胸片计算机辅助诊断——系统有效性、图像的一致化处理及去噪
|
摘要
随着医学影像数字化的发展,对影像的智能化理解成为一种必然趋势,计算机辅助诊断(Computer-aided Diagnosis,CAD)系统已经成为了医学影像学研究的热点之一,并逐步进入了医学临床应用。计算机辅助诊断系统的目的在于提高读片医生的工作质量和效率,提高诊断的准确性和一致性,减少读片时间,降低工作强度。
Χ线胸片结合CAD系统进行肺癌普查是一个较为经济可行的方案。CR(计算机摄影)、DR(计算机数字摄影)的普遍应用以及计算机辅助诊断系统被用于X线普查的辅助,使得孤立性肺结节病变的发现率不断提高。
文章在介绍CAD系统基本概念的基础上,阐述了Kodak ChestCAD系统在胸片疾病诊断中的基本工作流程,并通过实验验证了该系统在帮助医生诊断胸片结节时的有效性。
接下来,文章从图像的一致化处理和保留纹理细节的噪声去除两方面入手,增强图像质量,从而达到提高CAD系统的性能和帮助医生诊断的目的。
一方面,由于医学成像设备的参数,病人摆位和成像环境的不同,使得X线胸片在亮度和对比度也经常存在极大差异,可能导致CAD系统检测结果不准确。对于这个问题,我们通过构建一条映射曲线设计了一种胸片图像的一致性处理算法,并通过实验验证了算法的有效性。
另一方面,明显的噪声也会显著地降低CAD系统检测结果的准确性。CAD系统在检测病灶时都依赖于纹理细节,而许多图像处理方法在去除噪声时也使得图像中的重要纹理细节信息因为被模糊而丢失。我们提出了一种新的基于上下文的自适应图像去噪算法,它能够有效地保持图像的强弱纹理边界,并通过实验验证了噪声图像经过该方法去噪后可以提高CAD系统的性能。
Due to the digitization of medical imaging, the intellectual understanding of medical images by computers has become an inevitable trend. The study of computer-aided diagnosis (CAD) is already a popular research field with many CAD systems gradually entering the stage of clinical application. The purpose of introducing CAD system is to improve the performance and efficiency of radiologists, increase accuracy and consistency in the diagnosis process and reduce working time and strength.
The combination of chest X-rays and CAD system is a practical and economical way of conducting large-scale lung cancer screening. The detection rate for solitary pulmonary nodules increases due to the massive use of CR(Computed Radiography) and DR(Digital Radiography) and also the assistance of CAD systems in the screening process.
This paper introduces some basic concepts of CAD system, elaborates on the basic work flow of Kodak ChestCAD system in the diagnosis of chest radiography and proves the effectiveness of this system in helping radiologists locating lung nodules.
The following part of this paper aims at improving image quality by consistent rendering and denoising, so as to improve the performance of the CAD system and facilitate the diagnosis for radiologists.
On one hand, the brightness and contrast of chest X-rays vary greatly because of the difference in the parameters of imaging equipment, patients’positions and imaging environments, which probably causes some inaccuracy in the diagnostic result of CAD systems. To approach this problem, we designed an image consistent rendering process for chest X-rays based on constructing a proper LUT mapping curve and proved its effectiveness by experiments.
On the other hand, obvious noise also decreases the accuracy in the diagnostic result of CAD system. The CAD algorithms rely on the texture details to detect lesions, but those details can be blurred and even lost by many existing denoising methods. We propose a novel context-based image denoising method. It can effectively preserve the strong and weak edges in the image and experimental results show that processing noised images with this method can improve the performance of CAD system.
