肝脏病变MRI诊断与计算机辅助诊断
|
摘要
原发性肝细胞癌(HCC)是恶性程度极高的肝脏原发性肿瘤,而肝硬化再生结节(RN)多进展为HCC,从肝硬化中检出小HCC,使HCC病人的成功手术治疗和治愈成为可能；另外,早期肝硬化(肝纤维化)的准确诊断,使肝纤维化病人有望通过内科治疗而痊愈。近几年,核磁共振成像(MRI)已成为肝硬化、HCC影像诊断最敏感的手段之一。但早期肝硬化组织往往形态学变化轻微,同时临床MRI诊断肝硬化缺乏客观量化指标,使常规MRI诊断早期肝硬化存在困难；虽然新型MRI对比剂超顺磁性氧化铁(SPIO)的成功应用,使RN、HCC结节的鉴别诊断水平得以进一步提高,但SPIO增强图像不能显示肝脏病灶的血流动力学特点,同时SPIO增强图像使肝脏病灶诊断的假阳性率提高,目前,RN、HCC结节的MRI鉴别诊断仍为一难题。
本文针对临床RN、HCC结节及早期肝硬化MRI诊断中存在的问题,将动物模型、临床影像诊断技术与医学图像处理和分析技术相结合,发挥多学科交叉的优势,进行了肝脏病变的MRI诊断与计算机辅助诊断研究。首先,通过建立大鼠HCC模型获取RN、HCC结节钆喷替酸葡甲胺(Gd-DTPA)与SPIO增强图像,评估了Gd-DTPA与SPIO双对比增强图像鉴别诊断RN、HCC结节的应用价值,进一步设计了一种基于纹理特征的神经网络分类器用于SPIO增强图像RN和HCC结节的分类识别；其次,应用本文分类器分类识别肝硬化及正常肝脏病人的MRI；最后,进一步采用遗传算法优化基于纹理特征的BP网络分类器。
本文所完成的主要研究工作如下：
(1)大鼠RN、HCC结节SPIO增强图像及SPIO与Gd-DTPA双对比增强图像分析,提高了RN、HCC结节的影像鉴别诊断能力。首先,通过小剂量二乙基亚硝胺诱癌法建立大鼠肝硬化、HCC模型,获得病理证实的肝脏结节106枚(RN 24, HCC 82)。将106枚肝脏结节的SPIO增强图像及SPIO与Gd-DTPA双对比增强图像与病理结果对照分析,结果显示SPIO增强图像HCC结节的对比噪声比(CNR)显著提高(P<0.05),而肝硬化组织的CNR没有显著改变(P>0.05)；SPIO与Gd-DTPA双对比增强图像的HCC结节检出敏感性(96.34%)较单一GD-DTPA增强图像的敏感性(89.02%)有一定的提高,但其统计学差异没有显著性(P>0.05)。上述研究结果表明大鼠RN、HCC结节鉴别中,SPIO增强图像能较平扫图像提高HCC结节的检出和定性诊断能力,SPIO增强图像是Gd-DTPA增强图像有益的补充。
(2)基于纹理特征的BP神经网络分类器用于大鼠及病人肝脏MRI病灶分类识别,获得了较高的分类识别率。首先,利用ISODATA算法自动分割T1WI大鼠肝脏边缘,比较与Otsu阈值选择法和迭代阈值选择法的分割效果,结果显示ISODATA算法更适合于类别未知组织的自动分割问题,为进一步寻找病灶感兴趣区域(ROI)奠定了基础。其次,手动选取大鼠RN、HCC结节SPIO增强图像和病人肝硬化组织及正常肝脏组织T1WI的ROI,并进行纹理特征提取。综合比较后选择基于灰度共生矩阵的纹理特征提取方案,结合医生经验提取二阶矩、对比度、相关、逆差矩、熵、方差六个类间差异显著的纹理特征。针对大鼠RN、HCC结节SPIO增强图像,进行上述纹理特征的统计学分析,选取两组间差异显著的六个纹理特征,并与医生临床经验对比获得一致结论,设计基于纹理特征的神经网络分类器,获得了91.67%的独立测试正确率；针对正常肝脏和肝硬化病人T1WI,统计学分析后选取两组间差异显著的四个纹理特征,设计基于纹理特征的神经网络分类器,获得了87.60%的独立测试正确识别率。两组结果表明本论文所设计的基于纹理特征地的BP网络分类算法适合于大鼠RN、HCC结节SPIO增强图像及肝硬化病人T1WI分类识别,但肝硬化病人T1WI分类识别略低。
(3)针对肝硬化病人T1WI分类识别,增加纹理特征并基于遗传算法进行BP网络优化。首先,灰度共生矩阵增加至四个方向提取纹理特征参数,每个方向确定14个纹理特征参数,共计54个特征参数。采用盒状图分析后,获得可分性较好的24个特征参数,从而实现特征选择。在输入特征较多的情况下,本文并未直接进行BP神经网络设计,而采用遗传算法辅助下的BP网络优化设计：通过遗传算法优化BP神经网络初始值的GA-BP算法,在保证分类准确率的前提下,不仅增加了BP网络的稳定性,同时使BP网络的平均收敛速度提高,获得了95.00%的独立测试分类准确率。
Primary hepatocellular carcinoma (HCC) is a malignant primary liver cancer, and HCC usually occurs as a complication of cirrhosis. Small HCC detected from cirrhotic background makes the successful surgical treatment of HCC and cure being possible. Fibrosis is an early stage of cirrhosis and fibrosis is expected to recover through medicine treatment if it is diagnosed accurately. In recent years, Magnetic resonance imaging (MRI) has been one of the most sensitive methods for diagnosis of cirrhosis and HCC. But it is difficult to diagnose of fibrosis by MRI because morphological changes of fibrosis are often mild. Successful application of novel MRI contrast agent-superparamagnetic iron oxide (SPIO) improves the ability of differential diagnosis regenerative nodule (RN) from HCC nodule, but the SPIO-enhanced imaging can not show the hemodynamic characteristics of hepatic lesions and the false positive rate of SPIO-enhanced imaging is high relatively. It is difficult to diagnose RN from HCC nodule by MRI now.
For the problems of imaging diagnosis of fibrosis and hepatic nodules correlation cirrhosis, this reseach integrates animal expetiments and clinical imaging with imaging processing and imaging analysis technology. Possessing advantages of multiple subject crossing, it carries through the investigation of MR imaging diagnosis and computer aided diagnosis of hepatic lesions. Firstly, this dissertation studies the diagnostic value of SPIO-enhanced imaging and combined SPIO-enhanced and Gadolinium DTPA (Gd-DTPA) enhanced imaging for differential diagnosis of RN and HCC nodule through establishment of rat HCC model. Secondly, a neural network (NN) classifier based on texture feature is designed to distinguish rat hepatic nodule in SPIO-enhanced imaging. Thirdly, clinical patient cirrhosis MR imaging is classified by the NN classifier based on texture feature. Lastly, the initial value of the NN classifier for clinical patient cirrhosis MR imaging is optimized by Ggenetic algorithm (GA) to improve the performance of the classifier. The main research work and contributions of this dissertation can be summarized as follows:
(1) Analysing rat RN and HCC nodule SPIO-enhanced imaging and combined SPIO and Gd-DTPA enhanced imaging and the ability of rat RN and HCC differentiated diagnosis is improved. Rat HCC model is induced by low dose diethylnitrosamine, a total of 106 nodules including 24 RN and 82 HCC are pathologically confirmed. SPIO-enhanced imaging and combined SPIO and Gd-DTPA enhanced imaging of rat 106 modules are analysized compared with pathology. The results show SPIO-enhanced imaging contrast noise ratio (CNR) of HCC nodule is higher than the pre-SPIO enhanced imaging (P< 0.05), SPIO-enhanced imaging CNR of cirrhosis tissue has not significantly changed (P>0.05). The sensitivity of the combined SPIO and Gd-DTPA imaging (96.34%) is higher than the sensitivity of Gd-DTPA imaging alone (89.02%), but the statistical difference is not significant (P>0.05). It is concluded that SPIO-enhanced imaging can improve rat HCC nodule detection and diagnosis capability compared with T2WI, and SPIO-enhanced imaging is a useful complement to the Gd-DTPA enhanced image for rat RN and HCC differentiated diagnosis.
