面向肺部CAD的三维ROI分割、特征提取与分类方法研究
|
摘要
肺癌是当今死亡率最高的癌症疾病之一,严重地威胁着人类的生命健康。为了提高肺癌患者的生存率,肺癌的早期诊断与治疗是关键。CT扫描是肺癌诊断的重要手段。肺癌在CT图像中是以肺结节的形式表现的,而肺部本身结构的复杂性和肺结节形状大小的多样性,使得富有经验的医生也很难及时发现图像中所有可能病变区域。同时随着多层螺旋CT的出现,医生需要处理的图像信息急剧增加,因此迫切需要一种有效的技术来减轻医生的工作,计算机辅助诊断方法就这样慢慢发展起来。它可以辅助医生对病变及其它感兴趣的区域进行定性乃至准确的定量分析,从而提高医疗诊断的效率,减轻医生的负担。
本文对面向肺部CAD的三维RO1分割、特征提取与分类方法进行研究。主要研究工作包括三部分：首先研究了肺部感兴趣区域(ROI)的三维分割算法。包括自适应三维区域生长算法和基于三维Hessian矩阵的算法,并对两种算法分割后的结果进行分析和比较。其次研究了感兴趣区域的三维特征提取算法。从全局考虑,提取了表面积、体积、球形度、凸形度、紧凑度和灰度均值6个特征；从局部考虑,提取了肺结节的局部体积形状指数特征。并根据肺结节的CT影像特征,对所提取的特征进行分析评价。最后根据感兴趣区域提取的有效特征,应用Fisher分类器和基于核的Fisher分类器对结节区域和正常组织进行分类。实验表明,改进的基于核的Fisher分类器比Fisher分类器具有更好的分类性能和准确性。
整体而言,本文充分考虑了CT图像的上下层信息,运用三维的方法对感兴趣区域进行分割、特征提取及分类算法的研究,取得了较好的效果,可以为肺部结节的判别提供辅助诊断依据。
Lung cancer is one of the most deadly diseases in the world, it threatens people's lives. To improve lung cancer patients'viability, the earliest diagnosis and therapy of lung cancer is essential. CT scanning has been the most important method of lung cancer examination. The lung cancer is represented with the pulmonary nodules in CT images, and since the complexity of the lung structure and the multiplicity of nodule's shape, even the seasoned doctor cannot easily discover all of the possible disease in time. However, with the advent of the multislice CT, the image data which were handled by the physician drastically increased. An assistance is cried for in order to lighten physician's burdens, so computer-aided diagnosis (CAD) is developed gradually. It may help doctors to analyze pathological changes and other regions of interest in character and even in accurate quantity, and release the doctors'burden.
In this thesis,3D ROI segmentation, feature extraction and classification methods for pulmonary CAD are researched. The main work includes the following three parts:first,3D segmentation algorithms of region of interest are researched. The whole ROI can be extracted by the proposed adaptive three-dimensional region growth algorithm and the algorithm based on three-dimensional Hessian filter. The results of two algorithms are analysed. Second, a scheme for three-dimensional feature extraction is presented. In the aspect of the spatial context features, six features which consisted of surface, volume, sphericity, convexity, compactness, mean are extracted; in the aspect of the local features, the volumetric shape index features are extracted. These extracted features are identified and evaluated according to medical symptoms. Last, the fisher classifier and the kernel-fisher classifier are designed to distinguish nodules from normal areas after the effective features extraction. The experiment results indicate that the performance of modified kernel-fisher is better than that of the fisher classifier.
On the whole, the algorithms put forward in this thesis for ROI segmentation, feature extraction and classification by use of three-dimensional methods have good detected result according to the spatial context messages. And it could be used as supplementary information for the diagnosis of the pulmonary nodules.