Χ线胸片结合CAD系统进行肺癌普查是一个较为经济可行的方案。CR(计算机摄影)、DR(计算机数字摄影)的普遍应用以及计算机辅助诊断系统被用于X线普查的辅助,使得孤立性肺结节病变的发现率不断提高。
文章在介绍CAD系统基本概念的基础上,阐述了Kodak ChestCAD系统在胸片疾病诊断中的基本工作流程,并通过实验验证了该系统在帮助医生诊断胸片结节时的有效性。
接下来,文章从图像的一致化处理和保留纹理细节的噪声去除两方面入手,增强图像质量,从而达到提高CAD系统的性能和帮助医生诊断的目的。
一方面,由于医学成像设备的参数,病人摆位和成像环境的不同,使得X线胸片在亮度和对比度也经常存在极大差异,可能导致CAD系统检测结果不准确。对于这个问题,我们通过构建一条映射曲线设计了一种胸片图像的一致性处理算法,并通过实验验证了算法的有效性。
另一方面,明显的噪声也会显著地降低CAD系统检测结果的准确性。CAD系统在检测病灶时都依赖于纹理细节,而许多图像处理方法在去除噪声时也使得图像中的重要纹理细节信息因为被模糊而丢失。我们提出了一种新的基于上下文的自适应图像去噪算法,它能够有效地保持图像的强弱纹理边界,并通过实验验证了噪声图像经过该方法去噪后可以提高CAD系统的性能。
Due to the digitization of medical imaging, the intellectual understanding of medical images by computers has become an inevitable trend. The study of computer-aided diagnosis (CAD) is already a popular research field with many CAD systems gradually entering the stage of clinical application. The purpose of introducing CAD system is to improve the performance and efficiency of radiologists, increase accuracy and consistency in the diagnosis process and reduce working time and strength.
The combination of chest X-rays and CAD system is a practical and economical way of conducting large-scale lung cancer screening. The detection rate for solitary pulmonary nodules increases due to the massive use of CR(Computed Radiography) and DR(Digital Radiography) and also the assistance of CAD systems in the screening process.
This paper introduces some basic concepts of CAD system, elaborates on the basic work flow of Kodak ChestCAD system in the diagnosis of chest radiography and proves the effectiveness of this system in helping radiologists locating lung nodules.
The following part of this paper aims at improving image quality by consistent rendering and denoising, so as to improve the performance of the CAD system and facilitate the diagnosis for radiologists.
On one hand, the brightness and contrast of chest X-rays vary greatly because of the difference in the parameters of imaging equipment, patients’positions and imaging environments, which probably causes some inaccuracy in the diagnostic result of CAD systems. To approach this problem, we designed an image consistent rendering process for chest X-rays based on constructing a proper LUT mapping curve and proved its effectiveness by experiments.
On the other hand, obvious noise also decreases the accuracy in the diagnostic result of CAD system. The CAD algorithms rely on the texture details to detect lesions, but those details can be blurred and even lost by many existing denoising methods. We propose a novel context-based image denoising method. It can effectively preserve the strong and weak edges in the image and experimental results show that processing noised images with this method can improve the performance of CAD system.
引文
[1] Cancer Facts & Figures 2008. American Cancer Society. http://www.cancer.org.
[2]彭卫军,刘士远.肺癌筛查方法及其评价[J].中华肿瘤防治杂志.2006,13(17):1-2.
[3]彭敏,李龙芸.肺癌的早期诊断[J].中华内科杂志.2003,42(12):884-886.
[4] Tang, A.W.K., Moss, H.A., Robertson, R.J.H. The solitary pulmonary nodule[J]. European Journal of Radiology. 2003, 45(1):69-77.
[5]郭启勇.实用放射学(第3版)[M].北京:人民卫生出版社.2007:512-513.
[6] Swensen, S.J., Silverstein, M.D., Edell E.S., et al. Solitary pulmonary nodules: Clinical prediction model versus physicians[J]. Mayo Clinic Proceedings. 1999, 74(4): 319-329.
[7] Murthy, S.C., Rice, T.W. The solitary pulmonary nodule: a primer on differential diagnosis[J]. Seminars in Thoracic and Cardiovascular Surgery. 2002, 14(3): 239-249.
[8]蒋磊,吴建彬,左翔,薛祖平,陈福康.胸部直接数字化摄影在肺癌筛查中的价值[J].实用医技杂志.2008,15(19):2577-2578.
[9] McAdams, H.P., Samei, E., Dobbins III, J., et al. Recent Advances in Chest Radiography[J]. Radiology. 2006, 241: 663-683.
[10] Doi, K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential[J]. Computerized Medical Imaging and Graphics. 2007, 31(4-5): 198-211.
