分享:
分享到微信朋友圈
X
临床研究
基于扩散张量成像的纹理分析鉴别肝细胞癌和肝内胆管细胞癌的价值
王嫚 刘爱连 赵莹 林涛 王楠 宋清伟 郭妍 李昕 吴艇帆

Cite this article as: Wang M, Liu AL, Zhao Y, et al. The value of texture analysis based on diffusion tensor imaging in distinguishing hepatocellular carcinoma from intrahepatic cholangiocarcinoma[J]. Chin J Magn Reson Imaging, 2021, 12(8): 15-21.引用本文:王嫚, 刘爱连, 赵莹, 等. 基于扩散张量成像的纹理分析鉴别肝细胞癌和肝内胆管细胞癌的价值[J]. 磁共振成像, 2021, 12(8): 15-21. DOI:10.12015/issn.1674-8034.2021.08.004.


[摘要] 目的 探讨扩散张量成像(diffusion tensor imaging,DTI)纹理分析鉴别肝细胞癌(hepatocellular carcinoma,HCC)和肝内胆管细胞癌(intrahepatic cholangiocarcinoma,ICC)的价值。材料与方法 回顾性研究于大连医科大学附属第一医院接受上腹部1.5 T MRI (Signa HDXT,GE healthcare)检查且病理证实为HCC (52例)和ICC (28例)的患者资料。重建DTI图像,生成ADC和各向异性分数(fractional anisotropy,FA)图。两位观察者(分别具有2年和8年影像诊断经验)勾画肿瘤所有层面的ROI,使用AK (artificial intelligent kit,GE Healthcare)软件提取纹理特征。采用组内相关系数评估一致性,独立样本t检验或Mann-Whitney U检验比较各参数的差异,对有统计学意义的参数绘制ROC曲线。采用Logistic回归建立联合诊断模型。DeLong检验比较单一参数与联合诊断效能的差异。结果 两观察者测量结果一致性良好(组内相关系数>0.75)。HCC组ADC信号强度图的最大值、平均值、方差、标准差及熵小于ICC组;能量、峰度及相关性大于ICC组(P<0.05)。HCC组FA信号强度图的最大值、方差、标准差及长游程优势小于ICC组;相关性及短游程优势大于ICC组(P<0.05)。其余参数差异无统计学意义(P>0.05)。单个参数中,ADC-相关性鉴别HCC与ICC的效能最高,AUC为0.856,敏感度为75.0%,特异度为82.1%。当ADC-能量、FA-最大值、FA-短游程优势3个参数联合或者ADC-能量、ADC-相关性、FA-最大值、FA-短游程优势4个参数联合时均可获得最佳诊断效能,AUC值为0.877,敏感度为78.6%,特异度为84.6%。Delong检验显示联合诊断与多个参数相比较效能有显著提升(P<0.05)。结论 基于DTI的纹理分析可以提供多个参数鉴别诊断HCC与ICC。
[Abstract] Objective To explore the value of texture analysis based on diffusion tensor imaging (DTI) in distinguishing hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC). Materials andMethods The data of patients who underwent 1.5 T MRI examination of upper abdomen in the first Affiliated Hospital of Dalian Medical University and pathologically confirmed HCC (52 cases) or ICC (28 cases) were studied retrospectively. The DTI images were reconstructed to generate apparent diffusion coefficient (ADC) and fractional anisotropy (FA) maps. ROIs covering the entire tumor were drawn on each slice of ADC and FA signal intensity maps by two observers (with 2 years and 8 years of imaging diagnosis experience). The texture parameters of the tumor were extracted by AK software. The intra-class correlation coefficient (ICC) was used to test the consistency of the data. The independent samples t test or Mann-Whitney U test was used to compare the differences of the parameters. ROC curves were plotted to analyze diagnostic efficiency. Using Logistic regression analysis to establish the predictive model. The Delong test was used to compare the difference in efficacy between combined diagnosis and single parameter.Results The data consistency of two observers was good (intraclass correlation coeficient >0.75). Maxintensity, mean value, variance, standard deviation and entropy of ADC signal intensity graph in HCC group were smaller than those in ICC group. Energy, kurtosis and correlation were larger than those in ICC group (P<0.05). Maxintensity, variance, standard deviation, and LRE of the FA signal intensity map in the HCC group were smaller than those of the ICC group; the correlation and SRE were larger than the ICC group (P<0.05). There was no statistical difference in other parameters (P>0.05). The diagnostic efficiency of ADC-correlation was the best, AUC, sensitivity and specificity scores were 0.856, 75.0%, 82.1%.The best diagnosis can be obtained when the three parameters of ADC-energy, FA-maximum, and FA-short-run advantage are combined or the four parameters of ADC-energy, ADC-correlation, FA-maximum, and FA-short-run advantage are combined. The AUC was 0.877, the sensitivity and the specificity scores were 78.6%, 84.6%. Delong test showed statistical differences between combined diagnosis and multiple parameters (P<0.05).Conclusions DTI-based texture analysis can provide multiple parameters for differential diagnosis of HCC and ICC.
[关键词] 肝细胞癌;肝内胆管细胞癌;扩散张量成像;纹理分析;磁共振成像
[Keywords] hepatocellular carcinoma;intrahepatic cholangiocarcinoma;diffusion tensor imaging;texture analysis;magnetic resonance imaging