本文对面向肺部CAD的三维RO1分割、特征提取与分类方法进行研究。主要研究工作包括三部分：首先研究了肺部感兴趣区域(ROI)的三维分割算法。包括自适应三维区域生长算法和基于三维Hessian矩阵的算法,并对两种算法分割后的结果进行分析和比较。其次研究了感兴趣区域的三维特征提取算法。从全局考虑,提取了表面积、体积、球形度、凸形度、紧凑度和灰度均值6个特征；从局部考虑,提取了肺结节的局部体积形状指数特征。并根据肺结节的CT影像特征,对所提取的特征进行分析评价。最后根据感兴趣区域提取的有效特征,应用Fisher分类器和基于核的Fisher分类器对结节区域和正常组织进行分类。实验表明,改进的基于核的Fisher分类器比Fisher分类器具有更好的分类性能和准确性。
整体而言,本文充分考虑了CT图像的上下层信息,运用三维的方法对感兴趣区域进行分割、特征提取及分类算法的研究,取得了较好的效果,可以为肺部结节的判别提供辅助诊断依据。
Lung cancer is one of the most deadly diseases in the world, it threatens people's lives. To improve lung cancer patients'viability, the earliest diagnosis and therapy of lung cancer is essential. CT scanning has been the most important method of lung cancer examination. The lung cancer is represented with the pulmonary nodules in CT images, and since the complexity of the lung structure and the multiplicity of nodule's shape, even the seasoned doctor cannot easily discover all of the possible disease in time. However, with the advent of the multislice CT, the image data which were handled by the physician drastically increased. An assistance is cried for in order to lighten physician's burdens, so computer-aided diagnosis (CAD) is developed gradually. It may help doctors to analyze pathological changes and other regions of interest in character and even in accurate quantity, and release the doctors'burden.
In this thesis,3D ROI segmentation, feature extraction and classification methods for pulmonary CAD are researched. The main work includes the following three parts:first,3D segmentation algorithms of region of interest are researched. The whole ROI can be extracted by the proposed adaptive three-dimensional region growth algorithm and the algorithm based on three-dimensional Hessian filter. The results of two algorithms are analysed. Second, a scheme for three-dimensional feature extraction is presented. In the aspect of the spatial context features, six features which consisted of surface, volume, sphericity, convexity, compactness, mean are extracted; in the aspect of the local features, the volumetric shape index features are extracted. These extracted features are identified and evaluated according to medical symptoms. Last, the fisher classifier and the kernel-fisher classifier are designed to distinguish nodules from normal areas after the effective features extraction. The experiment results indicate that the performance of modified kernel-fisher is better than that of the fisher classifier.
On the whole, the algorithms put forward in this thesis for ROI segmentation, feature extraction and classification by use of three-dimensional methods have good detected result according to the spatial context messages. And it could be used as supplementary information for the diagnosis of the pulmonary nodules.
引文
1.余松林.医学统计学[M].北京：人民卫生出版社,2002,1-350.
2. Yuan He, Advanced Conceptes for Intelligent Vision Systems [M], Berlin: Springer Berlin/Heidelberg.2006,4179:946-954.
3.吴龙海.高分辨率CT图像的肺部病变计算机辅助诊断研究[D],合肥：中国科学技术大学,2008.
4. Muhm J. R., Miler W. E., Fontana R. S.. Lung cancer detected during a screening program using four-month chest radiography [J]. Radiology,1983,148:606-615.
5.朱翠玲,赵斌,吴恩惠.现代医学影像学：工程与临床[M],济南：科学技术出版社,2000,198-293.
6.谢宝.第1讲CT读片基础知识[J],中国医刊,2000,35(12)：45-47.
7.于甬华,袁双虎.计算机辅助诊断在医学影像学领域的应用和发展[J],现代医学仪器与应用,2007,19(5)：4-7.
8. T. Okumura, TMiwa, Jun-ichiKako. Automatic Detecion of Pulmonary Cancers in Chest CT Images by Variable N-Quoit Filter [J], Fourteenth International Conference on Pattern Recognition,1998,2(8):1671-1673.
9. W. J. Kostis, A. P. Reeves, D. J. Yankelevitz. Three-dimensional segmentation of solitary pulmonary nodules from helical CT scans, Proceedings of computer- assisted radiology and surgery [J], Elsevier Science,1999:203-207.
10. Samuel G. Armato, et al. Computerized detection of pulmonary nodules on CT scans [J], RadioGraphics,1999,19:1303-1311.
11.Kawata Y., Niki H. N., Ohmatsu. Computerized analysis of 3-D pulmonary nodule images in surrounding and internal structure feature spaces [J], IEEE transactions on medical imaging,2001,2(2):889-892.