(2) Classifying rat and patient hepatic MR imaging using BP NN classifier based on texture feature and higher classification rate is obtained. Firstly, rat hepatic edge is automatically segmented by ISODATA algorithm, Otsu threshold algorithms and Iterative threshold algorithms on T1WI. The result shows that ISODATA algorithm is suitable for unknown category imaging automatical segmentation, and it is preparative for lesion region of interest (ROI) searching next step. Secondly, ROI of rat RN and HCC nodule SPIO-enhanced imaging and patient cirrhosis and normal liver parenchyma T1WI are cut manually and feature characteristics are extracted by Gray Level Coourrence Matrix (GLCM). Feature extraction program based on texture is chosen after comprehensive comparision. Six texture characteristic parameters including Angular Second Moment, Contrast, Correlation, Inverse Difference Moment, Entropy, and Variance are extracted according to the medical experience. For rat RN and HCC nodule SPIO-enhanced imaging, six texture characteristic parameters which are significantly different between two classes are chosen after statistical analysis. The result is same as clinical doctor's subjective visuality. Then a BP NN classifier is designed and classified.The independent test classification accuracy of the BP NN classifier is 91.67%. For patient cirrhosis and normal liver parenchyma T1WI, four texture characteristic parameters which are significantly different between two classes are chosen after statistical analysis. Then a BP NN classifier is designed and classified.The independent test classification accuracy of the BP NN classifier is 87.60%. It is concluded that the BP NN classifier based on texture feature is suitable for classification rat RN and HCC nodule SPIO-enhanced imaging and patient cirrhosis and normal liver parenchyma T1WI, but classification accuracy of the patient cirrhosis and normal liver parenchyma T1WI is low relatively.
(3) For classification clinical patient cirrhosis T1WI, increasing the number of the texture feature parameters and optimizing the NN classifier by GA. Firstly, direction of GLCM is improved from one to four for extracting texture feature and 14 texture characteristic parameters of every direction are determined. Then feature is chosen by Box Plot. Among the 54 texture feature parameters,24 texture feature parameters which are preferably separability between cirrhosis and normal liver tissue are chosen. In the case of more input texture feature parameters, the BP NN classifier is optimized by GA and GA-BP algorithm is obtained. GA-BP algorithm is a mixed algorithm that the initial value of the BP NN classifier is optimized by GA. On ensuring the classification accuracy, stability and average convergence rate of GA-BP algorithm classifier is improved and the independent test classification accuracy of the GA-BP algorithm classifier is 95%.
本文针对临床RN、HCC结节及早期肝硬化MRI诊断中存在的问题,将动物模型、临床影像诊断技术与医学图像处理和分析技术相结合,发挥多学科交叉的优势,进行了肝脏病变的MRI诊断与计算机辅助诊断研究。首先,通过建立大鼠HCC模型获取RN、HCC结节钆喷替酸葡甲胺(Gd-DTPA)与SPIO增强图像,评估了Gd-DTPA与SPIO双对比增强图像鉴别诊断RN、HCC结节的应用价值,进一步设计了一种基于纹理特征的神经网络分类器用于SPIO增强图像RN和HCC结节的分类识别；其次,应用本文分类器分类识别肝硬化及正常肝脏病人的MRI；最后,进一步采用遗传算法优化基于纹理特征的BP网络分类器。
本文所完成的主要研究工作如下：
(1)大鼠RN、HCC结节SPIO增强图像及SPIO与Gd-DTPA双对比增强图像分析,提高了RN、HCC结节的影像鉴别诊断能力。首先,通过小剂量二乙基亚硝胺诱癌法建立大鼠肝硬化、HCC模型,获得病理证实的肝脏结节106枚(RN 24, HCC 82)。将106枚肝脏结节的SPIO增强图像及SPIO与Gd-DTPA双对比增强图像与病理结果对照分析,结果显示SPIO增强图像HCC结节的对比噪声比(CNR)显著提高(P<0.05),而肝硬化组织的CNR没有显著改变(P>0.05)；SPIO与Gd-DTPA双对比增强图像的HCC结节检出敏感性(96.34%)较单一GD-DTPA增强图像的敏感性(89.02%)有一定的提高,但其统计学差异没有显著性(P>0.05)。上述研究结果表明大鼠RN、HCC结节鉴别中,SPIO增强图像能较平扫图像提高HCC结节的检出和定性诊断能力,SPIO增强图像是Gd-DTPA增强图像有益的补充。
(2)基于纹理特征的BP神经网络分类器用于大鼠及病人肝脏MRI病灶分类识别,获得了较高的分类识别率。首先,利用ISODATA算法自动分割T1WI大鼠肝脏边缘,比较与Otsu阈值选择法和迭代阈值选择法的分割效果,结果显示ISODATA算法更适合于类别未知组织的自动分割问题,为进一步寻找病灶感兴趣区域(ROI)奠定了基础。其次,手动选取大鼠RN、HCC结节SPIO增强图像和病人肝硬化组织及正常肝脏组织T1WI的ROI,并进行纹理特征提取。综合比较后选择基于灰度共生矩阵的纹理特征提取方案,结合医生经验提取二阶矩、对比度、相关、逆差矩、熵、方差六个类间差异显著的纹理特征。针对大鼠RN、HCC结节SPIO增强图像,进行上述纹理特征的统计学分析,选取两组间差异显著的六个纹理特征,并与医生临床经验对比获得一致结论,设计基于纹理特征的神经网络分类器,获得了91.67%的独立测试正确率；针对正常肝脏和肝硬化病人T1WI,统计学分析后选取两组间差异显著的四个纹理特征,设计基于纹理特征的神经网络分类器,获得了87.60%的独立测试正确识别率。两组结果表明本论文所设计的基于纹理特征地的BP网络分类算法适合于大鼠RN、HCC结节SPIO增强图像及肝硬化病人T1WI分类识别,但肝硬化病人T1WI分类识别略低。
(3)针对肝硬化病人T1WI分类识别,增加纹理特征并基于遗传算法进行BP网络优化。首先,灰度共生矩阵增加至四个方向提取纹理特征参数,每个方向确定14个纹理特征参数,共计54个特征参数。采用盒状图分析后,获得可分性较好的24个特征参数,从而实现特征选择。在输入特征较多的情况下,本文并未直接进行BP神经网络设计,而采用遗传算法辅助下的BP网络优化设计：通过遗传算法优化BP神经网络初始值的GA-BP算法,在保证分类准确率的前提下,不仅增加了BP网络的稳定性,同时使BP网络的平均收敛速度提高,获得了95.00%的独立测试分类准确率。
Primary hepatocellular carcinoma (HCC) is a malignant primary liver cancer, and HCC usually occurs as a complication of cirrhosis. Small HCC detected from cirrhotic background makes the successful surgical treatment of HCC and cure being possible. Fibrosis is an early stage of cirrhosis and fibrosis is expected to recover through medicine treatment if it is diagnosed accurately. In recent years, Magnetic resonance imaging (MRI) has been one of the most sensitive methods for diagnosis of cirrhosis and HCC. But it is difficult to diagnose of fibrosis by MRI because morphological changes of fibrosis are often mild. Successful application of novel MRI contrast agent-superparamagnetic iron oxide (SPIO) improves the ability of differential diagnosis regenerative nodule (RN) from HCC nodule, but the SPIO-enhanced imaging can not show the hemodynamic characteristics of hepatic lesions and the false positive rate of SPIO-enhanced imaging is high relatively. It is difficult to diagnose RN from HCC nodule by MRI now.
For the problems of imaging diagnosis of fibrosis and hepatic nodules correlation cirrhosis, this reseach integrates animal expetiments and clinical imaging with imaging processing and imaging analysis technology. Possessing advantages of multiple subject crossing, it carries through the investigation of MR imaging diagnosis and computer aided diagnosis of hepatic lesions. Firstly, this dissertation studies the diagnostic value of SPIO-enhanced imaging and combined SPIO-enhanced and Gadolinium DTPA (Gd-DTPA) enhanced imaging for differential diagnosis of RN and HCC nodule through establishment of rat HCC model. Secondly, a neural network (NN) classifier based on texture feature is designed to distinguish rat hepatic nodule in SPIO-enhanced imaging. Thirdly, clinical patient cirrhosis MR imaging is classified by the NN classifier based on texture feature. Lastly, the initial value of the NN classifier for clinical patient cirrhosis MR imaging is optimized by Ggenetic algorithm (GA) to improve the performance of the classifier. The main research work and contributions of this dissertation can be summarized as follows:
(1) Analysing rat RN and HCC nodule SPIO-enhanced imaging and combined SPIO and Gd-DTPA enhanced imaging and the ability of rat RN and HCC differentiated diagnosis is improved. Rat HCC model is induced by low dose diethylnitrosamine, a total of 106 nodules including 24 RN and 82 HCC are pathologically confirmed. SPIO-enhanced imaging and combined SPIO and Gd-DTPA enhanced imaging of rat 106 modules are analysized compared with pathology. The results show SPIO-enhanced imaging contrast noise ratio (CNR) of HCC nodule is higher than the pre-SPIO enhanced imaging (P< 0.05), SPIO-enhanced imaging CNR of cirrhosis tissue has not significantly changed (P>0.05). The sensitivity of the combined SPIO and Gd-DTPA imaging (96.34%) is higher than the sensitivity of Gd-DTPA imaging alone (89.02%), but the statistical difference is not significant (P>0.05). It is concluded that SPIO-enhanced imaging can improve rat HCC nodule detection and diagnosis capability compared with T2WI, and SPIO-enhanced imaging is a useful complement to the Gd-DTPA enhanced image for rat RN and HCC differentiated diagnosis.
