分享:
分享到微信朋友圈
X
临床研究
不同感兴趣区勾画方法测量集成MRI和DWI定量参数对乳腺良恶性病变的鉴别诊断价值
宋美娜 董磊 何花 孙杰 宋江 高娜 王志军

Cite this article as: Song MN, Dong L, He H, et al. Value of syMRI and DWI quantitative parameters measured using different regions of interest method in differentiating benign and malignant breast lesions[J]. Chin J Magn Reson Imaging, 2022, 13(6): 17-22, 27.本文引用格式:宋美娜, 董磊, 何花, 等. 不同感兴趣区勾画方法测量集成MRI和DWI定量参数对乳腺良恶性病变的鉴别诊断价值[J]. 磁共振成像, 2022, 13(6): 17-22, 27. DOI:10.12015/issn.1674-8034.2022.06.004.


[摘要] 目的 探讨不同感兴趣区(region of interest,ROI)勾画方法测量集成MRI (synthetic MRI,syMRI)和扩散加权成像(diffusion weighted imaging,DWI)定量参数对乳腺良恶性肿块型病变的鉴别诊断价值。材料和方法 纳入122例乳腺MRI检查诊断为肿块型病变的患者,术前均接受常规MRI (T2加权成像、DWI和动态增强扫描)和syMRI检查。两名医师分别采用整体勾画法放置第一个ROI,记作“tumor”;最大层面勾画法放置第二个ROI,记作“max”;在肿瘤强化最明显的区域放置第三个ROI,记作“local”。依据动态增强图像肿瘤强化位置在表观扩散系数(apparent diffusion coefficient,ADC)图像上测量ADC值(ADCtumor、ADCmax和ADClocal)、在T1 mapping图像上测量T1弛豫时间值(T1tumor、T1max和T1local)、在T2 mapping图像上测量T2弛豫时间值(T2tumor、T2max和T2local)、在质子密度(proton density,PD) mapping图像上测量质子密度值(PDtumor、PDmax和PDlocal)。其中医师1间隔一个月后重复上述测量。计算组内相关系数(inter-class correlation coefficient,ICC)评价测量结果的可重复性。比较乳腺良性及恶性肿块型病变间各项参数的差异,利用受试者工作特征(receiver operate characteristic,ROC)曲线下面积(area under the curve,AUC)评估syMRI、DWI及二者联合对良恶性病变的鉴别诊断效能,AUC的比较采用Delong检验。结果 采用不同ROI勾画方法,同一医师两次测量以及不同医师之间测量所测得的ADC值、T1值、T2值及PD值可重复性好(ICC范围0.929~0.992)。三种ROI勾画方法所得ADC值、T2值、PD值在乳腺良恶性病变间差异均具有统计学意义(P值均<0.001)。多因素逻辑回归分析显示ADClocal、T2tumor、PDlocal是预测乳腺癌的独立影响因素,比值比(odds ratio,OR)分别为0.001、0.917、1.267,P值分别为0.013、0.039、0.043。ROC曲线分析显示ADClocal+T2tumor+PDlocal鉴别诊断乳腺良恶性病变的AUC最大(0.953),敏感度为95.2%、特异度84.2%、准确度为91.0%、阳性预测值为93.0%、阴性预测值为88.8%。ADClocal+T2tumor+PDlocal与ADClocal的诊断效能差异无统计学意义(AUC分别为0.953、0.942,P=0.143)。结论 syMRI、DWI定量参数对鉴别乳腺良恶性肿块均具有一定诊断价值,联合诊断模型效能与ADClocal诊断效能相当。
[Abstract] Objective To investigate the value of synthetic magnetic resonance imaging (syMRI) and diffusion weighted imaging (DWI) quantitative parameters measured using different regions of interest (ROI) method in differentiating benign and malignant breast mass-like lesions.Materials and Methods All patients underwent MRI scanning, including T2WI, dynamic contrast-enhanced MRI, DWI and syMRI. Two readers used the method of holistic drawing of lesions to outline ROI respectively, the ROI was drawn along the edge and recorded as "tumor". At the maximum slice of the lesion, the ROI was drawn along the edge and recorded as "max". In the solid area with the most obvious tumor enhancement, the ROI was drawn and recorded as "local". At the same time, apparent diffusion coefficient (ADC) values (ADCtumor, ADCmax and ADClocal) were measured in the ADC map according to the tumor enhancement position in the dynamic enhanced image. T1 relaxation time (T1tumor, T1max and T1local) were measured on T1 mapping images. T2 relaxation time (T2tumor, T2max and T2local) were measured on T2 mapping images. Proton density (PDtumor, PDmax and PDlocal) were measured on PD mapping images. The reader 1 repeated the above measurement after an interval of one month. The inter-class correlation coefficient (ICC) was calculated to evaluate the repeatability of the results measured by different physicians using different ROI delineation methods. Comparison of the differences of various parameters between benign and malignant breast lesions. The receiver operating characteristic (ROC) area under the curve (AUC) was used to evaluate the differential diagnosis performance of syMRI, DWI and their combination in benign and malignant lesions. Delong test was used to compare AUC.Results Using different ROI delineation methods, the ADC value, T1 value, T2 value and PD value measured by the same physician twice and between different physicians were reproducible (ICC range 0.929-0.992). The ADC value, T2 value and PD value obtained by the three ROI delineation methods were all significantly different between benign and malignant breast lesions (all P<0.001). Multivariate logistic regression analysis showed that ADClocal, T2tumor and PDlocal were independent variables in the diagnosis of breast cancer. The OR values were 0.001, 0.917 and 1.267, respectively (P=0.013, 0.039 and 0.043). ROC curve analysis showed that ADClocal+T2tumor+PDlocal had the highest AUC for differential diagnosis of benign and malignant breast lesions (0.953), with a sensitivity of 95.2%, a specificity of 84.2%, an accuracy of 91.0%, a positive predictive value of 93.0%, and a negative predictive value of 88.8%. There was no significant difference in diagnostic efficiency between ADClocal+T2tumor+PDlocal and ADClocal (AUC=0.953, 0.942; P=0.143).Conclusions For breast mass lesions, syMRI and DWI parameters are helpful to differentiating malignancy from benign lesions, the diagnostic performance of combined parameters model was comparable to that of the ADC value.
[关键词] 乳腺肿瘤;磁共振成像;集成磁共振成像;扩散加权成像;感兴趣区
[Keywords] breast neoplasms;magnetic resonance imaging;synthetic magnetic resonance imaging;diffusion weighted imaging;region of interest

