基于MRI的HIFU靶区及声通道组织自动识别研究
|
摘要
高强聚焦超声刀(High Intensity Focused Ultrasound, HIFU),是一种新型的现代无创、微创治疗高科技设备,通过体外照射治疗机体内部实体肿瘤。有别于其它现代相关治疗设备,其作用机理上是将高能量的超声波在作用于治疗靶区,瞬间使指定的组织出现高温凝固坏死。放射治疗技术是与类似的肿瘤治疗手段中应用最广泛的体外照射技术。放射治疗是通过射线治疗过程使组织逐渐变性坏死,其疗效一般在数周才予体现。靶区的误差一般不会造成病人立即死亡等安全问题。因此,在实际临床工作中为了提高肿瘤的放射治疗效果,放射区域常常是大于肿瘤实际靶区,行业上已经确定了靶区定位误差3cm许可的准则。然而如果我们将这种准则用于HIFU治疗,将会造成重要脏器结构损伤的后果。
对一些特殊部位或区域,为了安全,HIFU治疗的治疗靶区就会小于肿瘤实际区域,结果将可能会影响治疗效果;因此,以精确定位为基础的HIFU精确治疗对提高HIFU肿瘤治疗效果有着十分重要的意义。
开展有关基于核磁共振成像(Magnetic Resonance Imaging, MRI)的计算机组织识别技术的研究,是建立HIFU精确定位、精确治疗的基础,也为HIFU治疗人员对声通道内的重要组织结构的识别、声阻抗的计算及最佳治疗通道的选择提供依据。
出于这方面需要,开展了“基于MRI影像的HIFU治疗靶区及声通道的组织自动识别”项目的研究工作,根据相关研究的结果撰写了这份报告。作者本人曾做了大量文献检索工作,迄今为止,尚未发现类似的研究成果或报告。
本项目研究的目的主要包括三个方面:(1)通过分析MRI图像,了解人体组织结构在MR值上的共同特性及其影响因素;(2)在人体组织结构特性研究的基础上,建立基于MRI图像的计算机组织识别模式;(3)基于MRI图像的计算机组织识别模式,结合相关的图像处理技术建立基于MRI、CT图像的HIFU治疗靶区自动勾画系统,提高HIFU治疗效果与安全性。其最终目标是为了建立适形的、精确的能量投放HIFU治疗计划系统提供依据与技术支撑。为了实现这一目标,我们把本项目的具体工作分成了以下二个部分。
第一部分人体组织的MRI影像特性研究
以子宫肌瘤为研究对象,开展人体组织结构(包括肿瘤)在MRI影像上的共同特性的分析研究。
研究方法:采用在Microsoft Visual C++ 6.0环境下为本研究专门开发的“专用MRI图像分析软件”,用配对设计的统计分析方法对所研究的子宫肌瘤的MRI T2原始序列图像的数据进行分析,研究它们在MR值上的变化规律。
研究材料:重庆医科大学附属二院GE MRI设备上检查获取的DICOM格式的子宫肌瘤原始图像数据。
结果显示:(1)在同一检查条件下,同一患者不同时期的子宫肌瘤影像中,MR值分布范围没有显著差异;(2)在同一检查条件下,不同患者间的子宫肌瘤影像中,MR值分布范围没有显著差异;(3)在同一检查条件下,子宫肌瘤在HIFU治疗前、后的MR值变化范围存在显著差异;(4)在子宫肌瘤HIFU声通道中可能涉及的几个主要组织结构(腹壁肌肉、充盈膀胱、正常子宫组织及子宫肌瘤)间,在MR值分布范围上存在显著差异。
结论:由此推断,在同一MRI检查设备,相同的检查条件下的MRI影像上,人体组织结构中同一组织结构存在一定的共性,而不同的组织结构存在一定的差异。这种共性和差异性从效果看,可以作为计算机进行自动识别组织的依据,为HIFU治疗声通道的声阻抗的精确计算和靶区的精确勾画打下基础,为进一步开发与完善HIFU治疗计划系统提供了条件。另外,HIFU治疗可以改变原有组织结构的MRI图像特性,这有助于研究新的HIFU治疗效果评价指标,对建立HIFU治疗评价体系有一定意义。
第二部分MRI组织识别模式的建立与靶区自动勾画研究
研究的内容主要包括:(1)以MRI等图像为基础,对生物组织结构进行识别分析的技术实施研究;(2)为HIFU治疗建立一个比较精确的靶区自动勾画系统,从而进一步地三维成像,为HIFU治疗精确控制提供依据,也有利于建立更为完整的HIFU治疗计划系统。
研究方法:以MRI等原始图像为背景,利用相关的计算机识别技术,建立对靶区组织结构的自动识别模式,对所识别的组织结构进行进一步分割等处理,最后勾画出较为精确的靶区轮廓。研究中发现,没有任何一种技术或算法能够独立完成图像自动勾画的要求,必须借助多种技术共同完成。因此,在通过反复实验与对比研究分析后,确定了以点、面结合的组合图像处理技术方案,其中以基于MRI特性的分割阈值为特点的靶区图像分割和以区域生长算法为基础的图像模板匹配技术为本研究中图像组织结构识别技术方案的核心。此外,研究中还应用了包括图像分割阈值获取、多种分割方法的组合采用、组织结构提取等技术方法,并结合研究要求探讨了相关原理及具体算法。
结果与结论:通过试验发现(1)此技术方案能够比较准确识别诸如MRI类图像中的指定组织结构;(2)针对医学图像(MRI、CT等),能够比较快速、准确的对治疗靶区中边界相对不模糊的图像可达到一次性准确勾画的效果。因此,此项研究结果将会有助于HIFU、放射等医学治疗中的精确定位与治疗,有较大的实际应用价值。此外,此方案由于在对医学图像(MRI、CT等)进行处理的同时,结合了相关的MRI、CT显像组织结构特性分析技术,可帮助医务人员识别一些特定组织结构,提高相应的诊治水平,对建立基于MRI的智能分析诊断类的医疗专家系统的开发与研究方面的工作有所帮助。其勾画方案,除可应用于HIFU外,也可辅助用于其它一些医学治疗手段中,如三维立体放射治疗、超级γ刀的放射治疗计划系统中,可帮助提高精确定位、减少并发症发生率及重要组织结构的保护作用,并可能有助于医疗技术的进一步发展与提高有着十分重要的意义。
HIFU (High Intensity Focused Ultrasound) therapy system, is a non-invasive and minimally-invasive device recently developed for clinical treatment of tumors. Ultrasound beams from an extracorporeal transducer penetrates the overlying tussue, and then focuses on and ablates the target tumor within the body. Different from other modern hi-tech therapy system, it’s mechanism of action is to focus high-energy ultrasound in the target area so as to induce instant temperature rise which then makes the tumor tissue coagulatively necrozed
Compared with other analogous tumor therapeutic tools, radiotherapy technology is the most widely used extracorporeal irradiation technology at present time. Therapeutic effect of radiotherapy is achieved by inducing gradual necrosis of the irradiated tissues, but so gradual that its curative effect only appears several weeks later. Targeting deviation usually does not cause such safety problems as patient death. Therefore, in clinical practice, the irradiated area often exceeds the tumor area for enhancing therapy efficacy. It’s determined that a deviation within 3cm is allowed in radiotherapy targeting. However, if this deviation was allowed in HIFU targeting, vital organs or structures would be damaged.
To guarantee safety of some special sites or areas, HIFU target areas are smaller than the actual tumor region. This safety concern inevitably is offsetting therapeutic efficacy of HIFU. Under such a background, to improve HIFU treatment accuracy based on targeting accuracy will mean significantly for efficacy enhancement.