(2) Classifying rat and patient hepatic MR imaging using BP NN classifier based on texture feature and higher classification rate is obtained. Firstly, rat hepatic edge is automatically segmented by ISODATA algorithm, Otsu threshold algorithms and Iterative threshold algorithms on T1WI. The result shows that ISODATA algorithm is suitable for unknown category imaging automatical segmentation, and it is preparative for lesion region of interest (ROI) searching next step. Secondly, ROI of rat RN and HCC nodule SPIO-enhanced imaging and patient cirrhosis and normal liver parenchyma T1WI are cut manually and feature characteristics are extracted by Gray Level Coourrence Matrix (GLCM). Feature extraction program based on texture is chosen after comprehensive comparision. Six texture characteristic parameters including Angular Second Moment, Contrast, Correlation, Inverse Difference Moment, Entropy, and Variance are extracted according to the medical experience. For rat RN and HCC nodule SPIO-enhanced imaging, six texture characteristic parameters which are significantly different between two classes are chosen after statistical analysis. The result is same as clinical doctor's subjective visuality. Then a BP NN classifier is designed and classified.The independent test classification accuracy of the BP NN classifier is 91.67%. For patient cirrhosis and normal liver parenchyma T1WI, four texture characteristic parameters which are significantly different between two classes are chosen after statistical analysis. Then a BP NN classifier is designed and classified.The independent test classification accuracy of the BP NN classifier is 87.60%. It is concluded that the BP NN classifier based on texture feature is suitable for classification rat RN and HCC nodule SPIO-enhanced imaging and patient cirrhosis and normal liver parenchyma T1WI, but classification accuracy of the patient cirrhosis and normal liver parenchyma T1WI is low relatively.
(3) For classification clinical patient cirrhosis T1WI, increasing the number of the texture feature parameters and optimizing the NN classifier by GA. Firstly, direction of GLCM is improved from one to four for extracting texture feature and 14 texture characteristic parameters of every direction are determined. Then feature is chosen by Box Plot. Among the 54 texture feature parameters,24 texture feature parameters which are preferably separability between cirrhosis and normal liver tissue are chosen. In the case of more input texture feature parameters, the BP NN classifier is optimized by GA and GA-BP algorithm is obtained. GA-BP algorithm is a mixed algorithm that the initial value of the BP NN classifier is optimized by GA. On ensuring the classification accuracy, stability and average convergence rate of GA-BP algorithm classifier is improved and the independent test classification accuracy of the GA-BP algorithm classifier is 95%.
引文
[1]周康荣.腹部CT[M].上海：上海医科大学出版社,1998.
[2]周康荣.体部磁共振成像[M].上海：上海医科大学出版社,2000.
[3]Ohno A, Ohta Y, Ohtomo K, et al. Magnetic resonance imaging in chronic liver disease evaluated in relation to hepatic fibrosis clinical and experimental results [J]. Radiat Med,1990,8:159-163.
[4]谭延斌,张敏鸣.肝纤维化的MRI研究进展[J].国际医学放射学杂志,2009,32(2)：139-142.
[5]罗琳,王劲.肝硬化结节自然病程的磁共振功能成像研究进展[J].国际医学放射学杂志,2009,32(3)：245-249.
[6]刑古生.肝细胞肝癌的CT及MR比较影像学分析[D].北京：中国医学科学院北京协和医学院博士研究生学位论文,2009.
[7]Willatt JM, Hussain HK, Adusumilli S, et al. Imaging of hepatocellular carcinoma in the cirrhotic liver:challenges and controversies [J]. Radiology,2008,247:311-330.
[8]Ward J, Robinson PJ. How to detect hepatocellular carcinoma in cirrhosis [J]. Eur Radiol, 2002,12(9):2258-72.
[9]Hanna RF, Aguirre DA, Kased N, et al. Cirrhosis-associated hepatocellular nodules: correlation of histopathologic and MR imaging features [J]. Radiographics,2008, 8:747-769.
[10]Xu H, Kong X, Xiong Y, et al. MR features of regenerative nodules and dysplastic nodules in the cirrhotic liver [J]. J Huazhong Univ Sci Technolog Med Sci,2005,25(5):601-3.
[11]Hussain HK, Syed I, Nghiem HV, et al. T2-weighted MR imaging in the assessment of cirrhotic liver [J]. Radiology,2004,230:637-644.
[12]钱林学,刘玉江,丁惠国.小肝癌的影像学诊断进展[J].世界华人消化杂志,2010,18(5)：479-486
[13]Ito K, Mitchell DG. Hepatic morphologic changes in cirrhosis:MR imaging findings [J]. Abdom. Imag,2000,25:456-461.
[14]Kubo H, Harada M, Ishikawa M, et al. Hemodynamic changes with liver fibrosis measured by dynamic contrast-enhanced MRI in the rat[J].Magn Reson Med Sci,2006,5:65-71.
[15]Hughes CF, Chavez AD, Schlang A, et al. Superparamagnetic iron oxides and low molecular weight gadolinium chelates aresynergistic for direct visualization of advanced liver fibrosis [J]. J Magn Reson Imaging,2007,26:728-737.
[16]Aguirre DA, Behling CA, Alpert E, et al. Liver fibrosis:noninvasive diagnosis with double contrast material-enhanced MR imaging [J]. Radiology,2006,239:425-437.
[17]汪登斌,陈克敏,周康荣.Mn-DPDP增强MRI的实验研究[J].上海医学影像杂志,2002,11(2)：148-151.
[18]Kim JI, Lee JM, Choi JY, et al. The value of gadobenate dimeglumine-enhanced delayed phase MR imaging for characterization of hepatocellular nodules in the cirrhotic liver[J].Invest Radiol,2008,43:202-210.
[19]Hagiwara M, Rusinek H, Lee VS, et al. Advanced liver fibrosis:diagnosis with 3D whole liver perfusion MR imaging-initial experience [J]. Radiology,2008,246:926-934.
[20]Annet L, Materne R, Danse E, et al. Hepatic flow parameters measured with MR imaging and Doppler US:correlations with degree of cirrhosis and portal hypertension[J]. Radiology,2003,229:409-414.
[21]Taouli B, Tolia AJ, Losada M, et al. Diffusion-weighted MRI for quantification of liver fibrosis:preliminary experience [J]. AJR,2007,189:799-806.
[22]Koinuma M, Ohashi I, Hanafusa K, et al. Apparent diffusion coefficient measurements with diffusion-weighted magnetic resonance imaging for evaluation of hepatic fibrosis[J].J Magn Reson Imaging,2005,22:80-85.
[23]Boulanger Y, Amara M, Lepanto L, et al. Diffusion-weighted MR imaging of the liver of hepatitis C patients [J].NMR Biomed,2003,16:132-136.
[24]Lim AK, Patel N, Hamilton G, et al. The relationship of in vivo 31P MR spectroscopy to histology in chronic hepatitis C[J]. Hepatology,2003,37:788-794.
[25]Noren B, Dahlqvist 0, Lundberg P, et al. Separation of advanced from mild fibrosis in diffuse liver disease using 31P magnetic resonance spectroscopy[J]. Eur J Radiol,2008, 66:313-320.
[26]Orlacchio A, Bolacchi F, Angelico M, et al. In vivo high-field 3-Tesla 1H MR spectroscopic assessment of liver fibrosis in HCV correlated chronic liver disease[J]. Radiol Med,2008,113:289-299.
[27]Huwart L, Peeters F, Sinkus R, et al. Liver fibrosis:non invasive assessment with MR elastography [J].NMR Biomed,2006,19:173-179.