王嫚 1   刘爱连 1*   赵莹 1   林涛 1   王楠 1   宋清伟 1   郭妍 2   李昕 2   吴艇帆 2  

1 大连医科大学附属第一医院放射科,大连 116011

2 通用电气医疗,上海 200000

刘爱连,E-mail:liuailian@dmu.edu.cn

全体作者均声明无利益冲突。


基金项目: 国家自然科学基金 61971091
收稿日期:2021-04-09
接受日期:2021-06-11
DOI: 10.12015/issn.1674-8034.2021.08.004
引用本文:王嫚, 刘爱连, 赵莹, 等. 基于扩散张量成像的纹理分析鉴别肝细胞癌和肝内胆管细胞癌的价值[J]. 磁共振成像, 2021, 12(8): 15-21. DOI:10.12015/issn.1674-8034.2021.08.004.

       肝细胞癌(hepatocellular carcinoma,HCC)是最常见的原发性肝脏恶性肿瘤。肝内胆管细胞癌(intrahepatic cholangiocarcinoma,ICC)的发生率仅次于HCC,恶性程度高,易转移且术后容易复发,长期生存率低于HCC[1, 2]。肝切除术和肝移植是早期HCC患者推荐的治疗方法,中晚期推荐经肝动脉栓塞化疗(transcatheter arterial chemoembolization,TACE)[3, 4]。对于ICC而言,手术切除是唯一可能治愈的方法[5]。早期准确鉴别两者对治疗方式的选择和预后的评估至关重要。组织活检是鉴别HCC与ICC的“金标准”,但此有创检查存在种植转移、出血、感染等风险[6]。常规磁共振检查在病变鉴别方面虽有较大优势,但在鉴别不典型病变时存在困难。扩散张量成像(diffusion tensor imaging,DTI)是一种在扩散加权成像(diffusion-weighted imaging,DWI)基础上发展而来的无创性检查技术,能在三维空间内定量描述水分子的扩散运动及相关性,更好地反映微观结构特点。既往二维ROI的传统测量方法无法反映肿瘤整体的异质性,而通过纹理分析进行的三维测量可获得多个定量参数评估肿瘤异质性。本研究旨在探究基于DTI的纹理分析鉴别HCC与ICC的价值。

1 材料与方法

1.1 一般资料

       回顾性研究于大连医科大学附属第一医院接受上腹部MRI检查且病理证实为HCC或ICC的患者资料。纳入标准:①经病理证实为HCC或ICC;②患者在术前1个月内进行上腹部MRI扫描,包括T1WI、T2WI、增强扫描和DTI序列。排除标准:①MRI扫描前进行其他抗肿瘤治疗,包括TACE、射频消融术、化疗及放疗;②DTI图像质量欠佳无法分析。最后入组80例患者,其中HCC组52例,男46例,女6例,年龄43~76 (59.48±9.48)岁;ICC组28例,男17例,女11例,年龄41~81 (61.64±8.28)岁。本研究经过大连医科大学附属第一医院医学伦理委员会批准(批准文号:PJ-KS-KY-2019-167),免除受试者知情同意。