宋美娜 1   董磊 2   何花 2   孙杰 1   宋江 1   高娜 1   王志军 2*  

1 宁夏医科大学临床医学院,银川 750004

2 宁夏医科大学总医院放射科,银川 750004

王志军,E-mail:wangzhijun2056@163.com

作者利益冲突声明:全体作者均声明无利益冲突。


基金项目: 国家自然科学基金 81860538
收稿日期:2022-01-20
接受日期:2022-05-31
中图分类号:R445.2  R737.9 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2022.06.004
本文引用格式:宋美娜, 董磊, 何花, 等. 不同感兴趣区勾画方法测量集成MRI和DWI定量参数对乳腺良恶性病变的鉴别诊断价值[J]. 磁共振成像, 2022, 13(6): 17-22, 27. DOI:10.12015/issn.1674-8034.2022.06.004.

       乳腺癌不仅是女性最常见的恶性肿瘤,也是女性癌症致死的主要原因,近年来其发病率在我国呈持续上升趋势[1, 2],通过医学影像学方法对乳腺病变的良恶性进行准确鉴别对于该类疾病的早期诊断与及时治疗有着十分重要的意义。MRI技术具有多方位、多序列成像以及软组织分辨率高等优点,已成为乳腺肿瘤的常用影像学检查手段。临床上,乳腺MRI通常是以病变形态学和血流动力学特征为基础对疾病进行诊断及鉴别诊断[3]。然而,这种评估方式往往存在主观性强及诊断特异度低的缺点。虽然基于扩散加权成像(diffusion weighted imaging,DWI)图像计算所得的定量参数——表观扩散系数(apparent diffusion coefficient,ADC)能够在一定程度上弥补传统MRI形态学诊断特异度较低的缺陷,但不同文献中报道的ADC值的诊断敏感度和特异度尚存在较大差异[4, 5]