[28]Rouviere 0, Yin M, Dresner MA, et al. MR elastography of the liver:preliminary results [J]. Radiology,2006,240:440-448.
[29]Yin M, Woollard J, Wang X, et al. Quantitative assessment of hepatic fibrosis in an animal model with magnetic resonance elastography [J].Magn Reson Med,2007,58: 346-353.
[30]Kim YK, Lee YH, Kim CS, et al. Added diagnostic value of T2-weightedMR imaging to gadolinium-enhanced three-dimensional dynamic MR imaging for the detection of small hepatocellular carcinomas[J]. Eur J Radiol 2007,3517:1-7.
[31]Vanden BIC, Hussain SM, Dwarkasing RS, et al. MR imaging of hepatocellular carcinoma: relationship between lesion size and imaging findings, including signal intensity and dynamic enhancement patterns [J]. J Magn Reson Imaging,2007,26:1548-1555.
[32]Yu JS, Lee JH, Chung JJ, et al. Small Hypervascular Hepatocellular Carcinoma:limited value of portal and delayed phases on dynamic magnetic resonance imaging [J]. Acta Radiol,2008,16:1-9.
[33]Tanimoto A, Kuribayashi S. Application of superparamagnetic iron oxide to imaging of hepatocellular carcinoma [J]. Eur J Radiol,2006,58:200-16.
[34]Macarini L, Milillo P, Cascavilla A, et al. MR characterisation of dysplastic nodules and hepatocarcinoma in the cirrhotic liver with hepatospecific superparamagnetic contrast agents:pathological correlation in explanted livers[J]. Radiol Med, 2009,114(8):1267-82.
[35]Yukisawa S, Okugawa H, Masuya Y, et al. Multidetector helical CT plus superparamagnetic iron oxide-enhanced MR imaging for focal hepatic lesions in cirrhotic liver:A comparison with multi-phase CT during hepatic arteriography [J]. Eur J Radiol, 2007,61:279-289.
[36]Arnold P, Ward J, Wilson D, et al. Superparamagnetic iron oxide (SPIO) enhancement in the cirrhotic liver:a comparison of two doses of ferumoxides in patients with advanced disease [J]. Magnetic Resonance Imaging,2003,21:695-700.
[37]Takeshita K, Nagashima I, Frui S, et al. Effect of superparamagnetic iron oxide-enhanced MRI of the liver with hepatocellular carcinoma and hyperplastic nodule [J]. J Comput Assist Tomogr,2002,26(3):451-5.
[38]Yoo HJ, Lee JM, Lee MW, et al. Hepatocellular carcinoma in cirrhotic liver: double-contrast-enhanced, high-resolution 3. OT-MR imaging with pathologic correlation [J]. Invest Radiol,2008,43:538-546.
[39]Guiu B, Loffroy R, Ben SD, et al. Combined SPIO-gadolinium magnetic resonanceimaging in cirrhotic patients:negative predictive value and role in screening for hepatocellular carcinoma[J]. Abdom Imaging,2008,33(5):520-528.
[40]Macarini L, Marini S, Milillo P, et al. Double-contrast MRI (DC-MRI) in the study of the cirrhotic liver:utility of administering Gd-DTPA as a complement to examinations in which SPIO liver uptake and distribution alterations (SPIO-LUDA) are present and in the identification and characterisation of focal lesions[J].Radiol Med, 2006,111:1087-1102.
[41]卜丽红,王景林卜,张暹东.MRIMn-DPDP增强扫描诊断小HCC的初步研究[J].哈尔滨医科大学学报,2003,37(5)：447-449.
[42]Morana G, Grazioli L, Schneider G, et al. Hypervascular hepatic lesions:dynamic and late enhancement pattern with Gd-BOPTA [J].Acad Radiol,2002,9(2):476-479.
[43]管生.肝脏弥散、灌注成像及肝细胞癌癌前病变的实验研究[D].上海：复旦大学中山学院影像医学与核医学博士论文,2005.
[44]Wang J, Zhang Y, Wolf RL, et al. Amplitude-modulated continuous arterial spin-labeling 3.0-T perfusion MR imaging with a single coil:feasibility study [J]. Radiology,2005,235:218-228.
[45]Yoshioka H, Takahashi N, Yamaguchi M, et al. Double arterial phase dynamic MRI with sensitivity encoding (SENSE) for hypervascular hepatocellular carcinomas[J]. J Magn Reson Imaging,2002,16:259-266.
[46]Xu H,Xie JX, Li X, et al. Perfusion-weighted MRI in evaluating the intranodular hemodynamic characteristics of dysplastic nodules and hepatocellular carcinomas in an experimental rat model [J].J Magn Reson Imaging,2008,27:102-109.
[47]Gourtsoyianni S, Papanikolaou N, Yarmenitis S,et al. Respiratory gated diffusion-weighted imaging of the liver:value of apparent diffusion coefficient measurements in the differentiation between most commonly encountered benign and malignant focal liver lesions [J].Eur Radiol,2008,18:486-492.
[48]Bruegel M, Holzapfel K, Gaa J, et al. Characterization of focal liver lesions by ADC measurements using a respiratory triggered diffusion-weighted single shot echo planar MR imaging technique [J].Eur Radiol,2008,18:477-485.
[49]Demir 01, Obuz F, Sagol 0, et al. Contribution of diffusion-weighted MRI to the differential diagnosis of hepatic masse s [J].Diagn Interv Radiol,2007,13:81-86.
[50]Zhao WD, Guan S, Zhou KR, et al. In vivo detection of metabolic changes by 1H-MRS in the DEN-induced hepatocellular carcinoma in wistar rat [J]. J Cancer Res Clin Oncol,2005,13:597-602.
[51]Towen RA, Foley LM, Painter DM. Hepatocarcinogenesis tumor grading correlated with in vivo image-guided 1H-NMR spectroscopy in a rat model [J].Toxicol Appl Phanmacol,2005,207 (2):237-244.
[52]舒荣宝,王成林.电子计算机辅助诊断(CAD)的原理及临床应用[J].中国CT和MRI杂志,2004：2(2)：56-56.
[53]Doi K. Computer-aided diagnosis in medical imaging:Historical review, current status and future potential [J]. Computerized Medical Imaging and Graphics.2007,31:198-211.
[54]Bradley J. Erickson, Bartholmai B. Computer-aided detection and diagnosis at the start of the third millennium [J].Journal of digital imaging.2002,15(2):59-68.
[55]龚声蓉,刘纯平,王强等.数字图像处理与分析[M].北京：清华大学出版社,2006.
[56]陈武凡,秦安,江少峰,等.医学图像分析的现状与展望[J].中国生物医学工程学报,2008,27(2)：175-181.
[57]Du XY, Li YJ, Yao DZ. A support vector machine based algorithm for magnetic resonance image segmentation [C]. Fourth International Conference on Natural Computation, China, jinan 2008,18-20:49-53.
[58]Grau V, Mewes AUJ, Alcaniz M, et al. Improved watershed transform for medical image segmentation using prior information [J]. IEEE Trans. on Med. Imaging,2004, 23(4):447-458.
[59]刘炎,邱天爽,郭冬梅.基于柔性形态学与聚类算法的X光胸片肺部分割[J].中国生物医学工程学报,2008,26(5)：684-689.
[60]Folkesson J, Samset E, Kwong RK, et al. Unifying statistical classification and geodesic active regions for segmentation of cardiac MRI [J]. IEEE Transactions on information technology in biomedicine.2008,12(3):328:334.
[61]Ballin AA, Galun M, Gomori JM, et al. Automatic segmentation and classification of multiple sclerosis in multichannel MRI [J].IEEE Transactionson biomedical engineering.2009,56(10):2461:2469
[62]Park H, Bland PH, Meyer CR. Construction of an abdominal probabilistic atlas and its application in segmentation [J]. IEEE Trans. on Med. Imaging,2003,22:483-492.
[63]康晓东.医学影像图像分割与存储若干问题的研究[D].天津：天津大学计算机科学与技术学院计算机科学与技术,2007.
[64]延伟东.图像特征提取方法的研究[D].西安：西北工业大学,2007.
[65]谢菲,图像纹理特征的提取和图像分类系统研究及实现[D].成都：电子科技大学,2009.
[66]江少锋.医学图像的特征自动提取及基于模糊特征的图像检索研究[D].广州：南方医科大学生物医学工程学院,2008.
[67]梁琰.肺部CT图像分析及特征提取研究[D].重庆：重庆大学,2007.