[11] Doi, K. Current status and future potential of computer-aided diagnosis in medical imaging[J]. British Journal of Radiology. 2005, 78: S3-S19.
[12] Verma, B.S., Indrajit I. Advent of digital radiography: Part 1 [J]. Indian Journal of Radiology and Imaging. 2008, 18(2): 113-116.
[13] Ogawa, E., Satoshi A. Quantitative analysis of imaging performance for computedradiography systems. Proceedings of SPIE. 1995, 2432: 421-431.
[14] Schaefer-Prokop, C., Neitzel, U., Venema, H.W., et al. Digital chest radiography: An update on modern technology, dose containment and control of image quality[J]. European Radiology. 2008, 18(9): 1818-1830.
[15] Van Ginneken, B., Ter Haar Romeny, B.M., Viergever, M.A. Computer-aided diagnosis in chest radiography: A survey[J]. IEEE Transactions on Medical Imaging. 2001, 20(12): 1228-1241.
[16] Yoshida, H. Local contralateral subtraction based on bilateral symmetry of lung for reduction of false positives in computerized detection of pulmonary nodules[J]. IEEE Transactions on Biomedical Engineering. 2004, 51(5): 778-789.
[17] Hardie, R.C., Rogers, S.K., Wilson, T., et al. Performance analysis of a new computer aided detection system for identifying lung nodules on chest radiographs[J]. Medical Image Analysis. 2008, 12(3): 240-258.
[18] Nakamura, K., Yoshida, M., 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(3): 823-830.
[19] Schilham, A.M., Van Ginneken, B., Loog, M. A computer-aided diagnosis system for detection of lung nodules in chest radiographs with an evaluation on a public database[J]. Medical Image Analysis. 2006, 10(2): 247-258.
[20] Forrest, J.V., Friedman, P.J. Radiologic errors in patients with lung cancer[J]. Western Journal of Medicine. 1981, 134(6): 485-490.
[21] Arzhaeva, Y., Tax, D., Van Ginneken, B. Improving computer-aided diagnosis of interstitial disease in chest radiographs by combining one-class and two-class classifiers. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2006, 6144 III, 614458.
[22] Abe, H., MacMahon, H., Shiraishi, J. et al. Computer-aided diagnosis in chest radiology.Seminars in Ultrasound, CT and MRI. 2004, 25(5): 432-437.
[23] Seghers, D., Loeckx, D., Maes, F., et al. Visual enhancement of interval changes using a temporal subtraction technique. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2007, 6512(Part 3), 65124R.
[24] Schildkraut, J.S., Heath, M.D., Senn, R.A., et al. Pulminary nodule detection in a chest radiograph. US Patent: WO/2007/018889.
[25] Cootes, T.F., Taylor, C.J., Cooper, D.H., et al. Active shape models-their training and application[J]. Computer Vision and Image Understanding. 1995, 61(1): 38-59.
[26] Lusted, L.B. Logical analysis in roentgen diagnosis[J]. Radiology. 1960, 74: 178-193.
[27] Dorfman, D.D., Berbaum, K.S., Metz, C.E. Receiver operating characteristic rating analysis: Generalization to the population of readers and patients with the Jackknife method[J]. Investigative Radiology. 1992, 27(9): 723-731.
[28] Hillis, S.L., Obuchowski, N., Schartz, K., et al. A comparison of the Dorfman-Berbaum-Metz and Obuchowshi-Rockette methods for receiver operating characteristic (ROC) data[J]. Statistics in Medicine. 2005, 24: 1579-1607.
[29] Hillis, S.L., Berbaum, K.S. Monte Carlo validation of the dorfman-berbaum-metz method using normalized pseudovalues and less data-based model simplification[J]. Academic Radiology. 2005, 12(12): 1534-1541.
[30] Matsumoto, T., Yoshimura, H., Giger, M.L., et al. Potential usefulness of computerized nodule detection in screening programs for lung cancer[J]. Investigative Radiology. 1992, 27(6): 471-475.
[31] Kobayshi, T., Xu, X.W., MachMahon, H., et al. Effect of a computer-aided diagnosis scheme on radiologists' performance in detection of lung nodules on radiographs[J]. Radiology. 1996, 199(3): 843-848.