The study and research into tissue identification based on magnetic resonance imaging (MRI) will shed a light upon accurate targeting and accurate treating in HIFU procedure, and it will also provide HIFU medical personnel with a set of principles as how to identify vital tissues or structures, to calculate accoustic impedance and to chooce the optimal accoustic path. In view of this, we initiate a project to study automatic identification of tissues in the target region and in acoustic path based on MRI images. I did not found similar reports after literature research.
Objective: (1) To study common imaging characteristics of human body tissues and to determine its influencing factors through MRI image analysis. (2) To establish a MRI-image-based tissue automatic recognition system. (3) To establish a MRI-and-CT-image-based target region automatic drawing system. The systems established would be helpful to enhance the degree of accuracy during HIFU 3D conformal treatment planning.
The research includes two parts as below:
Part one MRI Imaging characteristics of human body tissues
Objective: To study common imaging characteristics of human body tissues (including tumors) by analyzing MRI images of hysteromyoma.
Methods: A“MRI image analysis software”developed specially for this study under Microsoft Visual C++ 6.0 was used. T2 MRI image data of hysteromyoma were analysed using matched-pair design.
Materials: Original image data of hysteromyoma in DICOM format from MRI system (GE) in 2nd Affiliated Hospital of Chongqing Medical University.
Results: (1) Under a same imaging condition and from the same patient, hysteromyoma MRI images obtained during different periods had similar MR figs section. (2) Under a same imaging condition, MR figs section from different patients had no significant difference. (3)Under a same MRI imaging condition, MR figs section before HIFU treatment and after HIFU treatment was significantly different. (4) Some principal tissues in acoustic path during hysteromyoma HIFU treatment, such as abdominal wall muscle, turgid bladder, normal hystera tissue, or hysteromyoma, had significantly-different imaging MR figs section.
Conclusions: On a same MRI device and under a same MRI imaging condition, the same kind of tissues in different persons have some commonalities, while different tissues have differences. It’s these commonalities and differences that provides support to automatic tissue identification, which enables accurate calculation of ultrasound-path acoustic impedance and accurate auto-drawing of the target region. All of the above developments will certainly optimate the HIFU treatment system. In addition, since HIFU treatment will change tissue’s MRI charateristics, our study also shows prospects to establish a new standard for therapeutic efficacy evaluation.
Part two Establish automatic tissue identification system and automatic target drawing system based on MRI images
Objective: (1) To analyze the way of automatic tissue recognitio based on MRI images. (2) To establish an automatic target drawing system, and thus to advance the current 3D imaging technology so as to improve accuracy of HIFU treatment. The ultimate objective is to optimate the HIFU treatment system.
Methods: We applied current computer recognition techniques in the establishment of automatic tissue identification system based on such images as MRI. The identified tissues were then segmented and further processed to draw a relatively accurate target region. Since a single technology or a single algorithm method was not able to meet requirements of automatic drawing in our study, we applied multiple techniques. After many times of comparative tests, we established an point-and-plane-integrated processing software system, whose core technology lies on the segmentation threshold based on MRI image characteristics, and on image matching technique based on sector accrete algorithm. We also used principles and algorithms such as image segmentation and threshold determination.
Results & conclusions: (1) The software system we established is effective in automatic tissue identification. (2) The software system can realize reliable and quick automatic drawing of such target regions whose boundary is clear in medical images (MRI, CT and etc.). Therefore, this study is helpful to realize accurate targeting and accurate treating in HIFU therapy and radiotherapy, which has significant value in clinical practice.
Additionally, we applied several image analysis techniques during the establishment of this software system, therefore, it also helps doctors to identify some special tissues , hence improve their medical proficiency. Our study also shed a light on the programming of intelligent diagnosis system. Besides its function in HIFU treatment, the automatic drawing program in the software system can also be used in other therapeutic modalities, such as 3D radiotherapy and super gammaknife therapy, to reduce complication and to protect important adjacent tissues and structures. Therefore, the software system developed in this study may promises to play a significant role in advancing medical technology.
对一些特殊部位或区域,为了安全,HIFU治疗的治疗靶区就会小于肿瘤实际区域,结果将可能会影响治疗效果;因此,以精确定位为基础的HIFU精确治疗对提高HIFU肿瘤治疗效果有着十分重要的意义。
开展有关基于核磁共振成像(Magnetic Resonance Imaging, MRI)的计算机组织识别技术的研究,是建立HIFU精确定位、精确治疗的基础,也为HIFU治疗人员对声通道内的重要组织结构的识别、声阻抗的计算及最佳治疗通道的选择提供依据。
出于这方面需要,开展了“基于MRI影像的HIFU治疗靶区及声通道的组织自动识别”项目的研究工作,根据相关研究的结果撰写了这份报告。作者本人曾做了大量文献检索工作,迄今为止,尚未发现类似的研究成果或报告。
本项目研究的目的主要包括三个方面:(1)通过分析MRI图像,了解人体组织结构在MR值上的共同特性及其影响因素;(2)在人体组织结构特性研究的基础上,建立基于MRI图像的计算机组织识别模式;(3)基于MRI图像的计算机组织识别模式,结合相关的图像处理技术建立基于MRI、CT图像的HIFU治疗靶区自动勾画系统,提高HIFU治疗效果与安全性。其最终目标是为了建立适形的、精确的能量投放HIFU治疗计划系统提供依据与技术支撑。为了实现这一目标,我们把本项目的具体工作分成了以下二个部分。
第一部分人体组织的MRI影像特性研究
以子宫肌瘤为研究对象,开展人体组织结构(包括肿瘤)在MRI影像上的共同特性的分析研究。
研究方法:采用在Microsoft Visual C++ 6.0环境下为本研究专门开发的“专用MRI图像分析软件”,用配对设计的统计分析方法对所研究的子宫肌瘤的MRI T2原始序列图像的数据进行分析,研究它们在MR值上的变化规律。
研究材料:重庆医科大学附属二院GE MRI设备上检查获取的DICOM格式的子宫肌瘤原始图像数据。
结果显示:(1)在同一检查条件下,同一患者不同时期的子宫肌瘤影像中,MR值分布范围没有显著差异;(2)在同一检查条件下,不同患者间的子宫肌瘤影像中,MR值分布范围没有显著差异;(3)在同一检查条件下,子宫肌瘤在HIFU治疗前、后的MR值变化范围存在显著差异;(4)在子宫肌瘤HIFU声通道中可能涉及的几个主要组织结构(腹壁肌肉、充盈膀胱、正常子宫组织及子宫肌瘤)间,在MR值分布范围上存在显著差异。
结论:由此推断,在同一MRI检查设备,相同的检查条件下的MRI影像上,人体组织结构中同一组织结构存在一定的共性,而不同的组织结构存在一定的差异。这种共性和差异性从效果看,可以作为计算机进行自动识别组织的依据,为HIFU治疗声通道的声阻抗的精确计算和靶区的精确勾画打下基础,为进一步开发与完善HIFU治疗计划系统提供了条件。另外,HIFU治疗可以改变原有组织结构的MRI图像特性,这有助于研究新的HIFU治疗效果评价指标,对建立HIFU治疗评价体系有一定意义。
第二部分MRI组织识别模式的建立与靶区自动勾画研究
研究的内容主要包括:(1)以MRI等图像为基础,对生物组织结构进行识别分析的技术实施研究;(2)为HIFU治疗建立一个比较精确的靶区自动勾画系统,从而进一步地三维成像,为HIFU治疗精确控制提供依据,也有利于建立更为完整的HIFU治疗计划系统。
研究方法:以MRI等原始图像为背景,利用相关的计算机识别技术,建立对靶区组织结构的自动识别模式,对所识别的组织结构进行进一步分割等处理,最后勾画出较为精确的靶区轮廓。研究中发现,没有任何一种技术或算法能够独立完成图像自动勾画的要求,必须借助多种技术共同完成。因此,在通过反复实验与对比研究分析后,确定了以点、面结合的组合图像处理技术方案,其中以基于MRI特性的分割阈值为特点的靶区图像分割和以区域生长算法为基础的图像模板匹配技术为本研究中图像组织结构识别技术方案的核心。此外,研究中还应用了包括图像分割阈值获取、多种分割方法的组合采用、组织结构提取等技术方法,并结合研究要求探讨了相关原理及具体算法。