[68]张勇,王瓛,郭秀花.基于医学图像纹理特征的疾病诊断方法及其应用现状[J].北京生物医学工程,2008,27(4)：444-447.
[69]薄华,马缚龙,焦李成.图像纹理的灰度共生矩阵计算问题的分析[J].电子学报,2006,34(1)：155-159.
[70]Herlidou M, Constansb J. M., Carsinc B, et al. MRI texture analysis on texture test objects, normal brain and intracranial tumors[J]. Magnetic Resonance Imaging,2003,21: 989-993.
[71]Yao JH, Chen J, Chow C. Breast Tumor Analysis in Dynamic Contrast Enhanced MRI Using Texture Features and Wavelet Transform [J]. IEEE Journal of selected topics in signal processing,2009,3 (1):94-100.
[72]Chevrefils C, Cheriet F, Aubin CE, et al. Texture analysis for automatic segmentation of intervertebral disks of scoliotic spines from MR Images [J]. IEEE Trans Inf Technol Biomed,2009,13 (4):608-620.
[73]汤井田,胡丹,龚智敏.基于SVM的图像纹理特征分类研究[J].计算机工程与科学,2008,130(18)：44-45,48.
[74]Haralick RM, Shanmugam K, Dinstein I. Textural features for image classification [J]. IEEE Trans Syst Man Cybernet [J].1973,3:610-21.
[75]Theodoridis S, Kutroumba K.模式识别[M].北京：电子工业出版社,2006.
[76]Marques JPS.模式识别-原理、方法及应用[M].北京：清华大学出版社,2003.
[77]Joo S, Yang YS, Moon WK, et al. Computer Aided Diagnosis of Solid Breast Nodules:Use of an Artificial Neural Network Based on Multiple Sonographic Features [J]. IEEE Trans Med Imaging,2004,23:1292-1200.
[78]雷元义,陈海东.人工神经网络及其在医学影像分析中的应用[J].医疗卫生装备.2005,26(2)：24-25.
[79]任爱军,高培毅.基于人工神经网络的颅内胶质瘤术后正常反应与肿瘤残存的MRI诊断[J].临床放射学杂志,2002,21(9)：675-678.
[80]Levman J, Leung T, Causer P, et al. Classification of dynamic contrast-enhanced magnetic resonance breast lesions by support vector machines [J]. IEEE Trans Med Imaging, 2008,27:688-696.
[81]Asada N, Doi K, MacMahon H., et al. Potential usefulness of an artificial neural network for differential diagnosis of interstitial lung disease:pilot study [J]. Radiology,1990,177:857-860.
[82]徐伟栋.乳腺X线图像的计算机辅助诊断技术研究[D].杭州：浙江大学生物医学工程与仪器科学学院,2005.
[83]哈章.基于灰阶超声序列图像的乳腺肿瘤计算机辅助诊断[D].合肥：中国科学技术大学生物医学工程,2008.
[84]Boniatis I, Panayiotakis G, Panagiotopoulos E. A computer-based system for the discrimination between normal and degenerated menisci from magnetic resonance images [C]. IEEE International Workshop on Imaging Systems and Techniques -IST 2008, Chania, Greece,2008,10-12:335-339.
[85]薛以锋,鲍旭东,马汉林等.基于CT图像的肺结节计算机辅助诊断系统[J].中国医学物理学杂志,2006,23(2)：93-96.
[86]杨振森.基于超声图像的前列腺病变计算机辅助诊断[D].合肥：中国科学技术大学生物医学工程博士学位论文,2009.
[87]张俊华.颈部淋巴结超声图像分析与应用[D].上海：复旦大学信息学院电子工程系,2007.
[88]陈菲.基于神经网络的超声肝图像识别研究[J].西南科技大学学报.2006,21(3)：36-40
[89]Tajima T, Zhang X, Kitagawa T, et al. Computer-aided detection (CAD) of hepatocellular carcinoma on multiphase CT images[C]. Proc. of SPIE Medical Imaging 2007: Computer-Aided Diagnosis.2007,65142Q-1-65:142Q-10.
[90]Chen EL, Chung PC, Chen CL, et al. An Automatic Diagnostic System for CT Liver Image Classification [J]. IEEE Trans Biomed Eng.1998,45:783-94.
[91]宋伟,金征宇,徐颖,等.计算机辅助诊断软件在CT多期扫描识别和评价小HCC中的应用[J].实用放射学杂志.2007,23(10)：1319-1321.
[92]董育宁,陈唏,杜超.智能肝肿瘤CT图像分割[J].中国生物医学工程学报,2007,26(4)：541-550.
[93]付树军,阮秋琦,李玉等.基于各向异性扩散方程的超声图像去噪与边缘增强[J].电子学报,2005,33(7)：1191-1195.
[94]Yue Yong, Croitoru MM, Bidani A, etal. Nonlinearm ulti-scale wavelet diffusion for speckle suppression and edge enhancem ent in ultrasound mi ages [J]. IEEE Trans Medical Imaging,2006,25(3):297-311.
[95]张继武,蒋大宗.小波技术在生物医学成像中的应用[J].国外医学生物医学工程分册,1996,19(6)：204-209.
[96]She K, PU HM, ZHENG FW. Fast Design of Independent Component Based on Single Pixel under Multisensing [J]. Journal of University of Electronic Science and Technology of China.2008,37 (3):421-424.
[97]Loh CH, Wu TC, Huang NE.Application of the empiricalmode decomposition—Hilbert spectrum method to identify near-faultground motion characteristics and structural responses[J]. Bulletin of the SeismologicalSociety ofAmerica,2001,91(5):1339-1357.
[98]Zhao X, Glahn D, Tan LH, et al.Comparison of TCA and ICA techniques in fMRI data processing[J]. J Magn Reson Imaging.2004,19(4):397-402
[99]Zhang XJ, Fujita H, Qin TF, et al. CAD on Liver Using CT and MRI[J]. Medical imaging and informatics.2008,4987:367-376.
[100]姜慧研,Hiroshi Fujita.多区域肝脏图像分割方法的研究[C]. Proceedings of the 6th world congress on intelligent control and automation. Dalian, China,2006,9992-9995:21-23.
[101]Gao L, Heath D, Kuszyk B, et al. Automatic liver segmentation technique for three-dimensional visualization of CT data [J]. Radiology,1996,201,359-364.
[102]Zhang X, Li W, Fujita H, et al. Automatic segmentation of hepatic tissue and 3D volume analysis of cirrhosis in multidetector row CT scans and MR imaging[J]. IEICE Trans. Inf.& Syst.2004, E87-D (8):2138-2147.
[103]Kato H, Kanematsu M, Zhang X, et al. Computer-aided diagnosis of hepatic fibrosis: Preliminary evaluation of MRI texture analysis using the finite difference method and an artificial neural network [J]. American Roentgen Ray Society.2007,189:117-122.
[104]童隆正,王磊,陈海荣,等.肝纤维化图像的灰度共生矩阵分析[J].首都医科大学学报.2003,24(3)：240-242.
[105]姜慧研,赵越,杨新风.基于MRI的肝硬化程度的计算机辅助诊断[J].东北大学学报(自然科学版).2007,28(1)：31-34.
[106]Zhang XJ, Fujita H, Kanematsu M, et al. Improving the Classification of Cirrhotic Liver by using Texture Features [C]. Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference. Shanghai, China,2005,867-870.
[107]Mala K, Sadasivam V. Wavelet based texture analysis of Liver tumor from Computed Tomography images for characterization using Linear Vector Quantization Neural Network[C]. Advanced Computing and Communications, ADCOM 2006 international Conference on, Surathkal,2006:267-70.
[108]陈真诚,周兆英,赵于前等.人体肝脏组织CT图像的分维特征研究[J].航天医学与医学工程.2005,18(3)：206-210.
[109]Zhang X, Kanematsu M, Fujita H, et al. Computerized classification of liver disease in MRI using artificial neural network [J]. IEIC Technical Report,2001,100(596):69-74.
[110]Gletsos M, Stavroula G, Mougiakakou, et al. A Computer-aided diagnostic system to characterize CT focal liver lesions:Design and optimization of a neural network classifier [J]. IEEE Trans on Information Technology in Biomedicine.2003,7(3): 153-162.
[111]Kako N, Zhang X, Li W, et al. Artificial neural network method for differentiation of focal liver disease in MR imaging[J]. Radiology,2002,255:646.
[112]Valavanis I, Mougiakakou SG, Nikita KS. A Computer aided diagnosis of CT focal liver lesions by an ensemble of neural network and statistical classifiers[C]. IEEE international joint conference on neural network.2004,3:1929-1933.