       集成MRI (synthetic MRI,syMRI)是一种新型的MRI技术[6],其基于多动态多回波-快速自旋回波(multi-dynamic multi-echo-fast spin echo,MDME-FSE)序列,通过一次扫描可同时对T1弛豫时间(longitudinal relaxation time,T1)、T2弛豫时间(transverse relaxation time,T2)和它们的反弛豫率R1 (1/T1)、R2 (1/T2)及质子密度(proton density,PD)进行量化,这些参数是评估组织特性的绝对定量指标,能够客观、定量地反映组织内部微观特征[7, 8]。其中,T1、T2及PD的相关研究最为广泛。既往研究表明syMRI定量图谱较传统的加权成像,具备更好的敏感度和特异度[9, 10],定量参数如T1、T2、PD值等具有较高的准确性和可重复性[11]。syMRI已应用于包括中枢神经[12]、骨骼[13, 14]、乳腺[15]、盆腹[16, 17]等多个部位的研究中。有研究显示syMRI可用于乳腺良恶性肿瘤的检出以及免疫组化指标和分子分型的预测[18, 19]。乳腺癌是一种高度异质性疾病,感兴趣区(region of interest,ROI)选择方法的不同可能影响定量参数的测量结果。本研究旨探讨不同ROI勾画方法测量syMRI及DWI参数对乳腺良恶性肿块型病变的鉴别诊断价值,希望能够为乳腺疾病的诊断提供新的线索。

1 资料与方法

1.1 一般资料

       本前瞻性研究通过宁夏医科大学总医院伦理委员会批准(批准文号:2020-657),全部受试者均签署知情同意书。纳入2020年7月至2021年12月临床怀疑有乳腺病变的患者资料。纳入标准:(1)所有患者均为肿块型病变并MRI检查后经穿刺活检或手术取得病理结果;(2)术前影像学检查包括:常规序列、DWI序列、超快速动态增强成像序列[(GE;基于笛卡尔排序的差分二次采样(differential subsampling with Cartesian ordering,DISCO)]、syMRI序列扫描,图像资料完整;(3) MRI检查前未做过手术、放化疗及激素治疗等任何治疗。排除标准:(1) MRI检查图像显示非肿块型强化病灶;(2)影像学及病理资料不全;(3)图像质量不佳(如伪影)导致无法勾画ROI者;(4)同侧乳房同时存在良性和恶性病变者。勾画ROI时,对于一侧乳房存在多个病灶的患者,选择病灶中最大者进行勾画及定量参数测量。

1.2 仪器及方法

       采用GE公司SIGNATM Architect 3.0 T磁共振扫描机,配套乳腺专用相控阵线圈。患者俯卧,双乳腺自然悬垂于线圈圆孔内,双臂置于头部两侧,足先进。具体参数:(1)常规扫描T1WI (TR 567 ms,TE 7 ms)及T2WI (TR 4812 ms,TE 92 ms);(2) DWI使用单次激发平面回波成像序列,b=800 mm2/s,TR 3210 ms,TE 73 ms,层厚4 mm,层间5.0 mm,矩阵 128×140,FOV 320 mm×320 mm,扫描时间为2 min 5 s;(3) syMRI序列使用MDME-FSE序列,TR 4378 ms,TE 19 ms,层厚5.0 mm,层间距5.0 mm,FOV 320 mm×320 mm,扫描时相总时间为4 min 32 s;(4) DISCO动态增强扫描序列TR 12 ms,TE 4 ms,层厚1 mm,层间距0 mm,FOV 320 mm×320 mm,对比剂注射前扫描一个蒙片,蒙片扫描结束后注射对比剂,对比剂注射速率为2.5 mL/s,注射完毕后用等量生理盐水冲管,扫描时相总时间为6 min 25 s。

1.3 图像处理及分析

       采用GE AW 4.7工作站READ Y View 软件对ADC值进行测量,在GE主机利用syMRI自动生成T1 mapping、T2 mapping和PD mapping图像,并分别测量T1、T2和PD值。由2名分别在乳腺影像诊断方面具有10年和8年诊断经验放射科医师采取双盲法独立进行图像分析及测量,其中医师1间隔一个月后重复上述测量。进行ROI勾画时,结合T1WI、T2WI、DWI和DISCO增强图像确定病灶的边界和范围,分别采用以下方法:(1)整体勾画法(在整个病灶每个层面沿边缘勾画ROI),记作“tumor”;(2)最大层面勾画法(在病变显示最大层面采用肿瘤轮廓法勾画第2个ROI),记作“max”,ROI尽量避开病灶出血、坏死、囊变区;(3)肿瘤强化最明显处勾画法(对照DISCO增强图像肿瘤强化最明显区域在定量图谱相应位置设置第3个ROI),记作“local”(根据肿瘤大小不同,ROI的范围为1.0~2.5 cm2)。