结果与结论:通过试验发现(1)此技术方案能够比较准确识别诸如MRI类图像中的指定组织结构;(2)针对医学图像(MRI、CT等),能够比较快速、准确的对治疗靶区中边界相对不模糊的图像可达到一次性准确勾画的效果。因此,此项研究结果将会有助于HIFU、放射等医学治疗中的精确定位与治疗,有较大的实际应用价值。此外,此方案由于在对医学图像(MRI、CT等)进行处理的同时,结合了相关的MRI、CT显像组织结构特性分析技术,可帮助医务人员识别一些特定组织结构,提高相应的诊治水平,对建立基于MRI的智能分析诊断类的医疗专家系统的开发与研究方面的工作有所帮助。其勾画方案,除可应用于HIFU外,也可辅助用于其它一些医学治疗手段中,如三维立体放射治疗、超级γ刀的放射治疗计划系统中,可帮助提高精确定位、减少并发症发生率及重要组织结构的保护作用,并可能有助于医疗技术的进一步发展与提高有着十分重要的意义。
HIFU (High Intensity Focused Ultrasound) therapy system, is a non-invasive and minimally-invasive device recently developed for clinical treatment of tumors. Ultrasound beams from an extracorporeal transducer penetrates the overlying tussue, and then focuses on and ablates the target tumor within the body. Different from other modern hi-tech therapy system, it’s mechanism of action is to focus high-energy ultrasound in the target area so as to induce instant temperature rise which then makes the tumor tissue coagulatively necrozed
Compared with other analogous tumor therapeutic tools, radiotherapy technology is the most widely used extracorporeal irradiation technology at present time. Therapeutic effect of radiotherapy is achieved by inducing gradual necrosis of the irradiated tissues, but so gradual that its curative effect only appears several weeks later. Targeting deviation usually does not cause such safety problems as patient death. Therefore, in clinical practice, the irradiated area often exceeds the tumor area for enhancing therapy efficacy. It’s determined that a deviation within 3cm is allowed in radiotherapy targeting. However, if this deviation was allowed in HIFU targeting, vital organs or structures would be damaged.
To guarantee safety of some special sites or areas, HIFU target areas are smaller than the actual tumor region. This safety concern inevitably is offsetting therapeutic efficacy of HIFU. Under such a background, to improve HIFU treatment accuracy based on targeting accuracy will mean significantly for efficacy enhancement.
The study and research into tissue identification based on magnetic resonance imaging (MRI) will shed a light upon accurate targeting and accurate treating in HIFU procedure, and it will also provide HIFU medical personnel with a set of principles as how to identify vital tissues or structures, to calculate accoustic impedance and to chooce the optimal accoustic path. In view of this, we initiate a project to study automatic identification of tissues in the target region and in acoustic path based on MRI images. I did not found similar reports after literature research.
Objective: (1) To study common imaging characteristics of human body tissues and to determine its influencing factors through MRI image analysis. (2) To establish a MRI-image-based tissue automatic recognition system. (3) To establish a MRI-and-CT-image-based target region automatic drawing system. The systems established would be helpful to enhance the degree of accuracy during HIFU 3D conformal treatment planning.
The research includes two parts as below:
Part one MRI Imaging characteristics of human body tissues
Objective: To study common imaging characteristics of human body tissues (including tumors) by analyzing MRI images of hysteromyoma.
Methods: A“MRI image analysis software”developed specially for this study under Microsoft Visual C++ 6.0 was used. T2 MRI image data of hysteromyoma were analysed using matched-pair design.
Materials: Original image data of hysteromyoma in DICOM format from MRI system (GE) in 2nd Affiliated Hospital of Chongqing Medical University.
Results: (1) Under a same imaging condition and from the same patient, hysteromyoma MRI images obtained during different periods had similar MR figs section. (2) Under a same imaging condition, MR figs section from different patients had no significant difference. (3)Under a same MRI imaging condition, MR figs section before HIFU treatment and after HIFU treatment was significantly different. (4) Some principal tissues in acoustic path during hysteromyoma HIFU treatment, such as abdominal wall muscle, turgid bladder, normal hystera tissue, or hysteromyoma, had significantly-different imaging MR figs section.
Conclusions: On a same MRI device and under a same MRI imaging condition, the same kind of tissues in different persons have some commonalities, while different tissues have differences. It’s these commonalities and differences that provides support to automatic tissue identification, which enables accurate calculation of ultrasound-path acoustic impedance and accurate auto-drawing of the target region. All of the above developments will certainly optimate the HIFU treatment system. In addition, since HIFU treatment will change tissue’s MRI charateristics, our study also shows prospects to establish a new standard for therapeutic efficacy evaluation.
Part two Establish automatic tissue identification system and automatic target drawing system based on MRI images
Objective: (1) To analyze the way of automatic tissue recognitio based on MRI images. (2) To establish an automatic target drawing system, and thus to advance the current 3D imaging technology so as to improve accuracy of HIFU treatment. The ultimate objective is to optimate the HIFU treatment system.
Methods: We applied current computer recognition techniques in the establishment of automatic tissue identification system based on such images as MRI. The identified tissues were then segmented and further processed to draw a relatively accurate target region. Since a single technology or a single algorithm method was not able to meet requirements of automatic drawing in our study, we applied multiple techniques. After many times of comparative tests, we established an point-and-plane-integrated processing software system, whose core technology lies on the segmentation threshold based on MRI image characteristics, and on image matching technique based on sector accrete algorithm. We also used principles and algorithms such as image segmentation and threshold determination.
Results & conclusions: (1) The software system we established is effective in automatic tissue identification. (2) The software system can realize reliable and quick automatic drawing of such target regions whose boundary is clear in medical images (MRI, CT and etc.). Therefore, this study is helpful to realize accurate targeting and accurate treating in HIFU therapy and radiotherapy, which has significant value in clinical practice.
Additionally, we applied several image analysis techniques during the establishment of this software system, therefore, it also helps doctors to identify some special tissues , hence improve their medical proficiency. Our study also shed a light on the programming of intelligent diagnosis system. Besides its function in HIFU treatment, the automatic drawing program in the software system can also be used in other therapeutic modalities, such as 3D radiotherapy and super gammaknife therapy, to reduce complication and to protect important adjacent tissues and structures. Therefore, the software system developed in this study may promises to play a significant role in advancing medical technology.