[32] Kakeda, S., Moriya, J., Sato, H., et al. Improved detection of lung nodules on chest radiographs using a commercial computer-aided diagnosis system[J]. American Journal of Roentgenology. 2004, 182(2): 505-510.
[33]蔡宇新,徐涛.基于ASM的图像中二维物体的定位方法研究[J].计算机应用.2003,23(6): 191-194.
[34] Couwenhoven, M., Senn, R., Foos, D. Enhancement method that provides direct and independent control of fundamental attributes of image quality for radiographic imagery. Proceedings of SPIE. 2004, 5367: 474-481.
[35] Pilkinton, D., Bitter, I., Summers, R.M., et al. The effect of edge-preserving image smoothing on automatic colonic polyp detection for CT colonography[J]. Progress in Biomedical Optics and Imaging. 2006, 6143: 984-991.
[36] Sinha, P.K., Hong, Q. An improved median filter[J]. IEEE Transactions on Medical Imaging. 1990, 9(3): 345-346.
[37] Elad, M. On the origin of the bilateral filter and ways to improve it[J]. IEEE Transactions on Image Processing. 2002, 11(10): 1141-1151.
[38] Keeling, S.L. Total variation based convex filters for medical imaging[J]. Applied Mathematics and Computation. 2003, 139(1): 101-119.
[39] Bouman, C., Sauer, K. A generalized Gaussian image model for edge-preserving MAP estimation[J]. IEEE Transactions on Image Processing. 1993, 2(3): 296-310.
[40] Sanches, J.M., Nascimento, J.C., Marques, J.S. Medical Image Noise Reduction Using the Sylvester–Lyapunov Equation[J]. Image Processing,IEEE Transactions on Medical Imaging. 2008, 17(9): 1522-1539.
[41] Schulze, M.A. Biomedical image processing with morphology-based nonlinear filters[D]. The University of Texas at Austin, 1994.
[42] Xiaolin, W., Nasir, M. Context-Based, adaptive, lossless image coding[J]. IEEE Transactions of Communications. 1997, 45(4): 437-444.
[43] Shiraishi, J., Katsuragawa, S., Ikezoe, J., et al. Development of a digital image database for chest radiographs with and without a lung nodule: Receiver operating characteristic analysis of radiologists' detection of pulmonary nodules[J]. American Journal of Roentgenology. 2000, 174(1): 71-74.
[44] Schilham, A., Van Ginneken, B. Simulating nodules in chest radiographs with real nodules from multi-slice CT images. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2006, 6144 III, 614456.
[45] Samei, E., Stebbins, S.A., Dobbins III, J.T., et al. Multiprojection correlation imaging for improved detection of pulmonary nodules[J]. American Journal of Roentgenology. 2007, 188(5): 1239-1245.
[46] Yeh, C., Wang, J.F., Wu, M.T., et al. A comparative study for 2D and 3D computer-aided diagnosis methods for solitary pulmonary nodules[J]. Computerized Medical Imaging and Graphics. 2008, 32(4): 270-276.
[47] Katsuragawa, S., Doi, K. Computer-aided diagnosis in chest radiography[J]. Computerized Medical Imaging and Graphics. 2007, 31(4-5): 212-223.
[48] Le, A., Mai, L., Liu, B.,The workflow and procedures for automatic integration of a computer-aided diagnosis workstation with a clinical PACS with real world examples. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2008, 6919, 69190U.
[2]彭卫军,刘士远.肺癌筛查方法及其评价[J].中华肿瘤防治杂志.2006,13(17):1-2.
[3]彭敏,李龙芸.肺癌的早期诊断[J].中华内科杂志.2003,42(12):884-886.
[4] Tang, A.W.K., Moss, H.A., Robertson, R.J.H. The solitary pulmonary nodule[J]. European Journal of Radiology. 2003, 45(1):69-77.
[5]郭启勇.实用放射学(第3版)[M].北京:人民卫生出版社.2007:512-513.
[6] Swensen, S.J., Silverstein, M.D., Edell E.S., et al. Solitary pulmonary nodules: Clinical prediction model versus physicians[J]. Mayo Clinic Proceedings. 1999, 74(4): 319-329.