12. Y. Lee, THara, H. Fujita. Automated Detection of Pulmonary Nodules in Helical CT Images Based on an Improved Template-Matching Technique [J], IEEE Transations On Medical Imaging,2001,20(7):595-604.
13.Kenji Suzuki, Feng Li, Shusuke Sone, and Kunio Doi. Computer-aided diagnostic scheme for distinction between benign and malignant nodules in thoracic low-dose CT by use of massive training artificial neural network [J], IEEE Transactions on medical imaging,2005,24(9):1138-1150.
14. Matthew S. B., Michael F. M., Jonathan G. Patient-Specific models for lung nodule detection and surveillance in CT images [J], IEEE Trans. Med. Imaging,2001,20(12): 1242-1250.
15. Rafael W, Patrik R, Andre Z, et al. Computer aided lung nodule detection on high resolution CT data [C]. Imaging Processing, Proceedings of SPIE,2002,4684: 677-688.
16. Rafael W, Andre Z. Optimal thresholding for 3D segmentation of pulmonary nodules in high resolution CT [C], Elsevier Science International Congress,2001,1230: 653-658.
17. Carol L N, Li F. An interactive system for CT lung nodule identification and examination [C], Elsevier Science International Congress,2001,1230:639-645.
18. Li F, Carol L N, Jianzhong Q, et al. Automatic detection of lung nodules from multi-slice low-dose CT images [J]. Proc. SPIE,2001,4322:1828-1835.
19.陈旭,庄天戈.胸部高分辨率CT片中肺实质的自动分割[J],上海交通大学学报,2002,36(7)：946-948.
20.陈旭,庄天戈等.窗口条件对共生纹理在磨玻璃影自动识别中的影响[J],上海交通大学学报,2003,37(5)：91-94.
21.陈旭,庄天戈等.复杂性测度特征在肺部HRCT图像分析中的应用[J],上海交通大学学报,2003,37(2)：224-227.
22.于立燕,于甬华等.原发性肺癌孤立性结节的自动提取[J],生物医学工程研究,2004,22(4)：19-21.
23.姜兴岳,耿道颖.计算机辅助诊断在医学影像学中的应用[J],中华现代影像学杂志,2005,9(2)：823-825.
24. T. Ishida, S. Katsuragawa, K. Nakamura, H. MacMahon, et al. Iterative image warping technique for temporal subtraction of sequential chest radiographs to detect interval change [J], Med.Phys,1999,26(7):1320-1329.
25. R. Wiemker, P. Rogalla, T. Blaffert, et al. Computer-aided detection(CAD)and volumetry of pulmonary nodules on high-resolution CT data [J], Med.Amundi,2003, 47(3):37-44.
26. S.G. Armato Ⅲ, M.L.Giger,K.Ashizawa, et. al. Automated lung segmentation in digital lateral chest radiographs [J], Med.Phys,1998,25(8):1507-1519.
27. R.Wiemker, P.Rogalla, T.Blaffert, et al. Computer-aided detection(CAD)and volumetry of pulmonary nodules on high-resolution CT data [J], Med.Amundi,2003, 47(3):37-44.
28.薛以锋.计算机辅助肺癌特征提取及分类方法的研究[D],南京：东南大学,2006.
29. Doi K, MacMahon H, Katsuragawa S, et al. Computer-aided Diagnosis in Radiology: Potential and Pitfalls [J], European Journal of Radiology,1999,31:97-109.
30. Bram VG, Bart MH. Computer-aided diagnosis in chest radiography: a survey [J], IEEE Transaction on Medical Imaging,2001,20(12):1228-1241.
31. Geoffrey D, Rubin, John K, et al. Pulmonary nodules on multi-detector row CT scans: performance comparison of radiologists and computer-aided detection [J], Radiology, 2005,234:274-283.
32.Nakamura K, Yoshida H, Engelmann R, et al. Computerized analysis of the likelihood of malignancy in solitary pulmonary nodules with use of artificial neural networks [J], Radiology,2000,214:823-830.
33. Xu CY, Prince JL, Handbook of medical imaging [M], American, Academic Press, 2000,159-169.
34.吴恩惠.医影像学[M],北京：人民卫生出版社,1993,115.
35.王曙燕,周明全等.医学图像的关联规则挖掘方法研究[J],计算机应用,2005,25(6)：1408-1409.