[113]Fukuda H, Ebara M, Kobayashi A, et al. An image analyzing system using an artificial neural network for evaluating the parenchymal echo pattern of cirrhotic liver and chronic hepatitis [J]. IEEE Trans Biomed Eng,1998,45:396-400.
[114]王莉.超顺磁性氧化铁增强MRI诊断肝占位的动物实验及临床研究[D].上海,第二军医大学影像医学与核医学博士论文,2002.
[115]赵卫东,管生,周康荣,等.DEN诱发大鼠肝细胞癌模型相关结节的MRI观察[J].中国医学影像技术,2005,(21(6)：828-830.
[116]陈晓荣,李肖,周翔平等.大鼠诱导型HCC MRI研究现状[J].华西医学.2005,20(1)：186-187.
[117]李肖,周翔平,唐承薇.SD大鼠诱发性HCC模型的建立[J].中华放射学杂志.2007.41(3)：312-316.
[118]吴伟,李肖,袁元,等.SD大鼠磁共振成像技术[J].四川动物,2005,24(4)：617-620
[119]冯子安,胡振明,施裕新,等.鼠肝MR成像技术参数探讨[J].影像诊断与介入放射学.2000,9(2)：84-86.
[120]边肇祺,张学工等.模式识别(第二版)[M].北京：清华大学出版社,1999.
[121]彭祖赠.模糊数学及其应用[M].武汉：武汉大学出版社,2002.
[122]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京：清华大学出版社,2005.
[123]吴瑞.基于遗传算法的模糊神经网络控制器的优化以及MATLAB与VC数据交换的研究[D].西安：西安理工大学,2007.
[124]黄丽.BP神经网络算法改进及应用研究[D].重庆：重庆师范大学,2008.
[2]周康荣.体部磁共振成像[M].上海：上海医科大学出版社,2000.
[3]Ohno A, Ohta Y, Ohtomo K, et al. Magnetic resonance imaging in chronic liver disease evaluated in relation to hepatic fibrosis clinical and experimental results [J]. Radiat Med,1990,8:159-163.
[4]谭延斌,张敏鸣.肝纤维化的MRI研究进展[J].国际医学放射学杂志,2009,32(2)：139-142.
[5]罗琳,王劲.肝硬化结节自然病程的磁共振功能成像研究进展[J].国际医学放射学杂志,2009,32(3)：245-249.
[6]刑古生.肝细胞肝癌的CT及MR比较影像学分析[D].北京：中国医学科学院北京协和医学院博士研究生学位论文,2009.
[7]Willatt JM, Hussain HK, Adusumilli S, et al. Imaging of hepatocellular carcinoma in the cirrhotic liver:challenges and controversies [J]. Radiology,2008,247:311-330.
[8]Ward J, Robinson PJ. How to detect hepatocellular carcinoma in cirrhosis [J]. Eur Radiol, 2002,12(9):2258-72.
[9]Hanna RF, Aguirre DA, Kased N, et al. Cirrhosis-associated hepatocellular nodules: correlation of histopathologic and MR imaging features [J]. Radiographics,2008, 8:747-769.
[10]Xu H, Kong X, Xiong Y, et al. MR features of regenerative nodules and dysplastic nodules in the cirrhotic liver [J]. J Huazhong Univ Sci Technolog Med Sci,2005,25(5):601-3.
[11]Hussain HK, Syed I, Nghiem HV, et al. T2-weighted MR imaging in the assessment of cirrhotic liver [J]. Radiology,2004,230:637-644.
[12]钱林学,刘玉江,丁惠国.小肝癌的影像学诊断进展[J].世界华人消化杂志,2010,18(5)：479-486
[13]Ito K, Mitchell DG. Hepatic morphologic changes in cirrhosis:MR imaging findings [J]. Abdom. Imag,2000,25:456-461.
[14]Kubo H, Harada M, Ishikawa M, et al. Hemodynamic changes with liver fibrosis measured by dynamic contrast-enhanced MRI in the rat[J].Magn Reson Med Sci,2006,5:65-71.
[15]Hughes CF, Chavez AD, Schlang A, et al. Superparamagnetic iron oxides and low molecular weight gadolinium chelates aresynergistic for direct visualization of advanced liver fibrosis [J]. J Magn Reson Imaging,2007,26:728-737.
[16]Aguirre DA, Behling CA, Alpert E, et al. Liver fibrosis:noninvasive diagnosis with double contrast material-enhanced MR imaging [J]. Radiology,2006,239:425-437.
[17]汪登斌,陈克敏,周康荣.Mn-DPDP增强MRI的实验研究[J].上海医学影像杂志,2002,11(2)：148-151.
[18]Kim JI, Lee JM, Choi JY, et al. The value of gadobenate dimeglumine-enhanced delayed phase MR imaging for characterization of hepatocellular nodules in the cirrhotic liver[J].Invest Radiol,2008,43:202-210.
[19]Hagiwara M, Rusinek H, Lee VS, et al. Advanced liver fibrosis:diagnosis with 3D whole liver perfusion MR imaging-initial experience [J]. Radiology,2008,246:926-934.
[20]Annet L, Materne R, Danse E, et al. Hepatic flow parameters measured with MR imaging and Doppler US:correlations with degree of cirrhosis and portal hypertension[J]. Radiology,2003,229:409-414.
[21]Taouli B, Tolia AJ, Losada M, et al. Diffusion-weighted MRI for quantification of liver fibrosis:preliminary experience [J]. AJR,2007,189:799-806.
[22]Koinuma M, Ohashi I, Hanafusa K, et al. Apparent diffusion coefficient measurements with diffusion-weighted magnetic resonance imaging for evaluation of hepatic fibrosis[J].J Magn Reson Imaging,2005,22:80-85.
[23]Boulanger Y, Amara M, Lepanto L, et al. Diffusion-weighted MR imaging of the liver of hepatitis C patients [J].NMR Biomed,2003,16:132-136.
[24]Lim AK, Patel N, Hamilton G, et al. The relationship of in vivo 31P MR spectroscopy to histology in chronic hepatitis C[J]. Hepatology,2003,37:788-794.
[25]Noren B, Dahlqvist 0, Lundberg P, et al. Separation of advanced from mild fibrosis in diffuse liver disease using 31P magnetic resonance spectroscopy[J]. Eur J Radiol,2008, 66:313-320.
[26]Orlacchio A, Bolacchi F, Angelico M, et al. In vivo high-field 3-Tesla 1H MR spectroscopic assessment of liver fibrosis in HCV correlated chronic liver disease[J]. Radiol Med,2008,113:289-299.
[27]Huwart L, Peeters F, Sinkus R, et al. Liver fibrosis:non invasive assessment with MR elastography [J].NMR Biomed,2006,19:173-179.
[28]Rouviere 0, Yin M, Dresner MA, et al. MR elastography of the liver:preliminary results [J]. Radiology,2006,240:440-448.
[29]Yin M, Woollard J, Wang X, et al. Quantitative assessment of hepatic fibrosis in an animal model with magnetic resonance elastography [J].Magn Reson Med,2007,58: 346-353.
[30]Kim YK, Lee YH, Kim CS, et al. Added diagnostic value of T2-weightedMR imaging to gadolinium-enhanced three-dimensional dynamic MR imaging for the detection of small hepatocellular carcinomas[J]. Eur J Radiol 2007,3517:1-7.
[31]Vanden BIC, Hussain SM, Dwarkasing RS, et al. MR imaging of hepatocellular carcinoma: relationship between lesion size and imaging findings, including signal intensity and dynamic enhancement patterns [J]. J Magn Reson Imaging,2007,26:1548-1555.
[32]Yu JS, Lee JH, Chung JJ, et al. Small Hypervascular Hepatocellular Carcinoma:limited value of portal and delayed phases on dynamic magnetic resonance imaging [J]. Acta Radiol,2008,16:1-9.
[33]Tanimoto A, Kuribayashi S. Application of superparamagnetic iron oxide to imaging of hepatocellular carcinoma [J]. Eur J Radiol,2006,58:200-16.
[34]Macarini L, Milillo P, Cascavilla A, et al. MR characterisation of dysplastic nodules and hepatocarcinoma in the cirrhotic liver with hepatospecific superparamagnetic contrast agents:pathological correlation in explanted livers[J]. Radiol Med, 2009,114(8):1267-82.
[35]Yukisawa S, Okugawa H, Masuya Y, et al. Multidetector helical CT plus superparamagnetic iron oxide-enhanced MR imaging for focal hepatic lesions in cirrhotic liver:A comparison with multi-phase CT during hepatic arteriography [J]. Eur J Radiol, 2007,61:279-289.