1.4 统计学分析

       用SPSS 26.0软件进行统计学分析。符合正态分布的计量资料采用均数±标准差进行描述,非正态分布的计量资料以M (Q1,Q3)表示。用ICC评价运用不同ROI勾画方法对不同医师间和同一医师两次测量结果的可重复性。正态分布资料采用t检验,非正态分布资料采用Mann-Whitney U检验比较乳腺肿块型良恶性病变间各参数的差异,将每种ROI勾画方法syMRI差异有统计学意义的定量参数采用多因素逻辑回归建立联合诊断模型。绘制受试者工作特征曲线(receiver operate characteristic,ROC),使用ROC曲线下面积(area under the curve,AUC)来评估各参数及联合诊断模型对乳腺良恶性病变的鉴别诊断效能,计算各参数及联合诊断模型的敏感度、特异度、准确度、阳性预测值及阴性预测值。运用Delong检验比较不同模型间AUC的差异。当P<0.05时认为差异有统计学意义。

2 结果

2.1 临床资料

       122例患者均为肿块型病变,年龄22~80 (50±13)岁,其中84例恶性病灶,包括非特殊浸润性癌70例,导管原位癌7例,乳腺粘液癌1例,乳腺导管内乳头状癌1例,其他恶性肿瘤5例(穿刺病理结果为浸润性癌,未细分类型);38例良性病灶,包括纤维腺瘤26例,导管内乳头状瘤6例,炎性病变6例。

2.2 观察者内及观察者间三种ROI勾画方法的重复性

       医师1前后两次分别采用三种ROI勾画方法测量定量参数,ADCtumor值、ADCmax值、ADClocal值、T1tumor值、T1max值、T1local值、T2tumor值、T2max值、T2local值、PDtumor值、PDmax值和PDlocal值在乳腺良恶性病变中采取三种ROI测量方法具有良好的观察内一致性,ICC分别为0.989、0.990、0.973、0.796、0.981、0.985、0.975、0.960、0.942、0.981、0.985、0.979。医师1与医师2分别采用三同ROI勾画方法测量上述定量参数,三种ROI测量方法均具有良好的观察间一致性,ICC分别为0.990、0.992、0.973、0.982、0.989、0.987、0.945、0.958、0.940、0.981、0.929、0.980。

2.3 三种ROI勾画方法所得定量参数对乳腺肿块型病变的鉴别诊断价值

       syMRI定量参数图像测得恶性肿块型病变T2tumor值、T2max值、T2local值、PDtumor值、PDmax值和PDlocal值均低于良性肿块型病变的相应定量参数,差异具有统计学意义(P值均<0.001) (图12)。乳腺良恶性肿块型病变间T1tumor值、T1max值、T1local值差异均无统计学意义(P值分别为0.532,0.842,0.565)。DWI三种ROI勾画方法所测得恶性肿块型病变ADCtumor值、ADCmax值、ADClocal值均低于良性肿块型病变的相应定量参数,差异具有统计学意义(P值均<0.001) (表1),ROC曲线结果显示单个参数鉴别诊断乳腺良恶性肿块型病变中ADClocal值的AUC值最大(0.942),敏感度为91.7%、特异度为89.5%、准确度为91.0%、阳性预测值为95.1%、阴性预测值为82.9%。

       多因素逻辑回归分析显示ADClocal、T2tumor、PDlocal是预测乳腺癌的独立影响因素,OR值分别为0.001、0.917、1.267,P分别为0.013、0.039、0.043 (表2)。ROC曲线分析显示ADClocal+T2tumor+PDlocal鉴别诊断乳腺良恶性病变的AUC最大(0.953),敏感度为95.2%、特异度为84.2%、准确度为91.0%、阳性预测值为93.0%、阴性预测值为88.8% (表3)。ADClocal+T2tumor+PDlocal、T2tumor+PDlocal及ADClocal的诊断乳腺良恶性病变的ROC曲线见图3。ADClocal+T2tumor+PDlocal与T2tumor+PDlocal,ADClocal与T2tumor+PDlocal的AUC差异均有统计学意义(P值分别0.003、0.023)。ADClocal+T2tumor+PDlocal与ADClocal的AUC差异无统计学意义(P值为0.143)。