引文
[1] A M Darwish,A K Jian.A rule based approach for visual pattern inspection[J].IEEE Transactions on PAMI.1998,10(1):56-68
[2] Antonio M Lopez,Felipe Lumbreras,John Serrat,et al.Evaluation of methods for ridge and valley detection[J].IEEE Trans.On Pattern Analysis and Machine Intelligence.1999,21(4):327-335
[3] B Wang,T O Binford,Generic.model-based estimation and detection of peaks in image surfaces[M]. Image Understanding Workshop.1996,913–922
[4] Bhanu B,Jones G Ⅲ,Ahn S.Recognizing articulated objects and objects articulation in SAR images[C].SPIE,Vol.3370,1998,493-507
[5] Cseke I.A fast segmentation scheme for white blood cell images[C].11th IAPR International Conf,C:Image,Speech and Signal Analysis.Vol.3,The Hage,The Netherlands,1992,530-533
[6] Dangman J.High confidence visual recognition of person by a test of statistical independence[J].IEEE Trans.on Pattern Anal.Machine Intelligence.1993,15(11): 1148-1161
[7] Eckstein F.Quantitative MRI of cartilage and bone:degenerative changes in osteoarthritis.NMR In Biomedicine [NMR Biomed].2006,Nov;Vol.19(7),822-54
[8] Gorman L. An Overview of Fingerprint Verification Technologies[J]. Information Security Technical Report.1998, 3(1):21-32
[9] Grossman Rachel,Hoffman Chen,Mardor Yae,et al.Quantitative MRI measurements of human fetal brain development in utero[J].NeuroImage.Nov,2006,Vol 33(2), 463-470
[10] Guleng Amu, Surong Hasi, Xingyu Yang, et al. Image Analysis by Pseudo-Jacobi-Fourier Moments[J].Applied Optics. 2004,43(10):2093-2101
[11] Hynynen K,Ghung AH.Thermal effects of focused ultrasound on the brain: determination with MR imaging[J].Radiology.1997,204,247-253
[12] Hynynen K,Unger E.MRI-guided noninvasive ultrasound surgery[J].Medical physics.1993, 20,107-115
[13] Jain A K.Fundamental of digital image processing[M].Prentice-Hall.Baltimore and London, 2nd Edition, 1989
[14] Javad Haddadnia. An Efficient Human Face Recognition System Using Pseudo Zernike Moment Invariant And Radial Basis Function Neural Network[J]. International Journal of Pattern Recognition and Arti_cial Intelligence. 2003, 17(1):41-62
[15] Jolesz FA.Magnetic resonance image-guided focused ultrasound surgery[J].Cancer journal. 2002,8,S100-112
[16] Jones III G,Bhanu B.Recognition of articulated and occluded objects[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.1999, 21: 603-613
[17] Kass M, Witbin A, Tetzopoulos D.Snake:Active contour model[J]. Int J. Comput Vision. 1987, 1 (4): 321-331
[18] Koo J H,Yoo S I.A structural matching for two-dimensional visual pattern inspection [C].Systems,Man,and Cybernetics,IEEE International Conference on.1998,Vol5:4429-4434
[19] Kuo C F,Shih C Y,Lee J Y.Automatic recognition of fabric weave patterns by a fuzzy C-means clustering Method[J].Textile Research Journal.2004,74(2):107-111
[20] Lei Tianhu,Wehrli Felix W.Magnetic resonance (MR) image inalysis-A statistical approach[J].International Journal of Image & Graphics.Jan. 2007, Vol.7(1),119-141
[21] Leung Kelvin K,Holden Mark,Saeed,Nadeem,et al.Automatic Quantification of Changes in Bone in Serial MR Images of Joints[J].IEEE Transactions on Medical Imaging.Dec,2006,Vol.25(12),1617-1626
[22] Meth R,Target/shadow segmentation and aspect estimation in synthetic aperture radar imagery[C].SPIE 1998,3370:188-196
[23] Mujdat cetin,William Clem Karl,David A.Castanon.Evaluation of a regularized SAR imaging technique based on recognition-oriented features[C].SPIE.April 2000,4053
[24] National Electrical Manufactures Association.Digital Imaging and Communications in Medicine(DICOM)[S].1999
[25] Negin M I,Chmielewski T A,Salganicoff Jr M,et al.An iris biometric system for public and personal use[J].Computer. 2000,3:70-75
[26] Prastawa Marcel,Gilmore John H,Lin Weili,et al.Automatic segmentation of MR images of the developing newborn brain[J].Medical Image Analysis. Oct,2005, Vol.9(5),457-466
[27] Q Z Ye,P E Danielsson.Inspection of Printed Circuit Boards by ConnectivityPreserving Shrinking[J].IEEE Transactions on PAMI. 1988,Vol1O(5):737-742
[28] Rouviere O,Raudrant A.MRI appearance of prostate following transrectal HIFU ablation of localized cancer[J].Eur Urology. 2001,40,03,265-274
