然而，不論是商用產品的電腦輔助偵測軟體或是仍處於實驗室階段的電腦輔助偵測與診斷軟體，都未臻理想，存在著很大的進步空間。故如何提升電腦輔助偵測軟體的效能，為一重要研究課題。
本研究針對肺部電腦斷層掃描影像，對在肺臟中的小腫瘤做了一系列的探討及分析，並提出一套方法希望電腦能透過這些資料自我學習如何辨識腫瘤，以期盼在未來協助醫師判斷及診斷。本研究所採用的肺部電腦斷層影像為美國國家癌症研究會(NATIONAL CANCER INSTITUTE，NCI)所提供的397位病人的電腦斷層掃描影像以及四位醫師在上面的標記。
本研究著重於小腫瘤的分析及判別，其流程分為三個部份，第一步先將肺區透過自適應的閥值及型態學的方法分割出來。再使用Hessian Matrix找出血管的樹狀結構並將血管標示出來。在有了肺區及血管資訊之後，將小腫瘤附近的肺壁及血管的資料連帶小腫瘤本身的資訊紀錄下來，以供後續分析。第二步將紀錄下來的資料依照四種不同類型的腫瘤做自動化分割，再對血管及肺壁附近的腫瘤做分割後的微調，使分割的結果更加精確。由分割後的結果，依據腫瘤本身的特徵及腫瘤附近的特徵做一系列的分析。第三步是由這些特徵資料訓練一個自動化的分類器學習如何辨識小腫瘤，經過基因演算法做特徵分類訓練分類器，再由學習後的分類器分類腫瘤。
本研究使用397位病人的肺部影像資料，將資料隨機抽選為不重複的訓練樣本及測試樣本。經過訓練樣本所訓練出的分類器，對於所抽選的測試樣本可達到平均83%以上的正確率。

Statistics from health department shows that the lung cancer has been on the second place for male and the first place for female in causes of death by malignant tumors for more than twenty consecutive years. Since the survival rate of late stage
lung cancer is relatively low, early diagnosis is critical to survival.
Low dose CT is capable of detecting lung tumors of length 1-2 mm in diameter, however the CT interpreting of which remains as a non-trivial task. For example, to identify round-shape tumor and tube-shape blood vessel structure, a physician must undergo a time-consuming and error-prone process to examine each CT carefully. And it is even more difficult when it comes to find small nodules in this way. A physician must go beyond naked-eyes examination for better precision and efficiency.
Computer aid has been proven to be able to assist physicians in improving detection rate of lung tumors in CT scans, and it is getting wider popularity in the business of medical instruments. However, detection software in either commercial products or laboratory developments has yet much room to improve. Therefore it is an essential subject of research to improve the performance of computer aided detection
software.
This research goes through a series of analysis on pulmonary nodules and presents a set of methods for computers to be able to learn identifying tumors, and it is hoped that these methods may serve as diagnosis aids for doctors in the future. NATIONAL CANCER INSTITUTE (NCI) provides essential data including computed tomography (CT) images of around 397 patients with doctors'' marking.
The first step of the research is to automatically identify nodules by computer and store the data of nodules and nearby lung wall and blood vessels. The second step is to automatically partition the data by the types of nodules and perform a series of characteristic analysis. The third step is to train a automatic operator that classifies tumors on the basis of the characteristic data obtained in previous step, and the trained operator is put to work identifying nodules and then be evaluated for performance.
With the tumor data of 397 patients as the basis of training, the automatic operator achieves a satisfactory performance. The classifier trained out of the samples can achieve 83% up correction rate on average on the selected samples.