1.2 一般征象分析

       记录患者的年龄、性别、HbsAg、AFP、CEA、CA19-9、肿瘤组织学分级、肿瘤大小、数量、病灶边界是否清晰、病灶邻近肝被膜是否皱缩、肝内胆管是否扩张、是否合并肝硬化、是否侵犯门静脉、是否有腹腔淋巴结肿大。

1.3 仪器与扫描参数

       所有患者采用1.5 T磁共振扫描仪(Signa HDXT,GE healthcare)进行上腹部MRI扫描,8通道腹部线圈,呼吸补偿。扫描前需禁食、禁水4 h。扫描序列及参数为:①轴位T1WI序列:采用快速扰相梯度回波序列,包括同相位T1WI和反向位T1WI,TR/TE=7.5 ms/4.5 ms,2.2 ms,FOV=40 cm×36 cm,矩阵=288×170,NEX=1,层厚4 mm,层间距1 mm;②轴位T2WI序列:采用快速自旋回波序列,TR/TE=6667 ms/93.8 ms,FOV=40 cm× 30 cm,矩阵=256×256,NEX=2,层厚6.5 mm,层间距1.5 mm;③轴位DTI序列:采用平面回波序列,TR/TE=5050 ms/81.9 ms,FOV=40 cm×40 cm,矩阵=96×96,NEX=2,b=0、600 s/mm2,在6个正交方向施加扩散梯度;④增强扫描序列:经肘静脉以2.5 mL/s的速度注射对比剂钆喷酸葡胺(gadolinium-diethylenetriamine pentaacetic acids,Gd-DTPA),剂量0.1 mmol/kg,分别于对比剂注射后15~20、55~60、180 s采用3D T1WI行动脉期、门静脉期及延迟期扫描。

1.4 图像分析及处理

       在GE AW4.6工作站,利用Functool软件对DTI图像进行重建,生成ADC和FA图。将ADC和FA信号强度图导入ITK-SNAP软件,由两位观察者(分别具有2年和8年影像诊断经验)分别沿肿瘤边缘勾画肿瘤全容积的ROI(肿瘤多发选取最大病灶分析) (见图12),将所有层面的ROI拟合成为三维容积感兴趣区(volume of interest,VOI)。使用AK软件(artificial intelligent kit,GE Healthcare)提取纹理参数,包括最大值、最小值、平均值、方差、标准差、能量、熵、峰度、偏度、相关性、长游程优势及短游程优势。

图1  男,51岁,肝左叶肝细胞癌。A~F:分别为T1WI、T2WI、增强扫描动脉期、门静脉期、延迟期、DTI图像; G~J:分别在ADC与FA信号强度图上勾画全肿瘤ROI,并生成相应的直方图
Fig. 1  Male, 51 years old, HCC in the left lobe of the liver. A—F: T1WI, T2WI, MRI enhanced scan of arterial phase, portal vein phase, delayed phase, DTI; G—J: ROIs were drawn covering the entire tumor on each slice of ADC and FA signal intensity maps, ADC and FA signal intensity histograms of HCC, respectively.
图2  男,56岁,肝方叶与肝右叶交界处肝内胆管细胞癌。A~F:分别为T1WI、T2WI、增强扫描动脉期、门静脉期、延迟期、DTI图像。G~J:分别在ADC与FA信号强度图上勾画全肿瘤ROI,并生成相应的直方图
Fig. 2  Male, 56 years old, ICC in the junction of the quadrate lobe and the right lobe of the liver. A—F: T1WI, T2WI, MRI enhanced scan of arterial phase, portal vein phase, delayed phase, DTI; G—J: ROIs were drawn covering the entire tumor on each slice of ADC and FA signal intensity maps, ADC and FA signal intensity histograms of ICC, respectively.