36. A.W.K. Tang, H.A. Moss, R.J.H. Robertson. The solitary pulmonary nodule [J], European Journal of Radiology,2003,45:69-77.
37.孟菲,冯爱娟.孤立性结节影像学研究进展[J],中华综合医学杂志,2003,5(1)：29-31.
38.李惠民,肖湘声.孤立结节诊断中CT容积显示的价值[J],中国医学计算机成像杂志,2005,(1)：30-35.
39. Rafael Wiemkera, Patrick Rogalla, et al. Computer-aided segmentation of pulmonary nodules:automated vasculature cutoff in thin-and thick-slice CT [J], International Congress Series,2003,1256:965-970.
40.罗述谦等.医学图像处理与分析[M],北京：科学出版社,2003,21-25.
41. Samuel G. Armato, et al. Computerized detection of pulmonary nodules on CT scans [J], RadioGraphics,1999,19:1303-1311.
42. Shiying Hu, Eric A, et al. Automatic lung segmentation for accurate quantitation of volumetric x-ray CT images [J], IEEE Trans Med Imag,2001,20(6):490-498.
43.Dajnowiec M, et al. Computer simulation for segmentation of lung nodules in CT images [J], IEEE International Conference on Systems, Man and Cybernetics,2004, 5(5):4491-4496.
44. Okumura T, Miwa, Jun-ichi T, Kako, et al. Automatic Detection of Lung Cancers in Chest CT Images by Variable N-Quoit Filter [J], Pattern Recognition,1998,2(8): 1671-1673.
45.Aylward SR, Bullitt E. Initialization, noise, singularities, and scale in height ridge traversal for tubular object centerline extraction [J], IEEE Transactions on Medical Imaging,2002,21(2):61-75.
46. Y.Sato, T.Araki, M.Hanayama, H.Naito, and S.Tamura. A viewpoint determination system for stenosis diagnosis and quantification in coronary angiographic image acquisition [J], IEEE Trans. Med. Imaging,1998,17(1):121-137.
47. Frangi AF, Niessen WJ, Vincken KL, et al. Multiscale vessel enhancement filtering [J], Lecture Notes in Computer Science,1998,1496:130-137.
48.舒中力.基于Level Set的心血管图像中轴提取[D],南京：东南大学,2005.
49.李燕,谭鸥,段会龙.三维医学图像可视化技术综述[J],中国图像图形学报,2001,6(2)：103-110.
50. Mangasarian O L.Generalized support vector machines [C], Cambridge: MIP Press, 2006,135-146.
51. Dehmeshki J, Ye X, Costello J. Shape based region growing using derivatives of 3D medical images:application to semiautomated detection of pulmonary nodules [J], ICIP,2003,1:1085-1089.
52. Monga O, Benayoun S. Using partial derivatives of 3D images to Extract Typical Surface Features [J], Comput Vis Image Understanding,1995,61:171-189.
53.Yoshida H, Nappi J. Three-dimensional computer-aided diagnosis scheme for detection of colonic polyps [J], IEEE Trans Med Imaging,2001,20(12):1261-1274.
54.边肇祺,张学工.模式识别(第2版)[M],北京：清华大学出版社,2000,280-282.
55.袁志发,周静芋.多元统计分析[M],北京：科学出版社,2002,216-220.
56. John Shawe-Taylor, Nello Cristianini.模式分析的核方法[M],北京：机械工业出版社,2004,55-69.
2. Yuan He, Advanced Conceptes for Intelligent Vision Systems [M], Berlin: Springer Berlin/Heidelberg.2006,4179:946-954.
3.吴龙海.高分辨率CT图像的肺部病变计算机辅助诊断研究[D],合肥：中国科学技术大学,2008.
4. Muhm J. R., Miler W. E., Fontana R. S.. Lung cancer detected during a screening program using four-month chest radiography [J]. Radiology,1983,148:606-615.
5.朱翠玲,赵斌,吴恩惠.现代医学影像学：工程与临床[M],济南：科学技术出版社,2000,198-293.
6.谢宝.第1讲CT读片基础知识[J],中国医刊,2000,35(12)：45-47.
7.于甬华,袁双虎.计算机辅助诊断在医学影像学领域的应用和发展[J],现代医学仪器与应用,2007,19(5)：4-7.