图1  女,31岁,右乳纤维腺瘤。1A:超快速动态增强成像序列(DISCO)动态增强(第3期)图像;1B:表观扩散系数伪彩图;1C:集成磁共振成像(syMRI) T2WI图像;1D:SyMRI T1-mapping图像;1E:SyMRI T2-mapping图像;1F:SyMRI PD-mapping图像。
图2  女,53岁,右乳腺非特殊型浸润性癌。2A:超快速动态增强成像序列(DISCO)动态增强(第3期)图像;2B:表观扩散系数伪彩图;2C:集成磁共振成像(syMRI) T2WI图像;2D:syMRI T1-mapping图像;2E:syMRI T2-mapping图像;2F:syMRI PD-mapping图像。
Fig. 1  Female, 31 years old, fibroadenoma of right breast. 1A: MRI dynamic enhancement (Phase 3 image); 1B: apparent diffusion coefficient (ADC)-mapping; 1C: synthetic magnetic resonance imaging (SyMRI) T2WI; 1D: SyMRI T1-mapping; 1E: SyMRI T2-mapping; 1F: SyMRI PD-mapping.
Fig. 2  Female, 53 years old, invasive ductal carcinoma of the right breast. 2A: MRI dynamic enhancement (Phase 3 image); 2B: apparent diffusion coefficient (ADC)-mapping; 2C: synthetic magnetic resonance imaging (SyMRI) T2WI; 2D: SyMRI T1-mapping; 2E: SyMRI T2-mapping; 2F: SyMRI PD-mapping.
图3  不同感兴趣区(ROI)勾画方法所得MRI定量参数诊断乳腺良恶性病变的受试者工作特征(ROC)曲线。PD为质子密度;ADC为表观扩散系数;tumor代表整体勾画法;max代表最大层面勾画法;local代表强化最明显处勾画法。
Fig.3  The receiver operate characteristic (ROC) curves of MRI quantitative parameters obtained by different region of interest (ROI) outlined methods in the diagnosis of benign and malignant breast lesions. PD is proton density; ADC is apparent diffusion coefficient; tumor stands for global sketching; Max stands for maximum level sketching; Local stands for enhanced delineation of the most obvious.
表1  乳腺良恶性病变间各定量参数比较
Tab. 1  Comparison of quantitative parameters between benign and malignant breast masses
表2  MRI定量参数多因素logistic回归分析
Tab. 2  Multivariate logistic regression analysis of MRI quantitative parameters
表3  MRI定量参数及联合诊断模型对乳腺肿块型病变的鉴别诊断价值
Tab. 3  The diagnostic value of MRI quantitative parameters and combined model for differentiating

3 讨论

       乳腺癌早期诊断和治疗有助于延长患者的生存期,提高生存质量,对疾病的发展和预后具有重要的意义。本研究首次采用3种不同的ROI勾画方法(整体勾画法、最大层面勾画法和强化最明显处勾画法)测量T1、T2、PD及ADC值,研究结果显示syMRI、DWI定量参数对鉴别乳腺良恶性肿块均具有一定诊断价值,多参数联合诊断模型(ADClocal+T2tumor+PDlocal)鉴别诊断乳腺良恶性病变的诊断效能最高,但与ADClocal的诊断效能相当。