1.5 统计学分析

       使用SPSS 21.0统计学软件进行统计分析。采用χ2检验比较两组间分类变量的差异,P<0.05表示差异有统计学意义。使用组内相关系数评估两观察者测量结果的一致性,组内相关系数>0.75代表一致性较好。若一致性较好,取两观察者测量数据的平均值进行统计学分析。采用Shapiro-Wilk检验连续变量是否符合正态分布。符合正态分布的数值变量用x¯±s表示,使用独立样本t检验比较两组间各参数的差异;不符合正态分布的数值变量以中位数(25百分位数,75百分位数)表示,使用Mann-Whitney U检验比较两组间各参数的差异。对有统计学意义的参数绘制ROC曲线。在ADC和FA信号强度图中分别选择效能最佳的两个参数[7],采用Logistic回归将这4个参数进行两两联合、三三联合、全部联合,并绘制ROC曲线评估联合诊断的效能。采用DeLong检验比较单个参数与联合诊断AUC值间差异,P<0.05表示差异有统计学意义。

2 结果

2.1 患者一般资料

       患者的临床资料见表1。两组患者性别、HbsAg、AFP、CEA、CA19-9、肿瘤组织学分级、病灶边界、肝被膜皱缩、胆管扩张、肝硬化及腹腔淋巴结肿大差异存在统计学意义(P<0.05),其余差异无统计学意义(P>0.05)。

表1  HCC组与ICC组患者临床资料比较
Tab. 1  Comparison of clinical data between HCC and ICC group

2.2 两观察者一致性检验结果

       两观察者测量各纹理参数一致性良好(组内相关系数均>0.75)。

2.3 HCC与ICC组间ADC与FA信号强度纹理参数结果比较

       HCC与ICC组间ADC与FA信号强度纹理参数结果比较见表2。HCC组ADC信号强度图的最大值、平均值、方差、标准差及熵小于ICC组(P<0.05);能量、峰度及相关性均大于ICC组(P<0.05)。HCC组FA信号强度图的最大值、方差、标准差、长游程优势小于ICC组(P<0.05);相关性及短游程优势大于ICC组(P<0.05)。其余参数差异无统计学意义(P>0.05)。

表 2  HCC与ICC组间ADC及FA信号强度纹理参数结果比较
Tab. 2  Comparison of ADC and FA texture parameters between HCC and ICC groups

2.4 ADC与FA信号强度纹理参数鉴别HCC与ICC的ROC分析以及联合诊断结果

       ADC与FA纹理参数鉴别HCC与ICC的ROC分析结果见表3,ROC曲线见图3、4。单个纹理参数中,ADC-相关性鉴别HCC与ICC的效能最高,AUC为0.857,敏感度为74.5%,特异度为82.1%。联合诊断的ROC曲线见图5、6。ADC-能量、FA-最大值、FA-短游程优势3个参数联合或ADC-能量、ADC-相关性、FA-最大值、FA-短游程优势4个参数联合时均可获得最佳诊断效能,AUC值、敏感度、特异度相同,分别为0.877、78.6%、84.6%。DeLong检验显示,ADC信号强度图的最大值、平均值、方差、标准差、熵、峰度以及FA信号强度图的最大值、方差、标准差、相关性、长游程优势、短游程优势的鉴别诊断效能显著低于上述3个参数或4个参数联合时的诊断效能,AUC值间差异有统计学意义(P<0.05),而ADC信号强度图的能量和相关性与联合诊断的AUC值间差异无统计学意义(P>0.05)。

图3  ADC信号强度图最大值、平均值、方差、标准差、能量、熵、峰度及相关性鉴别HCC与ICC的ROC曲线分析
图4  FA信号强度图最大值、方差、标准差、相关性、长游程优势及短游程优势鉴别HCC与ICC的ROC曲线分析
图5  ADC-能量、FA-最大值、FA-短游程优势联合时鉴别HCC与ICC的ROC曲线分析 图6 ADC-能量、ADC-相关性、FA-最大值、FA-短游程优势联合时鉴别HCC与ICC的ROC曲线分析
Fig. 3  ROC curve of maxintensity, mean value, variance, standard deviation, energy, entropy, kurtosis, correlation of ADC signal intensity for differentiating HCC and ICC.
Fig. 4  ROC curve of maxintensity, variance, standard deviation, correlation, LRE, SRE of FA signal intensity for differentiating HCC and ICC.
Fig. 5  ROC curve when ADC-energy, FA-maxintensity and FA-SRE were combined for differentiating HCC and ICC. Fig.6 ROC curve when ADC-energy, ADC-correlation, FA-maxintensity and FA-SRE were combined for differentiating HCC and ICC.
表 3  HCC与ICC组ADC和FA信号强度纹理参数的ROC分析结果
Tab. 3  ROC analysis of ADC and FA texture parameters between HCC and ICC groups