[29] Ruan S,Jaggi C,Bloyet D,et al.Brain Tissue Classification in MR Image Based on a 3D MRF Model[J]. Proceeding of the 20th Annual International Conference of the IEEE Engineering in Medical and Biology Society. Vol 20, No 2,1998
[30] Thuroff S. High-intensity focused ultrasound: complications and adverse events[J]. Molecular Urology. 2000,43,183-187
[31] Victor B,Bowyer K,Sarkar S.An evaluation of face and ear biometrics[A]. In: Proceedings of International,Conference on Pattern Recognition[C].Quebec City, Canada:Image Procssing and Pattern Recognition Press. 2002.429-432
[32] Visioli AG,ter Haar GR.High-intensive focused ultrasound combine chemotherapy in the treatment of localised tumours[J].European Journal of Ultrasound. 1999,9,1, 11-18
[33] Wermser D,Haussmann G,Liedtkc C E.Segmentation of blood smears by hierarchical thresholding[J].Computer Vision,Graphics,and Image Processing. 1984, l25,589-598
[34] Wu F,Chen W.Pathological study of extracor-poreally ablated hepatic cellular carcinoma with high-intensity focused ultrasound[J].Chinese Journal of Oncology. 2001,23,03,237-239
[35] Xianghua Xie,Majid Mirmehdi.RAGS:region-aided geometric snake[J].IEEE Trans Image Process. 2004,13(5):640-652
[36] Z L Ping, R G. Wu, Y L Sheng. Describing image with Chebyshev-Fourier Moments[J]. J.Opt.Soc.Am. 2002,19(9):1748-1754
[37] 毛淑华 雷伯录. DICOM 标准及应用[J].科技广场. 2006,11,60-62
[38] 王大志,黄劼,徐树英.采用灰度直方图的孔型识别和二值化阈值自动匹配[J]. 兵工自动化. 2005,(02)
[39] 王建平,盛军等.基于小波分析的二值化阈值选取[J].合肥工业大学学报. 2001,24(4):559-562
[40] 王忠礼,穆志纯.边缘检测在外耳识别技术中的应用[J].电脑开发与应用. 2002,15(12), 5,6,9
[41] 王隽.基于 MRI 的脑图像分割和三维重建[D].河北工业大学. 2004
[42] 王渭滨,田军委,黄永宣等.整形外科面部轮廓正面图像的边缘检测和模型建立[J].南方医科大学学报. 2006,V26,N5,667-669
[43] 王煦法,庄镇泉,王东生.C 语言图像处理程序设计[M].合肥:中国科学技术大学出版社. 1994
[44] 王颖杰,李晔雄,夏廷毅.“前列腺癌靶区勾画”专题研讨会简介. 中华放射肿瘤学杂志[J]. 2006,15(4):328
[45] 王鹏伟,吴秀清,张名成.基于多尺度形态学融合的分水岭图像分割方法[J].数据采集与处理. 2006,21(4),398-402
[46] 冯若.高强聚焦超声无创性外科的兴起[J].临床超声医学杂志. 1999,2:65-67
[47] 包桂秋,林喜荣,苏晓生等.基于人体生物特征的身份识别技术发展概况[J].清华大学学(自然科学版). 2001,41(4/5):72-76
[48] 卢岷,邹建中.高强度聚焦超声治疗肿瘤的术中影像学监控和疗效评价[J].中华超声影像学杂志. 2001,7:438-440
[49] 叶德荣,李坤成,李永忠等.基于图像分割的 MRI、CT 脑图像自动配准的探讨[J].医疗设备信息. 2000,15(3),10-12
[50] 田捷,包尚联,周明全.医学影像处理与分析[M].北京:电子工业部出版社. 2003
[51] 田捷,赵明昌,何晖光.集成化医学影像算法平台理论与实践[M].北京:清华大学出版社. 2004.183-209
[52] 伍尤富.图像处理中边缘检测研究方法[J].舰船电子工程. 2006,(04)
[53] 伍亚军,叶安培,王霄英等.基于 MRI 的动物肝脏脉管分割与三维重建[J].中国医学影像技术. 2005,V21,N8,1275-1277
[54] 伍烽,王智彪,陈文直等.高强度聚焦超声体外治疗肝肿瘤的剂量学研究[J].中华物理医学与康复杂志. 2000,22:267-269
[55] 刘怡光,游志胜,曹丽萍.一种快速图像识别方法[J]. 计算机工程. 2005,(14)
[56] 刘波.基于图像分割的非线性医学图像匹配[D].华中科技大学. 2004
[57] 刘勍,张久文,温志贤等.基于数学形态学的图像处理[J].天水师范学院学报. 2004,(02)
[58] 刘素华,韩萍.基于遗传算法的模糊模式识别及其应用[J].计算机工程与设计. 2005,(04)
[59] 吕晓东.医学数字影像通讯(DICOM)标准及其应用[J]. 医疗卫生装备.2001, (1):43-44
[60] 孙中原,张继武,常桂然.一种基于形态学方法优化的种子填充算法及其在脑组织 CT 影像自动分割中的应用[J].中国临床医学影像杂志. 2002,V13,N6,384-387
[61] 孙世良,温海燕,张连阳.现代大肠癌诊断与治疗[M].重庆:重庆出版社. 2004.467-469
[62] 孙冬梅,裘正定.生物特征识别技术综述[J].电子学报. 2001,29(12A),1744-1748
[63] 孙佳,彭玉华.基于模板匹配的掌纹识别自适应阈值选取方法[J].山东大学学报(工学版).2005,(05)
[64] 劳丽,吴效明,朱学峰.模糊集理论在图像分割中的应用综述[J].中国体视学与图像分析.2006,V11,N3,200-205
[65] 吴小艳,王维庆,杨春祥等.几种基于模板匹配法的数字图像识别算法分析[J].计量技术,2005,(06)
[66] 宋志勇,文玉梅,徐颖等.HIFU 治疗系统中 B 超图像的雾状伪像分析[J].数理医学杂志. 2007,2,20:(1)
[67] 张大波,郭利,巴巍. 基于第二代小波变换的心电信号分析算法[J].数据采集与处理. 2006,21(4),450-453
[68] 张立恒,陈炜.基于小波变换和贝叶斯理论的图像分割算法[J].电子测量技术. 2006,29(5),74-77
[69] 张洪峰,钱晓平.HIFU 肿瘤热疗机的计算机辅助治疗系统[J].中国医疗器械杂志. 2002,V26,N2,96-99
[70] 张益贞.Visual C++实现 MPEG/JPEG 编解码技术[M].北京:人民邮电出版社. 2002,10-66
[71] 张敏贵,周德龙,潘泉等.生物特征识别及研究现状[J].生物物理学报,2002,18 (2):156-162
[72] 张雯,郝重阳.基于区域增长的可视人彩色图像分割[J]. 电子技术应用 . 2005,31:3,43-44
[73] 张瑞兰,华晶,安巍力等.数字图像处理在医学影像方面的应用[J].医学信息.2006,V19,N3,400-401
[74] 李军宏,潘泉,程咏梅等.一种基于高阶统计量的图像识别算法研究[J]. 系统工程与电子技术. 2005,(06)
[75] 李忠杰,杨关良,徐小杰.自动阈值选取与边缘检测相结合的图像分割算法[J].海军工程大学学报. 2002,(04)
[76] 李前文.图像整合技术与肿瘤靶区的勾画[J].临床肿瘤学杂志.2005.03.15, 98-100
[77] 李晓红.卢光明颈部磁共振血管造影的伪影及消除方法[J].中国医学影像技术.1999,15(增刊),32-33
[78] 李彬,陈武凡.基于模糊聚类空间模型的非均匀 MR 图像分割[J].医疗卫生装备. 2006,27(2),3-4
[79] 李璟,朱善安.医学图像分割技术[J].生物医学工程学杂志.2006,V23,N4,891-894
[80] 杨竹,王智彪.高强度聚焦超声治疗子宫肌瘤的病理学研究[J].中华超声影像学杂志. 2003,12, 674-676
[81] 杨宏伟,张云.基于图像处理的血液细胞特征提取 [J]. 生物信息学 .2006, V4N2,76-78
[82] 杨晓强,李斌,魏生民.基于解剖知识模型的医学图像分割方法研究[J].航天医学与医学工程. 2005,V18,N1,62-65
[83] 汪伟,周洁敏. 高强度聚焦超声治疗症状性子宫肌瘤的初步临床研究[J]. 中华超声影像学杂志. 2002,11,161-163
[84] 邱明辉,张纪国.基于 DICOM 标准的医学图像显示与处理系统的研究与实现[J].医疗设备信息. 2004V19N11,18-20
[85] 陈立新,李文杰,黄晓延等.治疗计划系统中 CT 值及图像质量保证的若干问题讨论[J].中国医学物理学杂志. 2003,10,14. 20(4):193-196