誌　謝 i
摘　要 ii
ABSTRACT iv
目錄 vi
表目錄 ix
圖目錄 x
第一章　緒論 1
1.1 前言 1
1.2 研究背景 1
1.3 研究目的 3
1.4 文獻回顧 4
1.5 論文架構 5
第二章　基礎理論 7
2.1 肺腫瘤簡介 7
2.1.1 肺的基本組織結構 7
2.1.2 肺腫瘤分類 7
2.1.3 肺腫瘤在影像上的表現 8
2.2 肺部電腦斷層影像 9
2.2.1 電腦斷層掃描 9
2.2.2 電腦斷層影像格式 10
2.3 影像處理相關理論 10
2.3.1簡介 10
2.3.2 肺區截取演算法 11
2.3.3 Hessian Matrix 12
2.3.4血管追蹤演算法 12
2.3.5 Otsu閥演算法 14
2.3.6 區域成長演算法 15
2.4基因演算法 16
2.4.1簡介 16
2.4.2相關應用 18
2.4.3結論 19
第三章　研究方法 21
3.1前言 21
3.2 研究資料 22
3.3 腫瘤截取 23
3.4 腫瘤分割演算法 23
3.5腫瘤本身所呈現的特徵 30
3.5.1 一般統計量 30
3.5.2 體積 30
3.5.3 表面積 30
3.5.4 球型分析 31
3.5.5 Shape Index 31
3.5.6 似圓性 32
3.6腫瘤週遭所呈現的特徵 32
3.6.1 集中值 32
3.6.2 與肺壁最近距離 33
3.6.3 血管 33
3.7腫瘤分類 33
3.7.1 訓練分類器 33
3.7.2 調整參數標準化 36
3.7.3 參考非腫瘤資訊 36
3.7.4 使用分類器分類待判斷區塊 36
3.8 結論 37
第四章　系統架構及實驗 38
4.1 系統架構與設計 38
4.1.1 資料來源程式 38
4.1.2 特徵分析程式 41
4.1.3 分類器訓練及測試程式 42
4.1.4 測試環境 45
4.2 實驗一：分類器可靠性驗證 46
4.2.1 測試方法 46
4.3 實驗二：大量樣本測試 47
4.3.1 測試方法 47
4.3.2 實驗結果 47
第五章　結論與未來研究方向 49
5.1結論 49
5.2未來研究方向 49
參考文獻 50

[1]Henschke CI, Mccauley DI, Yankelevitz DF. "Early Lung Cancer Action Project: A Summary of the Findings on Baseline Screening." The Oncologist, vol. 6, 2001, pp. 147-152.
[2] Heelan RT, Flehinger BJ, Melamed MR. "Non–small-cell lung cancer: results of the New York screening program." Radiology, vol.151, 1984, pp.289–293.
[3]Yuan R, Vos PM, Cooperberg PL. "Computer-Aided Detection in Screening CT for Pulmonary Nodules." AJR, vol. 186, 2006, pp.1280–1287.
[4]Das M, Muhlenbruch G, Mahnken AH, et al. "Small Pulmonary Nodules: Effect of Two Computer-aided Detection Systems on Radiologist Performance." Radiology vol. 241, no. 2, 2006, pp.564-571.
[5]Gierada DS, Pilgram TK. "Lung Cancer: Interobserver Agreement on Interpretation of Pulmonary Findings at Low-Dose CT Screening." Radiology, vol. 246, no. 1, 2008, pp.265-272.
[6]Armato III SG, Giger ML, MacMahon H. "Automated detection of lung nodules in CT scans: Preliminary results." Med. Phys, 2001, vol.28, no.8.
[7]Lee Y, Hara T, and Fujita H et al.. "Automated Detection of Pulmonary Nodules in Helical CT Images Based on an Improved Template-Matching Technique." IEEE Transactions on Medic Imaging ,2001, vol.20, no.7.
[8]Gurcan MN, Sahiner B, Petrick N. "Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computer-aided diagnosis system." Medical Physics, 2002, vol.29 no.11, pp.2552-2558.
[9]Brown MS, Goldin JG, Suh RD. "Lung Micronodules: Automated Method for Detection at Thin-Section CT—Initial Experience." Radiology, 2003, vol. 226, pp.256-262
[10]Suzuki K, Armato SG III, Li F. "Massive training artificial neural network MTANN for reduction of false positives in computerized detection of lung nodules in low-dose computed tomography." Medical Physics, 2003, vol. 30, no.7, pp. 1602-1617.
[11]McNitt-Gray MF, Hart EM, Wyckoff N." A pattern classification approach to characterizing solitary pulmonary nodules imaged on high resolution CT: Preliminary results." Medical Physics, 1999, vol.26, no.6, pp.880-888.
[12]Shah SK, McNitt-Gray MF, Rogers SR. " Computer-aided Diagnosis of the Solitary Pulmonary Nodule." Acad Radiol, 2005, vol. 12, pp.570-575.