3 讨论

       HCC与ICC是最常见的肝脏原发性恶性肿瘤,由于治疗方法及预后不同,及早对两者进行鉴别诊断对提高患者生存质量、延长生存期限尤为重要。影像学检查在鉴别诊断方面有较大优势。常规MRI平扫结合动态增强扫描时HCC的“快进快出”,ICC的延迟强化伴肝被膜皱缩与胆管扩张是鉴别两者的典型表现。但临床上一些不典型的HCC与合并肝硬化背景并且无胆管扩张征象的ICC影像学表现有一定的相似之处。此时常规的MRI平扫结合增强鉴别两者较为困难[8, 9]

       近年来,DWI已广泛应用在肝脏良恶性肿瘤的鉴别诊断、肝脏肿瘤分级等方面[10, 11]。DWI的ADC值能够反映水分子的扩散程度,值越小,扩散运动越受限[12],通过ADC值可以判断病变的性质。但DWI仅能在某一水平面内反映水分子的扩散运动,无法评估组织的各向异性。DTI是一种在DWI基础上发展而来的无需对比增强的无创性检查技术,具有多个成像参数,其特有的FA值能够反映水分子扩散运动的方向性,FA值越大,代表方向性越好[13, 14],从而在三维立体空间反映水分子的扩散运动及相关性,更精细地反映微观结构。近年来,DTI技术在评估神经系统的病变中已广泛应用。在肝脏,诊断和评价慢性病毒性肝炎纤维化及肝硬化、肝脏局灶性病变的检出和鉴别等方面DTI也发挥了重要作用[15, 16]。陈丽华等[17]探讨了DTI技术术前鉴别HCC与ICC的价值,发现FA值较ADC值更能有效地鉴别ICC与HCC,AUC值为0.760。然而,传统ADC和FA值的测量受人为因素的影响较大,ROI放置位置的不同、只对肿瘤几个层面进行测量,这些因素都可能导致实验结果无法准确反映肿瘤整体的异质性。对肿瘤所有层面进行勾画并通过纹理分析获得多个定量参数能够更准确地反映肿瘤的异质性并且有助于提升鉴别诊断效能。

       本研究结果表明,HCC多好发于男性,常合并有乙肝感染和肝硬化;AFP为HCC肿瘤标志物,通常有升高;因假包膜的存在肿瘤边界多清晰;与杭轶等[18]、Wildner等[19]的研究结果一致。ICC组的CEA、CA19-9阳性率显著高于HCC组,说明二者对鉴别ICC具有较大的意义,既往研究[20]中将二者联合诊断AFP阴性的ICC,诊断率得到提高。ICC起源于胆管上皮细胞,癌细胞沿胆管侵犯会引起胆管扩张,靠近边缘的ICC可导致局部肝叶萎缩影像学上表现为邻近肝被膜的皱缩;因肿瘤组织学分级多为低分化,侵袭性较强,常会导致肿瘤侵犯门静脉以及向淋巴结转移,病灶边界多不清晰;这与Wildner等[19]、王悦超等[21]的研究结果相符。本研究结果显示,HCC组ADC信号强度图的最大值、平均值大于ICC组,分析原因可能与HCC和ICC组织学成分和构成有关。HCC的肿瘤细胞密度较ICC更高,细胞间的间隙小,水分子扩散运动受限的程度高,因此ADC值降低;ICC中央由大量疏松纤维组织构成,肿瘤细胞多位于周边可排列成腺管样,促进了水分子的扩散运动,使ADC值增加[1722]。方差和标准差代表强度水平分布的不对称程度;能量反映强度水平分布的均匀性,熵是能量值的倒数,能量值越小,熵值越大,图像越不均匀;峰度衡量强度水平的集中与分散程度,值越大,表明图像的灰度更集中于平均值;相关性表示像素在整个图像上与其相邻像素的相关程度,值越大,相关性越大。这些参数均可用于衡量组织的均匀性。本研究中,HCC组ADC的方差、标准差、熵小于ICC组,能量、峰度以及相关性大于ICC组,证明了ICC比HCC具有更高的异质性。根据本研究的结果推测可能是:与HCC相比,ICC分化程度较低,恶性程度更高,细胞异型性更明显,坏死、出血、囊变及侵犯周围组织可能性更大[20],因此ICC的均匀性较差。在FA信号强度图中,HCC组最大值、长游程优势小于ICC组,短游程优势及相关性大于ICC组。推测ICC组FA的最大值较高可能是ICC内的腺管与较多的纤维成分导致水分子的方向性更好[12],因此FA值较高。短游程优势与长游程优势反映图像的光滑与粗糙程度,短游程优势越大,图像的纹理光滑,长游程优势越大,图像的纹理越粗糙[23, 24],表明ICC较HCC更不均匀,与ADC参数得出的结果一致。本研究中ADC-相关性鉴别HCC与ICC效能最佳(AUC=0.856)。在方勇超等[25]的研究中,常规MRI动态增强扫描鉴别ICC的准确率为75%,相比之下基于DTI的纹理分析具有一定的优势。为了避免单一参数造成结果偏倚,故建立Logistic回归模型,当ADC-能量、FA-最大值、FA-短游程优势3个参数联合或者ADC-能量、ADC-相关性、FA-最大值、FA-短游程优势4个参数联合时得到最佳诊断效能(AUC=0.877),并且比多个单一纹理参数的鉴别效能有显著提升。