[36]Arnold P, Ward J, Wilson D, et al. Superparamagnetic iron oxide (SPIO) enhancement in the cirrhotic liver:a comparison of two doses of ferumoxides in patients with advanced disease [J]. Magnetic Resonance Imaging,2003,21:695-700.
[37]Takeshita K, Nagashima I, Frui S, et al. Effect of superparamagnetic iron oxide-enhanced MRI of the liver with hepatocellular carcinoma and hyperplastic nodule [J]. J Comput Assist Tomogr,2002,26(3):451-5.
[38]Yoo HJ, Lee JM, Lee MW, et al. Hepatocellular carcinoma in cirrhotic liver: double-contrast-enhanced, high-resolution 3. OT-MR imaging with pathologic correlation [J]. Invest Radiol,2008,43:538-546.
[39]Guiu B, Loffroy R, Ben SD, et al. Combined SPIO-gadolinium magnetic resonanceimaging in cirrhotic patients:negative predictive value and role in screening for hepatocellular carcinoma[J]. Abdom Imaging,2008,33(5):520-528.
[40]Macarini L, Marini S, Milillo P, et al. Double-contrast MRI (DC-MRI) in the study of the cirrhotic liver:utility of administering Gd-DTPA as a complement to examinations in which SPIO liver uptake and distribution alterations (SPIO-LUDA) are present and in the identification and characterisation of focal lesions[J].Radiol Med, 2006,111:1087-1102.
[41]卜丽红,王景林卜,张暹东.MRIMn-DPDP增强扫描诊断小HCC的初步研究[J].哈尔滨医科大学学报,2003,37(5)：447-449.
[42]Morana G, Grazioli L, Schneider G, et al. Hypervascular hepatic lesions:dynamic and late enhancement pattern with Gd-BOPTA [J].Acad Radiol,2002,9(2):476-479.
[43]管生.肝脏弥散、灌注成像及肝细胞癌癌前病变的实验研究[D].上海：复旦大学中山学院影像医学与核医学博士论文,2005.
[44]Wang J, Zhang Y, Wolf RL, et al. Amplitude-modulated continuous arterial spin-labeling 3.0-T perfusion MR imaging with a single coil:feasibility study [J]. Radiology,2005,235:218-228.
[45]Yoshioka H, Takahashi N, Yamaguchi M, et al. Double arterial phase dynamic MRI with sensitivity encoding (SENSE) for hypervascular hepatocellular carcinomas[J]. J Magn Reson Imaging,2002,16:259-266.
[46]Xu H,Xie JX, Li X, et al. Perfusion-weighted MRI in evaluating the intranodular hemodynamic characteristics of dysplastic nodules and hepatocellular carcinomas in an experimental rat model [J].J Magn Reson Imaging,2008,27:102-109.
[47]Gourtsoyianni S, Papanikolaou N, Yarmenitis S,et al. Respiratory gated diffusion-weighted imaging of the liver:value of apparent diffusion coefficient measurements in the differentiation between most commonly encountered benign and malignant focal liver lesions [J].Eur Radiol,2008,18:486-492.
[48]Bruegel M, Holzapfel K, Gaa J, et al. Characterization of focal liver lesions by ADC measurements using a respiratory triggered diffusion-weighted single shot echo planar MR imaging technique [J].Eur Radiol,2008,18:477-485.
[49]Demir 01, Obuz F, Sagol 0, et al. Contribution of diffusion-weighted MRI to the differential diagnosis of hepatic masse s [J].Diagn Interv Radiol,2007,13:81-86.
[50]Zhao WD, Guan S, Zhou KR, et al. In vivo detection of metabolic changes by 1H-MRS in the DEN-induced hepatocellular carcinoma in wistar rat [J]. J Cancer Res Clin Oncol,2005,13:597-602.
[51]Towen RA, Foley LM, Painter DM. Hepatocarcinogenesis tumor grading correlated with in vivo image-guided 1H-NMR spectroscopy in a rat model [J].Toxicol Appl Phanmacol,2005,207 (2):237-244.
[52]舒荣宝,王成林.电子计算机辅助诊断(CAD)的原理及临床应用[J].中国CT和MRI杂志,2004：2(2)：56-56.
[53]Doi K. Computer-aided diagnosis in medical imaging:Historical review, current status and future potential [J]. Computerized Medical Imaging and Graphics.2007,31:198-211.
[54]Bradley J. Erickson, Bartholmai B. Computer-aided detection and diagnosis at the start of the third millennium [J].Journal of digital imaging.2002,15(2):59-68.
[55]龚声蓉,刘纯平,王强等.数字图像处理与分析[M].北京：清华大学出版社,2006.
[56]陈武凡,秦安,江少峰,等.医学图像分析的现状与展望[J].中国生物医学工程学报,2008,27(2)：175-181.
[57]Du XY, Li YJ, Yao DZ. A support vector machine based algorithm for magnetic resonance image segmentation [C]. Fourth International Conference on Natural Computation, China, jinan 2008,18-20:49-53.
[58]Grau V, Mewes AUJ, Alcaniz M, et al. Improved watershed transform for medical image segmentation using prior information [J]. IEEE Trans. on Med. Imaging,2004, 23(4):447-458.
[59]刘炎,邱天爽,郭冬梅.基于柔性形态学与聚类算法的X光胸片肺部分割[J].中国生物医学工程学报,2008,26(5)：684-689.
[60]Folkesson J, Samset E, Kwong RK, et al. Unifying statistical classification and geodesic active regions for segmentation of cardiac MRI [J]. IEEE Transactions on information technology in biomedicine.2008,12(3):328:334.
[61]Ballin AA, Galun M, Gomori JM, et al. Automatic segmentation and classification of multiple sclerosis in multichannel MRI [J].IEEE Transactionson biomedical engineering.2009,56(10):2461:2469
[62]Park H, Bland PH, Meyer CR. Construction of an abdominal probabilistic atlas and its application in segmentation [J]. IEEE Trans. on Med. Imaging,2003,22:483-492.
[63]康晓东.医学影像图像分割与存储若干问题的研究[D].天津：天津大学计算机科学与技术学院计算机科学与技术,2007.
[64]延伟东.图像特征提取方法的研究[D].西安：西北工业大学,2007.
[65]谢菲,图像纹理特征的提取和图像分类系统研究及实现[D].成都：电子科技大学,2009.
[66]江少锋.医学图像的特征自动提取及基于模糊特征的图像检索研究[D].广州：南方医科大学生物医学工程学院,2008.
[67]梁琰.肺部CT图像分析及特征提取研究[D].重庆：重庆大学,2007.
[68]张勇,王瓛,郭秀花.基于医学图像纹理特征的疾病诊断方法及其应用现状[J].北京生物医学工程,2008,27(4)：444-447.
[69]薄华,马缚龙,焦李成.图像纹理的灰度共生矩阵计算问题的分析[J].电子学报,2006,34(1)：155-159.
[70]Herlidou M, Constansb J. M., Carsinc B, et al. MRI texture analysis on texture test objects, normal brain and intracranial tumors[J]. Magnetic Resonance Imaging,2003,21: 989-993.
[71]Yao JH, Chen J, Chow C. Breast Tumor Analysis in Dynamic Contrast Enhanced MRI Using Texture Features and Wavelet Transform [J]. IEEE Journal of selected topics in signal processing,2009,3 (1):94-100.
[72]Chevrefils C, Cheriet F, Aubin CE, et al. Texture analysis for automatic segmentation of intervertebral disks of scoliotic spines from MR Images [J]. IEEE Trans Inf Technol Biomed,2009,13 (4):608-620.
[73]汤井田,胡丹,龚智敏.基于SVM的图像纹理特征分类研究[J].计算机工程与科学,2008,130(18)：44-45,48.
[74]Haralick RM, Shanmugam K, Dinstein I. Textural features for image classification [J]. IEEE Trans Syst Man Cybernet [J].1973,3:610-21.
[75]Theodoridis S, Kutroumba K.模式识别[M].北京：电子工业出版社,2006.
[76]Marques JPS.模式识别-原理、方法及应用[M].北京：清华大学出版社,2003.
[77]Joo S, Yang YS, Moon WK, et al. Computer Aided Diagnosis of Solid Breast Nodules:Use of an Artificial Neural Network Based on Multiple Sonographic Features [J]. IEEE Trans Med Imaging,2004,23:1292-1200.