[13]Way TW, Sahiner B, Chan HP. "Computer-aided diagnosis of pulmonary nodules on CT scans: Improvement of classification performance with nodule surface features." Medical Physics, 2009, vol.36, no.7, pp.3086-3098.
[14]Lee MC, Wiemker R, Boroczky L. "Impact of segmentation uncertainties on computer-aided diagnosis of pulmonary nodules." Int J CARS, 2008, vol. 3, pp.551–558.
[15]Kawata Y, Niki N, Ohmatsu H. "Quantitative surface characterization of pulmonary nodules based on thin-section CT images." IEEE Trans Mucl Sci, 1998, vol. 45, no.4, pp.2132-2138.
[16]Kostis WJ, Reeves AP, Yankelevitz DF, Henschke CI. "Three-dimensional segmentation and growth-rate estimation of small pulmonary nodules in helical CT images." IEEE Trans Med Imaging, 2003, vol.22, pp.1259–1274.
[17]Reeves AP, Chan AB, Yankelevitz DF, Henschke CI, Kressler B, Kostis WJ. "On measuring the change in size of pulmonary nodules." IEEE Transactions on Medical Imaging, 2006, vol.25, pp.435–450.
[18]S. Hu and J. Reinhardt, “Automatic lung segmentation for accurate quantitation of volumetric x-ray ct images,” IEEE Transactions on Medical Imaging, vol. 20, no. 6, pp. 490–498, June 2001.
[19]M. Sonka, V. Hlavac, and R. Boyle, Image Processing, Analysis and Machine Vision. Pacific Grove, CA: PWS, 1999.
[20]A. F. Frangi, W. J. Niessen, K. L. Vincken, and M. A. Viergever, "Multiscale vessel enhancement filtering," MICCAI, 1998, vol. 1496 pp. 130–137.
[21]N. Otsu, “A threshold selection method from gray level histogram,” IEEE Transactions on Systems, Man and Cybernetics, SMC - 9, 1979, pp. 62-66.
[22]RK Justice, EM Stokely, "3-D Segmentation of MR Brain Images Using Seeded Region Growing," 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 1083-1084.
[23]E. Alexandre, L. Cuadra, M. Rosa and F. Lopez-Ferreras, "Feature Selection for Sound Classification in Hearing Aids Through Restricted Search Driven by Genetic Algorithms," IEEE Transactions on Audio, Speech, and Language Processing, vol. 15, 2007, pp. 2249-2256.
[24]T. W. Manikas, K. Ashenayi and R. L. Wainwright, "Genetic Algorithms for Autonomous Robot Navigation," Instrumentation & Measurement Magazine, IEEE, vol. 10, 2007, pp. 26-31.
[25]William J. Kostis, Anthony P. Reeves, David F. Yankelevitz and Claudia I. Henschke " Three-Dimensional Segmentation and Growth-Rate Estimation of Small Pulmonary Nodules in Helical CT Images," IEEE Transactions on Medical Imagine, vol. 22, 2003, pp. 1259-1273.
[26]Xiangwei Zhang et. al., “A New Method for Spherical Object Detection and Its Application to Computer Aided Detection of Pulmonary Nodules in CT Images,” MICCAI 2007, Part I, LNCS 4791. 2007; 842–849.
[27]J. Holland, Adaptation in Natural and Artificial System, USA, MA: MIT Press Cambridge, 1992.
[28]Awai K, Murao K, Ozawa A, "Pulmonary Nodules: Estimation of Malignancy at Thin-Section Helical CT—Effect of Computer-aided Diagnosis on Performance of Radiologists," Radiology, vol. 239, 2006, pp.276-284.
[29]Way TW, Hadjiiski LM, Sahiner B, "Computer-aided diagnosis of pulmonary nodules on CT scans: Segmentation and classification using 3D active contours," Medical Physics, vol. 33, 2006, pp.2323-2337.
[30]Li F, Aoyama M, Shiraishi J, "Radiologists’ performance for differentiating benign from malignant lung nodules on high-resolution CT using computer-estimated likelihood of malignancy," AJR, vol.183, 2004, pp.1209-1215.
[31]彰化基督教醫院肺癌醫療暨衛教網頁
http://www2.cch.org.tw/lungcancer/