       本研究的局限性:样本量较小,结果可能存在偏倚,需扩大样本量进一步验证;沿着肿瘤边缘进行病灶勾画时,难以避免人为因素的干扰;本文采用手动分割肿瘤图像,费时耗力,未来会尝试采用全自动或半自动分割方法简化分割流程;分析的纹理参数有限,未结合T2WI与增强图像共同进行纹理分析,将在未来纳入更多的参数构建影像组学模型。

       综上所述,基于DTI的纹理分析可以提供多种定量参数,具有反映肿瘤异质性的潜力,对HCC与ICC的鉴别诊断有一定价值。

1
安澜, 曾红梅, 冉显会, 等. 肝细胞癌和肝内胆管细胞癌流行病学研究进展[J]. 中国肿瘤, 2020, 29(11): 879-884. DOI: 10.11735/j.issn.1004-0242.2020.11.A015.
An L, Zeng HM, Ran XH, et al. Progress on epidemiology of hepatocellular carcinoma and intrahepatic cholangiocarcinoma[J]. Chin Cancer, 2020, 29(11): 879-884. DOI: 10.11735/j.issn.1004-0242.2020.11.A015.
2
Rahnemai-Azar AA, Pandey P, Kamel I, et al. Monitoring outcomes in intrahepatic cholangiocarcinoma patients following hepatic resection[J]. Hepat Oncol, 2016, 3(4): 223-239. DOI: 10.2217/hep-2016-0009.
3
谭运华, 商阳阳, 张涛, 等. 肝细胞癌综合治疗的现状和前景[J]. 临床肝胆病杂志, 2019, 35(8): 1858-1860. DOI: CNKI:SUN:LCGD.0.2019-08-061.
Tan YH, Shang YY, Zhang T, et al. Current status and perspectives of multimodality therapy for hepatocellular carcinoma[J]. J Clin Hepatol, 2019, 35(8): 1858-1860. DOI: CNKI:SUN:LCGD.0.2019-08-061.
4
朱姗薇, 郑绯, 潘一敏, 等. 肝细胞癌(HCC)的常用治疗方法概述及奥沙利铂在晚期HCC治疗中的临床应用[J].首都食品与医药, 2020, 27(14): 44-46. DOI: CNKI:SUN:YYSD.0.2020-14-034.
Zhu SW, Zheng F, Pan YM, et al. Overview of common treatment methods for hepatocellular carcinoma (HCC) and clinical application of oxaliplatin in the treatment of advanced HCC[J]. Capit Food Med, 2020, 27(14): 44-46. DOI: CNKI:SUN:YYSD.0.2020-14-034.
5
Dodson RM, Weiss MJ, Cosgrove D, et al. Intrahepatic cholangiocarcinoma: management options and emerging therapies[J]. J Am Coll Surg, 2013, 217(5): 736-750. DOI: 10.1016/j.jamcollsurg.2013.05.021.
6
Cillo U, Fondevila C, Donadon M, et al. Surgery for cholangiocarcinoma[J]. Liver Int, 2019, 16: 143-155. DOI: 10.1111/liv.14089.
7
王楠, 刘爱连, 李烨, 等. 基于单源双能CT平扫图像的纹理分析对肝脓肿和肝转移瘤的鉴别价值[J]. 放射学实践, 2019, 34(11): 1246-1250. DOI: 10.13609/j.cnki.1000-0313.2019.11.015.
Wang N, Liu AL, Li Y, et al. Differentiation of liver abscess and metastatic tumor: based on texture analysis on non-enhanced spectral CT[J]. Radiol Pract, 2019, 34(11): 1246-1250. DOI: 10.13609/j.cnki.1000-0313.2019.11.015.
8
肖扬锐, 毛卫波, 李炳荣, 等. 富动脉血供肿块型肝内胆管细胞癌的MRI特征及其病理基础[J]. 温州医科大学学报, 2020, 50(10): 800-804. DOI: CNKI:SUN:WZYX.0.2020-10-005.
Xiao YR, Mao WB, Li BR, et al. The MRI features and pathology of intrahepatic mass-forming cholangiocarcinoma with rich arterial blood supply[J]. J Wenzhou Med Univ, 2020, 50(10): 800-804. DOI: CNKI:SUN:WZYX.0.2020-10-005.