[86] 陈宇杰,陆叙林,王辉.用 Visual C++实现对彩色医学图像的分割[J].医疗设备信息. 2001,02,15;16(2):4-6
[87] 陈虎,王守尊,周朝辉.基于数学形态学的图像边缘检测方法研究[J].工程图学学报. 2004,(02)
[88] 陈受田.腹部肿瘤超声诊断技术[M].重庆:重庆出版社. 2004,316-322
[89] 郑亚琴,田心.医学图像配准技术研究进展[J].国际生物医学工程杂志. 2006, V29N2,88-92
[90] 郑成勇,李红,高景丽.基于最优分类数及遗传算法的灰度图像多阈值自动分割法[J].计算机工程与应用. 2003,(23)
[91] 侯振杰,麻硕士,裴喜春.一种基于遗传算法的骨髓细胞图像分割方法[J].计算机工程与科学. 2006,28(10),63-65
[92] 姜慧研,司岳鹏,雒兴刚.基于改进的大津方法与区域生长的医学图像分割[J].东北大学学报(自然科学版). 2006,(04)
[93] 胡阳秋,高小榕,高上凯. 医学图像DICOM格式转换软件的设计与实现[J]. 北京生物医学工程.2000,19(4),193-197
[94] 胡松柳,徐向英.非小细胞肺癌三维适形放射治疗的靶区勾画相关研究[J].黑龙江医学. 2006,30(5):344-345
[95] 胡瑢华,刘国平,张华等.基于灰度形态学的图像处理在焊缝识别中的应用[J].计算机工程与应用. 2005,(27)
[96] 赵奎,蔡美峰.模糊 C 均值聚类算法在结构面组织识别中的应用[J].金属矿山. 2002,1,13-15
[97] 项晖,庄天戈.基于 CT/MRI 图像的三维近距离放射治疗计划系统[J].中国医疗器械杂志. 2002V26N6,398-401
[98] 饶海涛,翁桂荣.基于数学形态学的图像边缘检测[J].苏州大学学报(自然科学版). 2004,(02)
[99] 倪宗瓒.医用统计学(第二版)[M].北京:人民卫生出版社,1998
[100] 翁璇,郑小林,彭承林.医学图像分割方法研究[J].中国医疗器械信息. 2006, V12N6,24-26
[101] 耿茵茵,蔡安妮,孙景鳌.自动图像阈值分割算法[J].计算机工程与应用. 2002,(17)
[102] 诸葛斌,周荷琴,唐雷.中国数字人数据集断层图像自动配准算法[J].中国生物医学工程学报. 2005,V24N4,443-447
[103] 贾云得.机器视觉[M].北京:科学出版社. 2000.34–63
[104] 郭宁宁,曾立波,吴琼水等.基于多光谱采集技术的骨髓涂片细胞分割方法[J].计算机工程与应用. 2005,31:171-174
[105] 郭娟,王广阔.灰度图像的最佳阈值分割算法[J].济宁师范专科学校学报. 2005, (03)
[106] 高卫东,刘基宏等.织物中纬纱排列参数的自动识别[J].棉纺织技术. 2002,30(1):28-31
[107] 高元桂,蔡幼铨,蔡祖龙.磁共振成像诊断学[M].北京:人民军医出版社. 1993
[108] 常淑芳,白晋.高强度聚焦超声定位损伤离体人子宫肌瘤的研究[J].中国超声医学杂志. 2001,17,97-100
[109] 常淑芳,顾美礼,伍烽等.高强度聚焦超声治疗子宫肌瘤的剂量学研究[J].中华物理医学与康复杂志. 2003,9,454
[110] 康维,欧阳成,王广志等.基于分水岭变换的彩色细胞图像分割[J].清华大学学报·自然科学版. 2006,V46N3,414-417
[111] 黄海欣等.CT/MR 图像融合在鼻咽癌三维适形放疗中靶区勾画的应用[J].医学文选. 2006,25(3):549-551
[112] 黄继英,梁星原.磁共振成像原理[M].西安:陕西科学技术出版社. 1998. 95-107
[113] 黄朝兵,蒋莹莹. 一种基于小波变换的低对比度图像增强方法[J].现代计算机(下半月版). 2006,12,82-83
[114] 温海燕, 张雁.C 语言实用技术[M]. 成都:成都科技大学出版社. 1996
[115] 程义民,王以孝,沈勇等.B 型超声心动图中心脏组织及结构的识别[J].中国生物医学工程学报. 2001,(06)
[116] 程咏梅,潘泉,张洪才等.计算机智能图像识别算法研究[J].计算机应用. 2004,(02)
[117] 程勇,陈卫国.医学图像后处理技术及其研究进展[J].医疗设备信息.2005, V20N9,34-37
[118] 韩冬林.加权平均算法在 DSA 图像处理中的应用[J].医疗卫生装备.2003, V24N7,41
[119] 翟亮,张洪峰.数字图像处理技术在超声刀治疗肿瘤中的应用[J].中国医学影像技术. 2002,V18N,S1 94
[120] 谭珂,郭光友,潘新华等.基于数字化可视人数据集的医学图像配准研究[J].军医进修学院学报.2006,V27N1,60-61
[121] 瞿洋,杨利平.Houhg 变换 OCR 图象倾斜矫正方法[J].中国图象图形学报 A辑. 2001,6(2):178-181
[2] Antonio M Lopez,Felipe Lumbreras,John Serrat,et al.Evaluation of methods for ridge and valley detection[J].IEEE Trans.On Pattern Analysis and Machine Intelligence.1999,21(4):327-335
[3] B Wang,T O Binford,Generic.model-based estimation and detection of peaks in image surfaces[M]. Image Understanding Workshop.1996,913–922
[4] Bhanu B,Jones G Ⅲ,Ahn S.Recognizing articulated objects and objects articulation in SAR images[C].SPIE,Vol.3370,1998,493-507
[5] Cseke I.A fast segmentation scheme for white blood cell images[C].11th IAPR International Conf,C:Image,Speech and Signal Analysis.Vol.3,The Hage,The Netherlands,1992,530-533
[6] Dangman J.High confidence visual recognition of person by a test of statistical independence[J].IEEE Trans.on Pattern Anal.Machine Intelligence.1993,15(11): 1148-1161
[7] Eckstein F.Quantitative MRI of cartilage and bone:degenerative changes in osteoarthritis.NMR In Biomedicine [NMR Biomed].2006,Nov;Vol.19(7),822-54
[8] Gorman L. An Overview of Fingerprint Verification Technologies[J]. Information Security Technical Report.1998, 3(1):21-32
[9] Grossman Rachel,Hoffman Chen,Mardor Yae,et al.Quantitative MRI measurements of human fetal brain development in utero[J].NeuroImage.Nov,2006,Vol 33(2), 463-470
[10] Guleng Amu, Surong Hasi, Xingyu Yang, et al. Image Analysis by Pseudo-Jacobi-Fourier Moments[J].Applied Optics. 2004,43(10):2093-2101
[11] Hynynen K,Ghung AH.Thermal effects of focused ultrasound on the brain: determination with MR imaging[J].Radiology.1997,204,247-253
[12] Hynynen K,Unger E.MRI-guided noninvasive ultrasound surgery[J].Medical physics.1993, 20,107-115
[13] Jain A K.Fundamental of digital image processing[M].Prentice-Hall.Baltimore and London, 2nd Edition, 1989
[14] Javad Haddadnia. An Efficient Human Face Recognition System Using Pseudo Zernike Moment Invariant And Radial Basis Function Neural Network[J]. International Journal of Pattern Recognition and Arti_cial Intelligence. 2003, 17(1):41-62
[15] Jolesz FA.Magnetic resonance image-guided focused ultrasound surgery[J].Cancer journal. 2002,8,S100-112