8. T. Okumura, TMiwa, Jun-ichiKako. Automatic Detecion of Pulmonary Cancers in Chest CT Images by Variable N-Quoit Filter [J], Fourteenth International Conference on Pattern Recognition,1998,2(8):1671-1673.
9. W. J. Kostis, A. P. Reeves, D. J. Yankelevitz. Three-dimensional segmentation of solitary pulmonary nodules from helical CT scans, Proceedings of computer- assisted radiology and surgery [J], Elsevier Science,1999:203-207.
10. Samuel G. Armato, et al. Computerized detection of pulmonary nodules on CT scans [J], RadioGraphics,1999,19:1303-1311.
11.Kawata Y., Niki H. N., Ohmatsu. Computerized analysis of 3-D pulmonary nodule images in surrounding and internal structure feature spaces [J], IEEE transactions on medical imaging,2001,2(2):889-892.
12. Y. Lee, THara, H. Fujita. Automated Detection of Pulmonary Nodules in Helical CT Images Based on an Improved Template-Matching Technique [J], IEEE Transations On Medical Imaging,2001,20(7):595-604.
13.Kenji Suzuki, Feng Li, Shusuke Sone, and Kunio Doi. Computer-aided diagnostic scheme for distinction between benign and malignant nodules in thoracic low-dose CT by use of massive training artificial neural network [J], IEEE Transactions on medical imaging,2005,24(9):1138-1150.
14. Matthew S. B., Michael F. M., Jonathan G. Patient-Specific models for lung nodule detection and surveillance in CT images [J], IEEE Trans. Med. Imaging,2001,20(12): 1242-1250.
15. Rafael W, Patrik R, Andre Z, et al. Computer aided lung nodule detection on high resolution CT data [C]. Imaging Processing, Proceedings of SPIE,2002,4684: 677-688.
16. Rafael W, Andre Z. Optimal thresholding for 3D segmentation of pulmonary nodules in high resolution CT [C], Elsevier Science International Congress,2001,1230: 653-658.
17. Carol L N, Li F. An interactive system for CT lung nodule identification and examination [C], Elsevier Science International Congress,2001,1230:639-645.
18. Li F, Carol L N, Jianzhong Q, et al. Automatic detection of lung nodules from multi-slice low-dose CT images [J]. Proc. SPIE,2001,4322:1828-1835.
19.陈旭,庄天戈.胸部高分辨率CT片中肺实质的自动分割[J],上海交通大学学报,2002,36(7)：946-948.
20.陈旭,庄天戈等.窗口条件对共生纹理在磨玻璃影自动识别中的影响[J],上海交通大学学报,2003,37(5)：91-94.
21.陈旭,庄天戈等.复杂性测度特征在肺部HRCT图像分析中的应用[J],上海交通大学学报,2003,37(2)：224-227.
22.于立燕,于甬华等.原发性肺癌孤立性结节的自动提取[J],生物医学工程研究,2004,22(4)：19-21.
23.姜兴岳,耿道颖.计算机辅助诊断在医学影像学中的应用[J],中华现代影像学杂志,2005,9(2)：823-825.
24. T. Ishida, S. Katsuragawa, K. Nakamura, H. MacMahon, et al. Iterative image warping technique for temporal subtraction of sequential chest radiographs to detect interval change [J], Med.Phys,1999,26(7):1320-1329.
25. R. Wiemker, P. Rogalla, T. Blaffert, et al. Computer-aided detection(CAD)and volumetry of pulmonary nodules on high-resolution CT data [J], Med.Amundi,2003, 47(3):37-44.
26. S.G. Armato Ⅲ, M.L.Giger,K.Ashizawa, et. al. Automated lung segmentation in digital lateral chest radiographs [J], Med.Phys,1998,25(8):1507-1519.
27. R.Wiemker, P.Rogalla, T.Blaffert, et al. Computer-aided detection(CAD)and volumetry of pulmonary nodules on high-resolution CT data [J], Med.Amundi,2003, 47(3):37-44.
28.薛以锋.计算机辅助肺癌特征提取及分类方法的研究[D],南京：东南大学,2006.
29. Doi K, MacMahon H, Katsuragawa S, et al. Computer-aided Diagnosis in Radiology: Potential and Pitfalls [J], European Journal of Radiology,1999,31:97-109.