9
Ciresa M, De Gaetano AM, Pompili M, et al. Enhancement patterns of intrahepatic mass-forming cholangiocarcinoma at multiphasic computed tomography and magnetic resonance imaging and correlation with clinicopathologic features[J]. Eur Rev Med Pharmacol Sci, 2015, 19(15): 2786-2797.
10
代小兵, 刘启榆, 吴俊辉, 等. 磁共振增强扫描弥散加权成像对肝脏良恶性肿瘤的鉴别诊断价值研究[J]. 实用肝脏病杂志, 2021, 24(2): 268-271. DOI: CNKI:SUN:GBSY.0.2021-02-031.
Dai XB, Liu QY, Wu JH, et al. Diagnosis of focal lesions of liver by diffusion-weighted magnetic resonance imaging[J]. J Pract Hepatol, 2021, 24(2): 268-271. DOI: CNKI:SUN:GBSY.0.2021-02-031.
11
Surov A, Pech M, Omari J, et al. Diffusion-weighted imaging reflects tumor grading and microvascular invasion in hepatocellular carcinoma[J]. Liver Cancer, 2021, 10(1): 10-24. DOI: 10.1159/000511384.
12
van Valerie Phi, Reiner CS, Klarhoefer M, et al. Diffusion tensor imaging of the abdominal organs: Influence of oriented intravoxel flow compartments[J]. NMR Biomed, 2019, 32(11): 4159. DOI: 10.1002/nbm.4159.
13
程启超, 王燕鸣, 李菲, 等. DTI及DTT在评估Ⅰ期子宫内膜癌肌层浸润深度的价值[J]. 磁共振成像, 2020, 11(7): 497-500. DOI: 10.12015/issn.1674-8034.2020.07.004.
Cheng QC, Wang YM, Li F, et al. The value of DTI and DTT in the depth of myometrial invasion of endometrial cancer stage Ⅰ[J]. Chin J Magn Reson Imaging, 2020, 11(7): 497-500. DOI: 10.12015/issn.1674-8034.2020.07.004.
14
Li R, Chang N, Liu YL, et al. The integrity of the substructure of the corpus callosum in patients with right classic trigeminal neuralgia[J]. J Craniofac Surg, 2021, 32(2): 632-636. DOI: 10.1097/SCS.0000000000007082.
15
Huang MP, Lu X, Wang XF, et al. Diffusion tensor imaging quantifying the severity of chronic hepatitis in rats[J]. BMC Med Imaging, 2020, 20(1): 74. DOI: 10.1186/s12880-020-00466-3.
16
Erturk SM, Ichikawa T, Kaya E, et al. Diffusion tensor imaging of cysts, hemangiomas, and metastases of the liver[J]. Acta Radiol, 2014, 55(6): 654-660. DOI: 10.1177/0284185113504916.
17
陈丽华, 刘爱连, 宋清伟, 等. 磁共振扩散张量成像鉴别诊断肝内胆管细胞癌与肝细胞癌[J]. 中国医学影像技术, 2017, 33(7): 993-997. DOI: 10.13929/j.1003-3289.201611140.
Chen LH, Liu AL, Song QW, et al. Diffusion tensor imaging in differential diagnosis of intrahepatic cholangiocarcinoma and hepatocellular carcinoma[J]. Chin J Med Imaging Technol, 2017, 33(7): 993-997. DOI: 10.13929/j.1003-3289.201611140.
18
杭轶, 杨小勇, 李文美. 肝内胆管癌与肝细胞癌临床特征的比较研究[J]. 中国普通外科杂志, 2015, 24(2): 175-179. DOI: CNKI:SUN:ZPWZ.0.2015-02-007.