[16] Jones III G,Bhanu B.Recognition of articulated and occluded objects[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.1999, 21: 603-613
[17] Kass M, Witbin A, Tetzopoulos D.Snake:Active contour model[J]. Int J. Comput Vision. 1987, 1 (4): 321-331
[18] Koo J H,Yoo S I.A structural matching for two-dimensional visual pattern inspection [C].Systems,Man,and Cybernetics,IEEE International Conference on.1998,Vol5:4429-4434
[19] Kuo C F,Shih C Y,Lee J Y.Automatic recognition of fabric weave patterns by a fuzzy C-means clustering Method[J].Textile Research Journal.2004,74(2):107-111
[20] Lei Tianhu,Wehrli Felix W.Magnetic resonance (MR) image inalysis-A statistical approach[J].International Journal of Image & Graphics.Jan. 2007, Vol.7(1),119-141
[21] Leung Kelvin K,Holden Mark,Saeed,Nadeem,et al.Automatic Quantification of Changes in Bone in Serial MR Images of Joints[J].IEEE Transactions on Medical Imaging.Dec,2006,Vol.25(12),1617-1626
[22] Meth R,Target/shadow segmentation and aspect estimation in synthetic aperture radar imagery[C].SPIE 1998,3370:188-196
[23] Mujdat cetin,William Clem Karl,David A.Castanon.Evaluation of a regularized SAR imaging technique based on recognition-oriented features[C].SPIE.April 2000,4053
[24] National Electrical Manufactures Association.Digital Imaging and Communications in Medicine(DICOM)[S].1999
[25] Negin M I,Chmielewski T A,Salganicoff Jr M,et al.An iris biometric system for public and personal use[J].Computer. 2000,3:70-75
[26] Prastawa Marcel,Gilmore John H,Lin Weili,et al.Automatic segmentation of MR images of the developing newborn brain[J].Medical Image Analysis. Oct,2005, Vol.9(5),457-466
[27] Q Z Ye,P E Danielsson.Inspection of Printed Circuit Boards by ConnectivityPreserving Shrinking[J].IEEE Transactions on PAMI. 1988,Vol1O(5):737-742
[28] Rouviere O,Raudrant A.MRI appearance of prostate following transrectal HIFU ablation of localized cancer[J].Eur Urology. 2001,40,03,265-274
[29] Ruan S,Jaggi C,Bloyet D,et al.Brain Tissue Classification in MR Image Based on a 3D MRF Model[J]. Proceeding of the 20th Annual International Conference of the IEEE Engineering in Medical and Biology Society. Vol 20, No 2,1998
[30] Thuroff S. High-intensity focused ultrasound: complications and adverse events[J]. Molecular Urology. 2000,43,183-187
[31] Victor B,Bowyer K,Sarkar S.An evaluation of face and ear biometrics[A]. In: Proceedings of International,Conference on Pattern Recognition[C].Quebec City, Canada:Image Procssing and Pattern Recognition Press. 2002.429-432
[32] Visioli AG,ter Haar GR.High-intensive focused ultrasound combine chemotherapy in the treatment of localised tumours[J].European Journal of Ultrasound. 1999,9,1, 11-18
[33] Wermser D,Haussmann G,Liedtkc C E.Segmentation of blood smears by hierarchical thresholding[J].Computer Vision,Graphics,and Image Processing. 1984, l25,589-598
[34] Wu F,Chen W.Pathological study of extracor-poreally ablated hepatic cellular carcinoma with high-intensity focused ultrasound[J].Chinese Journal of Oncology. 2001,23,03,237-239
[35] Xianghua Xie,Majid Mirmehdi.RAGS:region-aided geometric snake[J].IEEE Trans Image Process. 2004,13(5):640-652
[36] Z L Ping, R G. Wu, Y L Sheng. Describing image with Chebyshev-Fourier Moments[J]. J.Opt.Soc.Am. 2002,19(9):1748-1754
[37] 毛淑华 雷伯录. DICOM 标准及应用[J].科技广场. 2006,11,60-62
[38] 王大志,黄劼,徐树英.采用灰度直方图的孔型识别和二值化阈值自动匹配[J]. 兵工自动化. 2005,(02)
[39] 王建平,盛军等.基于小波分析的二值化阈值选取[J].合肥工业大学学报. 2001,24(4):559-562
[40] 王忠礼,穆志纯.边缘检测在外耳识别技术中的应用[J].电脑开发与应用. 2002,15(12), 5,6,9
[41] 王隽.基于 MRI 的脑图像分割和三维重建[D].河北工业大学. 2004
[42] 王渭滨,田军委,黄永宣等.整形外科面部轮廓正面图像的边缘检测和模型建立[J].南方医科大学学报. 2006,V26,N5,667-669
[43] 王煦法,庄镇泉,王东生.C 语言图像处理程序设计[M].合肥:中国科学技术大学出版社. 1994
[44] 王颖杰,李晔雄,夏廷毅.“前列腺癌靶区勾画”专题研讨会简介. 中华放射肿瘤学杂志[J]. 2006,15(4):328
[45] 王鹏伟,吴秀清,张名成.基于多尺度形态学融合的分水岭图像分割方法[J].数据采集与处理. 2006,21(4),398-402
[46] 冯若.高强聚焦超声无创性外科的兴起[J].临床超声医学杂志. 1999,2:65-67
[47] 包桂秋,林喜荣,苏晓生等.基于人体生物特征的身份识别技术发展概况[J].清华大学学(自然科学版). 2001,41(4/5):72-76
[48] 卢岷,邹建中.高强度聚焦超声治疗肿瘤的术中影像学监控和疗效评价[J].中华超声影像学杂志. 2001,7:438-440
[49] 叶德荣,李坤成,李永忠等.基于图像分割的 MRI、CT 脑图像自动配准的探讨[J].医疗设备信息. 2000,15(3),10-12
[50] 田捷,包尚联,周明全.医学影像处理与分析[M].北京:电子工业部出版社. 2003
[51] 田捷,赵明昌,何晖光.集成化医学影像算法平台理论与实践[M].北京:清华大学出版社. 2004.183-209