30. Bram VG, Bart MH. Computer-aided diagnosis in chest radiography: a survey [J], IEEE Transaction on Medical Imaging,2001,20(12):1228-1241.
31. Geoffrey D, Rubin, John K, et al. Pulmonary nodules on multi-detector row CT scans: performance comparison of radiologists and computer-aided detection [J], Radiology, 2005,234:274-283.
32.Nakamura K, Yoshida H, Engelmann R, et al. Computerized analysis of the likelihood of malignancy in solitary pulmonary nodules with use of artificial neural networks [J], Radiology,2000,214:823-830.
33. Xu CY, Prince JL, Handbook of medical imaging [M], American, Academic Press, 2000,159-169.
34.吴恩惠.医影像学[M],北京：人民卫生出版社,1993,115.
35.王曙燕,周明全等.医学图像的关联规则挖掘方法研究[J],计算机应用,2005,25(6)：1408-1409.
36. A.W.K. Tang, H.A. Moss, R.J.H. Robertson. The solitary pulmonary nodule [J], European Journal of Radiology,2003,45:69-77.
37.孟菲,冯爱娟.孤立性结节影像学研究进展[J],中华综合医学杂志,2003,5(1)：29-31.
38.李惠民,肖湘声.孤立结节诊断中CT容积显示的价值[J],中国医学计算机成像杂志,2005,(1)：30-35.
39. Rafael Wiemkera, Patrick Rogalla, et al. Computer-aided segmentation of pulmonary nodules:automated vasculature cutoff in thin-and thick-slice CT [J], International Congress Series,2003,1256:965-970.
40.罗述谦等.医学图像处理与分析[M],北京：科学出版社,2003,21-25.
41. Samuel G. Armato, et al. Computerized detection of pulmonary nodules on CT scans [J], RadioGraphics,1999,19:1303-1311.
42. Shiying Hu, Eric A, et al. Automatic lung segmentation for accurate quantitation of volumetric x-ray CT images [J], IEEE Trans Med Imag,2001,20(6):490-498.
43.Dajnowiec M, et al. Computer simulation for segmentation of lung nodules in CT images [J], IEEE International Conference on Systems, Man and Cybernetics,2004, 5(5):4491-4496.
44. Okumura T, Miwa, Jun-ichi T, Kako, et al. Automatic Detection of Lung Cancers in Chest CT Images by Variable N-Quoit Filter [J], Pattern Recognition,1998,2(8): 1671-1673.
45.Aylward SR, Bullitt E. Initialization, noise, singularities, and scale in height ridge traversal for tubular object centerline extraction [J], IEEE Transactions on Medical Imaging,2002,21(2):61-75.
46. Y.Sato, T.Araki, M.Hanayama, H.Naito, and S.Tamura. A viewpoint determination system for stenosis diagnosis and quantification in coronary angiographic image acquisition [J], IEEE Trans. Med. Imaging,1998,17(1):121-137.
47. Frangi AF, Niessen WJ, Vincken KL, et al. Multiscale vessel enhancement filtering [J], Lecture Notes in Computer Science,1998,1496:130-137.
48.舒中力.基于Level Set的心血管图像中轴提取[D],南京：东南大学,2005.
49.李燕,谭鸥,段会龙.三维医学图像可视化技术综述[J],中国图像图形学报,2001,6(2)：103-110.
50. Mangasarian O L.Generalized support vector machines [C], Cambridge: MIP Press, 2006,135-146.
51. Dehmeshki J, Ye X, Costello J. Shape based region growing using derivatives of 3D medical images:application to semiautomated detection of pulmonary nodules [J], ICIP,2003,1:1085-1089.
52. Monga O, Benayoun S. Using partial derivatives of 3D images to Extract Typical Surface Features [J], Comput Vis Image Understanding,1995,61:171-189.
53.Yoshida H, Nappi J. Three-dimensional computer-aided diagnosis scheme for detection of colonic polyps [J], IEEE Trans Med Imaging,2001,20(12):1261-1274.
54.边肇祺,张学工.模式识别(第2版)[M],北京：清华大学出版社,2000,280-282.
55.袁志发,周静芋.多元统计分析[M],北京：科学出版社,2002,216-220.
56. John Shawe-Taylor, Nello Cristianini.模式分析的核方法[M],北京：机械工业出版社,2004,55-69.