[78]雷元义,陈海东.人工神经网络及其在医学影像分析中的应用[J].医疗卫生装备.2005,26(2)：24-25.
[79]任爱军,高培毅.基于人工神经网络的颅内胶质瘤术后正常反应与肿瘤残存的MRI诊断[J].临床放射学杂志,2002,21(9)：675-678.
[80]Levman J, Leung T, Causer P, et al. Classification of dynamic contrast-enhanced magnetic resonance breast lesions by support vector machines [J]. IEEE Trans Med Imaging, 2008,27:688-696.
[81]Asada N, Doi K, MacMahon H., et al. Potential usefulness of an artificial neural network for differential diagnosis of interstitial lung disease:pilot study [J]. Radiology,1990,177:857-860.
[82]徐伟栋.乳腺X线图像的计算机辅助诊断技术研究[D].杭州：浙江大学生物医学工程与仪器科学学院,2005.
[83]哈章.基于灰阶超声序列图像的乳腺肿瘤计算机辅助诊断[D].合肥：中国科学技术大学生物医学工程,2008.
[84]Boniatis I, Panayiotakis G, Panagiotopoulos E. A computer-based system for the discrimination between normal and degenerated menisci from magnetic resonance images [C]. IEEE International Workshop on Imaging Systems and Techniques -IST 2008, Chania, Greece,2008,10-12:335-339.
[85]薛以锋,鲍旭东,马汉林等.基于CT图像的肺结节计算机辅助诊断系统[J].中国医学物理学杂志,2006,23(2)：93-96.
[86]杨振森.基于超声图像的前列腺病变计算机辅助诊断[D].合肥：中国科学技术大学生物医学工程博士学位论文,2009.
[87]张俊华.颈部淋巴结超声图像分析与应用[D].上海：复旦大学信息学院电子工程系,2007.
[88]陈菲.基于神经网络的超声肝图像识别研究[J].西南科技大学学报.2006,21(3)：36-40
[89]Tajima T, Zhang X, Kitagawa T, et al. Computer-aided detection (CAD) of hepatocellular carcinoma on multiphase CT images[C]. Proc. of SPIE Medical Imaging 2007: Computer-Aided Diagnosis.2007,65142Q-1-65:142Q-10.
[90]Chen EL, Chung PC, Chen CL, et al. An Automatic Diagnostic System for CT Liver Image Classification [J]. IEEE Trans Biomed Eng.1998,45:783-94.
[91]宋伟,金征宇,徐颖,等.计算机辅助诊断软件在CT多期扫描识别和评价小HCC中的应用[J].实用放射学杂志.2007,23(10)：1319-1321.
[92]董育宁,陈唏,杜超.智能肝肿瘤CT图像分割[J].中国生物医学工程学报,2007,26(4)：541-550.
[93]付树军,阮秋琦,李玉等.基于各向异性扩散方程的超声图像去噪与边缘增强[J].电子学报,2005,33(7)：1191-1195.
[94]Yue Yong, Croitoru MM, Bidani A, etal. Nonlinearm ulti-scale wavelet diffusion for speckle suppression and edge enhancem ent in ultrasound mi ages [J]. IEEE Trans Medical Imaging,2006,25(3):297-311.
[95]张继武,蒋大宗.小波技术在生物医学成像中的应用[J].国外医学生物医学工程分册,1996,19(6)：204-209.
[96]She K, PU HM, ZHENG FW. Fast Design of Independent Component Based on Single Pixel under Multisensing [J]. Journal of University of Electronic Science and Technology of China.2008,37 (3):421-424.
[97]Loh CH, Wu TC, Huang NE.Application of the empiricalmode decomposition—Hilbert spectrum method to identify near-faultground motion characteristics and structural responses[J]. Bulletin of the SeismologicalSociety ofAmerica,2001,91(5):1339-1357.
[98]Zhao X, Glahn D, Tan LH, et al.Comparison of TCA and ICA techniques in fMRI data processing[J]. J Magn Reson Imaging.2004,19(4):397-402
[99]Zhang XJ, Fujita H, Qin TF, et al. CAD on Liver Using CT and MRI[J]. Medical imaging and informatics.2008,4987:367-376.
[100]姜慧研,Hiroshi Fujita.多区域肝脏图像分割方法的研究[C]. Proceedings of the 6th world congress on intelligent control and automation. Dalian, China,2006,9992-9995:21-23.
[101]Gao L, Heath D, Kuszyk B, et al. Automatic liver segmentation technique for three-dimensional visualization of CT data [J]. Radiology,1996,201,359-364.
[102]Zhang X, Li W, Fujita H, et al. Automatic segmentation of hepatic tissue and 3D volume analysis of cirrhosis in multidetector row CT scans and MR imaging[J]. IEICE Trans. Inf.& Syst.2004, E87-D (8):2138-2147.
[103]Kato H, Kanematsu M, Zhang X, et al. Computer-aided diagnosis of hepatic fibrosis: Preliminary evaluation of MRI texture analysis using the finite difference method and an artificial neural network [J]. American Roentgen Ray Society.2007,189:117-122.
[104]童隆正,王磊,陈海荣,等.肝纤维化图像的灰度共生矩阵分析[J].首都医科大学学报.2003,24(3)：240-242.
[105]姜慧研,赵越,杨新风.基于MRI的肝硬化程度的计算机辅助诊断[J].东北大学学报(自然科学版).2007,28(1)：31-34.
[106]Zhang XJ, Fujita H, Kanematsu M, et al. Improving the Classification of Cirrhotic Liver by using Texture Features [C]. Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference. Shanghai, China,2005,867-870.
[107]Mala K, Sadasivam V. Wavelet based texture analysis of Liver tumor from Computed Tomography images for characterization using Linear Vector Quantization Neural Network[C]. Advanced Computing and Communications, ADCOM 2006 international Conference on, Surathkal,2006:267-70.
[108]陈真诚,周兆英,赵于前等.人体肝脏组织CT图像的分维特征研究[J].航天医学与医学工程.2005,18(3)：206-210.
[109]Zhang X, Kanematsu M, Fujita H, et al. Computerized classification of liver disease in MRI using artificial neural network [J]. IEIC Technical Report,2001,100(596):69-74.
[110]Gletsos M, Stavroula G, Mougiakakou, et al. A Computer-aided diagnostic system to characterize CT focal liver lesions:Design and optimization of a neural network classifier [J]. IEEE Trans on Information Technology in Biomedicine.2003,7(3): 153-162.
[111]Kako N, Zhang X, Li W, et al. Artificial neural network method for differentiation of focal liver disease in MR imaging[J]. Radiology,2002,255:646.
[112]Valavanis I, Mougiakakou SG, Nikita KS. A Computer aided diagnosis of CT focal liver lesions by an ensemble of neural network and statistical classifiers[C]. IEEE international joint conference on neural network.2004,3:1929-1933.
[113]Fukuda H, Ebara M, Kobayashi A, et al. An image analyzing system using an artificial neural network for evaluating the parenchymal echo pattern of cirrhotic liver and chronic hepatitis [J]. IEEE Trans Biomed Eng,1998,45:396-400.
[114]王莉.超顺磁性氧化铁增强MRI诊断肝占位的动物实验及临床研究[D].上海,第二军医大学影像医学与核医学博士论文,2002.
[115]赵卫东,管生,周康荣,等.DEN诱发大鼠肝细胞癌模型相关结节的MRI观察[J].中国医学影像技术,2005,(21(6)：828-830.
[116]陈晓荣,李肖,周翔平等.大鼠诱导型HCC MRI研究现状[J].华西医学.2005,20(1)：186-187.
[117]李肖,周翔平,唐承薇.SD大鼠诱发性HCC模型的建立[J].中华放射学杂志.2007.41(3)：312-316.
[118]吴伟,李肖,袁元,等.SD大鼠磁共振成像技术[J].四川动物,2005,24(4)：617-620
[119]冯子安,胡振明,施裕新,等.鼠肝MR成像技术参数探讨[J].影像诊断与介入放射学.2000,9(2)：84-86.
[120]边肇祺,张学工等.模式识别(第二版)[M].北京：清华大学出版社,1999.
[121]彭祖赠.模糊数学及其应用[M].武汉：武汉大学出版社,2002.
[122]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京：清华大学出版社,2005.
[123]吴瑞.基于遗传算法的模糊神经网络控制器的优化以及MATLAB与VC数据交换的研究[D].西安：西安理工大学,2007.
[124]黄丽.BP神经网络算法改进及应用研究[D].重庆：重庆师范大学,2008.