[52] 伍尤富.图像处理中边缘检测研究方法[J].舰船电子工程. 2006,(04)
[53] 伍亚军,叶安培,王霄英等.基于 MRI 的动物肝脏脉管分割与三维重建[J].中国医学影像技术. 2005,V21,N8,1275-1277
[54] 伍烽,王智彪,陈文直等.高强度聚焦超声体外治疗肝肿瘤的剂量学研究[J].中华物理医学与康复杂志. 2000,22:267-269
[55] 刘怡光,游志胜,曹丽萍.一种快速图像识别方法[J]. 计算机工程. 2005,(14)
[56] 刘波.基于图像分割的非线性医学图像匹配[D].华中科技大学. 2004
[57] 刘勍,张久文,温志贤等.基于数学形态学的图像处理[J].天水师范学院学报. 2004,(02)
[58] 刘素华,韩萍.基于遗传算法的模糊模式识别及其应用[J].计算机工程与设计. 2005,(04)
[59] 吕晓东.医学数字影像通讯(DICOM)标准及其应用[J]. 医疗卫生装备.2001, (1):43-44
[60] 孙中原,张继武,常桂然.一种基于形态学方法优化的种子填充算法及其在脑组织 CT 影像自动分割中的应用[J].中国临床医学影像杂志. 2002,V13,N6,384-387
[61] 孙世良,温海燕,张连阳.现代大肠癌诊断与治疗[M].重庆:重庆出版社. 2004.467-469
[62] 孙冬梅,裘正定.生物特征识别技术综述[J].电子学报. 2001,29(12A),1744-1748
[63] 孙佳,彭玉华.基于模板匹配的掌纹识别自适应阈值选取方法[J].山东大学学报(工学版).2005,(05)
[64] 劳丽,吴效明,朱学峰.模糊集理论在图像分割中的应用综述[J].中国体视学与图像分析.2006,V11,N3,200-205
[65] 吴小艳,王维庆,杨春祥等.几种基于模板匹配法的数字图像识别算法分析[J].计量技术,2005,(06)
[66] 宋志勇,文玉梅,徐颖等.HIFU 治疗系统中 B 超图像的雾状伪像分析[J].数理医学杂志. 2007,2,20:(1)
[67] 张大波,郭利,巴巍. 基于第二代小波变换的心电信号分析算法[J].数据采集与处理. 2006,21(4),450-453
[68] 张立恒,陈炜.基于小波变换和贝叶斯理论的图像分割算法[J].电子测量技术. 2006,29(5),74-77
[69] 张洪峰,钱晓平.HIFU 肿瘤热疗机的计算机辅助治疗系统[J].中国医疗器械杂志. 2002,V26,N2,96-99
[70] 张益贞.Visual C++实现 MPEG/JPEG 编解码技术[M].北京:人民邮电出版社. 2002,10-66
[71] 张敏贵,周德龙,潘泉等.生物特征识别及研究现状[J].生物物理学报,2002,18 (2):156-162
[72] 张雯,郝重阳.基于区域增长的可视人彩色图像分割[J]. 电子技术应用 . 2005,31:3,43-44
[73] 张瑞兰,华晶,安巍力等.数字图像处理在医学影像方面的应用[J].医学信息.2006,V19,N3,400-401
[74] 李军宏,潘泉,程咏梅等.一种基于高阶统计量的图像识别算法研究[J]. 系统工程与电子技术. 2005,(06)
[75] 李忠杰,杨关良,徐小杰.自动阈值选取与边缘检测相结合的图像分割算法[J].海军工程大学学报. 2002,(04)
[76] 李前文.图像整合技术与肿瘤靶区的勾画[J].临床肿瘤学杂志.2005.03.15, 98-100
[77] 李晓红.卢光明颈部磁共振血管造影的伪影及消除方法[J].中国医学影像技术.1999,15(增刊),32-33
[78] 李彬,陈武凡.基于模糊聚类空间模型的非均匀 MR 图像分割[J].医疗卫生装备. 2006,27(2),3-4
[79] 李璟,朱善安.医学图像分割技术[J].生物医学工程学杂志.2006,V23,N4,891-894
[80] 杨竹,王智彪.高强度聚焦超声治疗子宫肌瘤的病理学研究[J].中华超声影像学杂志. 2003,12, 674-676
[81] 杨宏伟,张云.基于图像处理的血液细胞特征提取 [J]. 生物信息学 .2006, V4N2,76-78
[82] 杨晓强,李斌,魏生民.基于解剖知识模型的医学图像分割方法研究[J].航天医学与医学工程. 2005,V18,N1,62-65
[83] 汪伟,周洁敏. 高强度聚焦超声治疗症状性子宫肌瘤的初步临床研究[J]. 中华超声影像学杂志. 2002,11,161-163
[84] 邱明辉,张纪国.基于 DICOM 标准的医学图像显示与处理系统的研究与实现[J].医疗设备信息. 2004V19N11,18-20
[85] 陈立新,李文杰,黄晓延等.治疗计划系统中 CT 值及图像质量保证的若干问题讨论[J].中国医学物理学杂志. 2003,10,14. 20(4):193-196
[86] 陈宇杰,陆叙林,王辉.用 Visual C++实现对彩色医学图像的分割[J].医疗设备信息. 2001,02,15;16(2):4-6
[87] 陈虎,王守尊,周朝辉.基于数学形态学的图像边缘检测方法研究[J].工程图学学报. 2004,(02)
[88] 陈受田.腹部肿瘤超声诊断技术[M].重庆:重庆出版社. 2004,316-322
[89] 郑亚琴,田心.医学图像配准技术研究进展[J].国际生物医学工程杂志. 2006, V29N2,88-92
[90] 郑成勇,李红,高景丽.基于最优分类数及遗传算法的灰度图像多阈值自动分割法[J].计算机工程与应用. 2003,(23)
[91] 侯振杰,麻硕士,裴喜春.一种基于遗传算法的骨髓细胞图像分割方法[J].计算机工程与科学. 2006,28(10),63-65
[92] 姜慧研,司岳鹏,雒兴刚.基于改进的大津方法与区域生长的医学图像分割[J].东北大学学报(自然科学版). 2006,(04)
[93] 胡阳秋,高小榕,高上凯. 医学图像DICOM格式转换软件的设计与实现[J]. 北京生物医学工程.2000,19(4),193-197
[94] 胡松柳,徐向英.非小细胞肺癌三维适形放射治疗的靶区勾画相关研究[J].黑龙江医学. 2006,30(5):344-345
[95] 胡瑢华,刘国平,张华等.基于灰度形态学的图像处理在焊缝识别中的应用[J].计算机工程与应用. 2005,(27)
[96] 赵奎,蔡美峰.模糊 C 均值聚类算法在结构面组织识别中的应用[J].金属矿山. 2002,1,13-15
[97] 项晖,庄天戈.基于 CT/MRI 图像的三维近距离放射治疗计划系统[J].中国医疗器械杂志. 2002V26N6,398-401
[98] 饶海涛,翁桂荣.基于数学形态学的图像边缘检测[J].苏州大学学报(自然科学版). 2004,(02)
[99] 倪宗瓒.医用统计学(第二版)[M].北京:人民卫生出版社,1998
[100] 翁璇,郑小林,彭承林.医学图像分割方法研究[J].中国医疗器械信息. 2006, V12N6,24-26
[101] 耿茵茵,蔡安妮,孙景鳌.自动图像阈值分割算法[J].计算机工程与应用. 2002,(17)
[102] 诸葛斌,周荷琴,唐雷.中国数字人数据集断层图像自动配准算法[J].中国生物医学工程学报. 2005,V24N4,443-447
[103] 贾云得.机器视觉[M].北京:科学出版社. 2000.34–63
[104] 郭宁宁,曾立波,吴琼水等.基于多光谱采集技术的骨髓涂片细胞分割方法[J].计算机工程与应用. 2005,31:171-174
[105] 郭娟,王广阔.灰度图像的最佳阈值分割算法[J].济宁师范专科学校学报. 2005, (03)
[106] 高卫东,刘基宏等.织物中纬纱排列参数的自动识别[J].棉纺织技术. 2002,30(1):28-31
[107] 高元桂,蔡幼铨,蔡祖龙.磁共振成像诊断学[M].北京:人民军医出版社. 1993
[108] 常淑芳,白晋.高强度聚焦超声定位损伤离体人子宫肌瘤的研究[J].中国超声医学杂志. 2001,17,97-100
[109] 常淑芳,顾美礼,伍烽等.高强度聚焦超声治疗子宫肌瘤的剂量学研究[J].中华物理医学与康复杂志. 2003,9,454
[110] 康维,欧阳成,王广志等.基于分水岭变换的彩色细胞图像分割[J].清华大学学报·自然科学版. 2006,V46N3,414-417
[111] 黄海欣等.CT/MR 图像融合在鼻咽癌三维适形放疗中靶区勾画的应用[J].医学文选. 2006,25(3):549-551
[112] 黄继英,梁星原.磁共振成像原理[M].西安:陕西科学技术出版社. 1998. 95-107
[113] 黄朝兵,蒋莹莹. 一种基于小波变换的低对比度图像增强方法[J].现代计算机(下半月版). 2006,12,82-83
[114] 温海燕, 张雁.C 语言实用技术[M]. 成都:成都科技大学出版社. 1996
[115] 程义民,王以孝,沈勇等.B 型超声心动图中心脏组织及结构的识别[J].中国生物医学工程学报. 2001,(06)
[116] 程咏梅,潘泉,张洪才等.计算机智能图像识别算法研究[J].计算机应用. 2004,(02)
[117] 程勇,陈卫国.医学图像后处理技术及其研究进展[J].医疗设备信息.2005, V20N9,34-37
[118] 韩冬林.加权平均算法在 DSA 图像处理中的应用[J].医疗卫生装备.2003, V24N7,41
[119] 翟亮,张洪峰.数字图像处理技术在超声刀治疗肿瘤中的应用[J].中国医学影像技术. 2002,V18N,S1 94
[120] 谭珂,郭光友,潘新华等.基于数字化可视人数据集的医学图像配准研究[J].军医进修学院学报.2006,V27N1,60-61
[121] 瞿洋,杨利平.Houhg 变换 OCR 图象倾斜矫正方法[J].中国图象图形学报 A辑. 2001,6(2):178-181