[7] Murthy, S.C., Rice, T.W. The solitary pulmonary nodule: a primer on differential diagnosis[J]. Seminars in Thoracic and Cardiovascular Surgery. 2002, 14(3): 239-249.
[8]蒋磊,吴建彬,左翔,薛祖平,陈福康.胸部直接数字化摄影在肺癌筛查中的价值[J].实用医技杂志.2008,15(19):2577-2578.
[9] McAdams, H.P., Samei, E., Dobbins III, J., et al. Recent Advances in Chest Radiography[J]. Radiology. 2006, 241: 663-683.
[10] Doi, K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential[J]. Computerized Medical Imaging and Graphics. 2007, 31(4-5): 198-211.
[11] Doi, K. Current status and future potential of computer-aided diagnosis in medical imaging[J]. British Journal of Radiology. 2005, 78: S3-S19.
[12] Verma, B.S., Indrajit I. Advent of digital radiography: Part 1 [J]. Indian Journal of Radiology and Imaging. 2008, 18(2): 113-116.
[13] Ogawa, E., Satoshi A. Quantitative analysis of imaging performance for computedradiography systems. Proceedings of SPIE. 1995, 2432: 421-431.
[14] Schaefer-Prokop, C., Neitzel, U., Venema, H.W., et al. Digital chest radiography: An update on modern technology, dose containment and control of image quality[J]. European Radiology. 2008, 18(9): 1818-1830.
[15] Van Ginneken, B., Ter Haar Romeny, B.M., Viergever, M.A. Computer-aided diagnosis in chest radiography: A survey[J]. IEEE Transactions on Medical Imaging. 2001, 20(12): 1228-1241.
[16] Yoshida, H. Local contralateral subtraction based on bilateral symmetry of lung for reduction of false positives in computerized detection of pulmonary nodules[J]. IEEE Transactions on Biomedical Engineering. 2004, 51(5): 778-789.
[17] Hardie, R.C., Rogers, S.K., Wilson, T., et al. Performance analysis of a new computer aided detection system for identifying lung nodules on chest radiographs[J]. Medical Image Analysis. 2008, 12(3): 240-258.
[18] Nakamura, K., Yoshida, M., 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(3): 823-830.
[19] Schilham, A.M., Van Ginneken, B., Loog, M. A computer-aided diagnosis system for detection of lung nodules in chest radiographs with an evaluation on a public database[J]. Medical Image Analysis. 2006, 10(2): 247-258.
[20] Forrest, J.V., Friedman, P.J. Radiologic errors in patients with lung cancer[J]. Western Journal of Medicine. 1981, 134(6): 485-490.
[21] Arzhaeva, Y., Tax, D., Van Ginneken, B. Improving computer-aided diagnosis of interstitial disease in chest radiographs by combining one-class and two-class classifiers. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2006, 6144 III, 614458.
[22] Abe, H., MacMahon, H., Shiraishi, J. et al. Computer-aided diagnosis in chest radiology.Seminars in Ultrasound, CT and MRI. 2004, 25(5): 432-437.
[23] Seghers, D., Loeckx, D., Maes, F., et al. Visual enhancement of interval changes using a temporal subtraction technique. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2007, 6512(Part 3), 65124R.
[24] Schildkraut, J.S., Heath, M.D., Senn, R.A., et al. Pulminary nodule detection in a chest radiograph. US Patent: WO/2007/018889.
[25] Cootes, T.F., Taylor, C.J., Cooper, D.H., et al. Active shape models-their training and application[J]. Computer Vision and Image Understanding. 1995, 61(1): 38-59.
[26] Lusted, L.B. Logical analysis in roentgen diagnosis[J]. Radiology. 1960, 74: 178-193.
[27] Dorfman, D.D., Berbaum, K.S., Metz, C.E. Receiver operating characteristic rating analysis: Generalization to the population of readers and patients with the Jackknife method[J]. Investigative Radiology. 1992, 27(9): 723-731.
[28] Hillis, S.L., Obuchowski, N., Schartz, K., et al. A comparison of the Dorfman-Berbaum-Metz and Obuchowshi-Rockette methods for receiver operating characteristic (ROC) data[J]. Statistics in Medicine. 2005, 24: 1579-1607.