Hang Y, Yang XY, Li WM. Comparative study of clinical features between intrahepatic cholangiocarcinoma and hepatocellular carcinoma[J]. Chin J Gener Surg, 2015, 24(2): 175-179. DOI: CNKI:SUN:ZPWZ.0.2015-02-007.
19
Wildner D, Bernatik T, Greis C, et al. CEUS in hepatocellular carcinoma and intrahepatic cholangiocellular carcinoma in 320 patients - early or late washout matters: a subanalysis of the DEGUM multicenter trial[J]. Ultraschall Med, 2015, 36(2): 132-9. DOI: 10.1055/s-0034-1399147.
20
张国柄, 徐江海. 联合CA19-9、CA125和CEA检测在AFP阴性的ICC鉴别诊断中的应用价值[J]. 实用肝脏病杂志, 2017, 20(6): 740-743. DOI: CNKI:SUN:GBSY.0.2017-06-025.
Zhang GB, Xu JH. Application of serum CA19- 9 , CA125 and CEA in diagnosis of serum AFP negative patients with intrahepatic cholangiocarcinoma[J]. J Pract Hepatol, 2017, 20(6): 740-743. DOI: CNKI:SUN:GBSY.0.2017-06-025.
21
王悦超. 肝内胆管细胞癌与肝细胞癌临床特征的对比分析[D]. 长春: 吉林大学, 2017.
Wang YC. A comparison and analysis of clinical features of intrahepatic cholangiocarcinoma and hepatocellular carcinoma[D]. Changchun: Jilin Univ, 2017.
22
Zou XL, Luo Y, Li Z, et al. Volumetric apparent diffusion coefficient histogram analysis in differentiating intrahepatic mass-forming cholangiocarcinoma from hepatocellular carcinoma[J]. J Magn Reson Imaging, 2019, 49(4): 975-983. DOI: 10.1002/jmri.26253.
23
Song LR, Yin JD. Application of texture analysis based on sagittal fat-suppression and oblique axial T2-weighted magnetic resonance imaging to identify lymph node invasion status of rectal cancer[J]. Front Oncol, 2020, 10: 1364. DOI: 10.3389/fonc.2020.01364.
24
翟乾, 苏云杉, 岳梨蓉, 等. 兔百草枯中毒CT肺纹理特征与肾灌注的相关性研究[J]. 国际医学放射学杂志, 2020, 43(4): 384-390. DOI: 10.19300/j.2020.L17950.
Zhai Q, Su YS, Yue LR, et al. Correlation between CT lung texture characteristics and renal perfusion in paraquat poisoning rabbit[J]. Int J Med Radiol, 2020, 43(4): 384-390. DOI: 10.19300/j.2020.L17950.
25
方勇超, 王强, 唐权, 等. MSCT与MRI动态增强扫描对高血供肝内胆管癌与肝细胞癌的诊断价值分析[J]. 实用肝脏病杂志, 2020, 23(6): 885-888. DOI: CNKI:SUN:GBSY.0.2020-06-034.
Fang YC, Wang Q, Tang Q, et al. Clinicalvalue of MSCT and MRl dynamic enhanced scans in diagnosing patients with intrahepatic cholangiocarcinoma and hepatocellular carcinoma with high blood supply[J]. J Pract Hepatol, 2020, 23(6): 885-888. DOI: CNKI:SUN:GBSY.0.2020-06-034.

上一篇 构建急性脑卒中机械取栓治疗后预后预测模型的研究
下一篇 基于T2WI的纹理分析在预测肝癌经导管肝动脉化疗栓塞术后早期复发的价值
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2