[29] Hillis, S.L., Berbaum, K.S. Monte Carlo validation of the dorfman-berbaum-metz method using normalized pseudovalues and less data-based model simplification[J]. Academic Radiology. 2005, 12(12): 1534-1541.
[30] Matsumoto, T., Yoshimura, H., Giger, M.L., et al. Potential usefulness of computerized nodule detection in screening programs for lung cancer[J]. Investigative Radiology. 1992, 27(6): 471-475.
[31] Kobayshi, T., Xu, X.W., MachMahon, H., et al. Effect of a computer-aided diagnosis scheme on radiologists' performance in detection of lung nodules on radiographs[J]. Radiology. 1996, 199(3): 843-848.
[32] Kakeda, S., Moriya, J., Sato, H., et al. Improved detection of lung nodules on chest radiographs using a commercial computer-aided diagnosis system[J]. American Journal of Roentgenology. 2004, 182(2): 505-510.
[33]蔡宇新,徐涛.基于ASM的图像中二维物体的定位方法研究[J].计算机应用.2003,23(6): 191-194.
[34] Couwenhoven, M., Senn, R., Foos, D. Enhancement method that provides direct and independent control of fundamental attributes of image quality for radiographic imagery. Proceedings of SPIE. 2004, 5367: 474-481.
[35] Pilkinton, D., Bitter, I., Summers, R.M., et al. The effect of edge-preserving image smoothing on automatic colonic polyp detection for CT colonography[J]. Progress in Biomedical Optics and Imaging. 2006, 6143: 984-991.
[36] Sinha, P.K., Hong, Q. An improved median filter[J]. IEEE Transactions on Medical Imaging. 1990, 9(3): 345-346.
[37] Elad, M. On the origin of the bilateral filter and ways to improve it[J]. IEEE Transactions on Image Processing. 2002, 11(10): 1141-1151.
[38] Keeling, S.L. Total variation based convex filters for medical imaging[J]. Applied Mathematics and Computation. 2003, 139(1): 101-119.
[39] Bouman, C., Sauer, K. A generalized Gaussian image model for edge-preserving MAP estimation[J]. IEEE Transactions on Image Processing. 1993, 2(3): 296-310.
[40] Sanches, J.M., Nascimento, J.C., Marques, J.S. Medical Image Noise Reduction Using the Sylvester–Lyapunov Equation[J]. Image Processing,IEEE Transactions on Medical Imaging. 2008, 17(9): 1522-1539.
[41] Schulze, M.A. Biomedical image processing with morphology-based nonlinear filters[D]. The University of Texas at Austin, 1994.
[42] Xiaolin, W., Nasir, M. Context-Based, adaptive, lossless image coding[J]. IEEE Transactions of Communications. 1997, 45(4): 437-444.
[43] Shiraishi, J., Katsuragawa, S., Ikezoe, J., et al. Development of a digital image database for chest radiographs with and without a lung nodule: Receiver operating characteristic analysis of radiologists' detection of pulmonary nodules[J]. American Journal of Roentgenology. 2000, 174(1): 71-74.
[44] Schilham, A., Van Ginneken, B. Simulating nodules in chest radiographs with real nodules from multi-slice CT images. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2006, 6144 III, 614456.
[45] Samei, E., Stebbins, S.A., Dobbins III, J.T., et al. Multiprojection correlation imaging for improved detection of pulmonary nodules[J]. American Journal of Roentgenology. 2007, 188(5): 1239-1245.
[46] Yeh, C., Wang, J.F., Wu, M.T., et al. A comparative study for 2D and 3D computer-aided diagnosis methods for solitary pulmonary nodules[J]. Computerized Medical Imaging and Graphics. 2008, 32(4): 270-276.
[47] Katsuragawa, S., Doi, K. Computer-aided diagnosis in chest radiography[J]. Computerized Medical Imaging and Graphics. 2007, 31(4-5): 212-223.
[48] Le, A., Mai, L., Liu, B.,The workflow and procedures for automatic integration of a computer-aided diagnosis workstation with a clinical PACS with real world examples. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2008, 6919, 69190U.