       MRI中不同组织弛豫时间的差异与组织自由水含量、水分子和大分子的随机运动、组织脂肪含量及顺磁性物质的存在等因素有关[20, 21]。其中自由水的含量对组织弛豫时间的影响最为明显,有研究[22]表明自由水含量越多,T2弛豫时间越长。本研究结果显示,无论采用哪种ROI勾画方法,syMRI乳腺良性病变的T2值均显著高于恶性病变,这可能是由于恶性病灶癌细胞增殖快,细胞密集,核浆比例增大,导致细胞外间隙缩小以及自由水含量减少,而部分细胞因供氧量不足发生坏死,坏死物质及炎性细胞在细胞间隙浸润,进一步加剧了细胞外间隙缩小的程度,使得自由水含量更少,以上原因可能是使乳腺恶性病变T2值较良性病变的T2值更短的主要原因[23]。此外,本研究结果显示乳腺良性病变的PD值显著高于恶性病变的PD值。车树楠等[24]报道PD值在乳腺良恶性组间差异无统计学意义,推测其原因可能是由于纳入恶性组的病变中包含了非肿块型强化的病灶,测量的ROI内含有少量正常的乳腺组织,而乳腺组织含有大量的脂肪成分,脂肪和水同样含有大量的质子,从而增加了恶性病变的PD值,从而造成两组间PD值差异无统计学意义。Matsuda等[25]研究显示乳腺恶性病变的T1值显著高于良性病变。本研究显示尽管恶性病变与良性病变相比T1值较高,但两者间差异无统计学意义(P>0.05),与Gao等[26]研究结果一致。分析原因为可能为T1值并不是乳腺良恶性鉴别最敏感的指标,同时它受多种因素的影响,如样本量的大小、病理类型、磁场强度等,需要扩大样本量和多中心进一步研究。

       DWI可显示体内水分子的随机运动,与良性肿瘤相比,恶性肿瘤通常表现为更加显著的水分子扩散受限。ADC值作为客观定量评估方式,已被广泛作为恶性肿瘤的生物学标记物,为乳腺癌患者病灶内水分子自由扩散受限程度的评估提供定量化的信息[27]。本研究ADClocal的ROI勾画是根据增强图像中肿瘤强化最明显的区域放置的,既往研究[28]结果均显示ADClocal较其他单一参数具有更高的诊断效能,分析原因可能为肿瘤内强化最明显的部位反映了肿瘤组织细胞最密集、自由水含量较少的区域,有研究发现肿瘤ADC值较低的区域与肿瘤组织密集区有较好的相关性[29]。因此,ADClocal能够较好地区分乳腺肿块型病变的良恶性。

       syMRI各参数的AUC显示使用单一参数诊断良恶性疾病的效能和特异度较低,分析原因为syMRI各良恶性病变定量参数存在一定程度的重叠,不能全面反映乳腺病变的组织微环境特征,单独应用可能存在一定的局限性,因此可通过运用多种成像技术及联合多种参数来提供更加全面、精准的诊断信息。本研究中多因素逻辑回归分析显示ADClocal、T2tumor、PDlocal是诊断乳腺恶性病变的独立影响因素。考虑到多参数联合可以充分利用各技术的优势,反映更全面的肿瘤信息,因此本研究分别建立T2tumor+PDlocal和ADClocal+T2tumor+PDlocal模型,并与其他参数比较诊断效能,结果显示ADClocal+T2tumor+PDlocal模型的AUC最高(0.953),且拥有较高的敏感度(95.2%)和特异度(84.2%),其AUC与ADClocal (AUC=0.942)相比差异无统计学意义(P=0.143),提示二者的诊断效能相当。然而,与DWI相比,syMRI可同时提取多种定量参数,从多个维度全面反映乳腺肿瘤的微观结构变化,从而可用于乳腺良恶性病变的鉴别诊断。徐良洲等[30]利用syMRI测量的8个特征性脑区的T1及T2值有很好的稳定性及可重复性,本研究也显示由syMRI提供的定量弛豫时间和ADC值在乳腺组织中均具有良好的不同观察者间和同一观察者内可重复性,表明syMRI定量参数可作为长期随访研究或多中心研究的有效指标。因此,syMRI在临床中具有广阔的应用前景。

       本研究的局限性:(1)由于非肿块型强化病变部位可能混杂有正常的腺体组织,ROI勾画较困难且病例数较少,因此没有纳入研究进行分析;(2)未对增强扫描前后弛豫参数的变化进行评估,将在后续的研究中加以补充;(3)本研究为单中心研究,需要多中心研究进一步验证。

       综上所述,本研究显示通过不同ROI勾画方法对乳腺syMRI定量参数进行测量具有良好的可重复性,syMRI及DWI定量参数对鉴别乳腺良恶性肿块型病变均具有较高的诊断价值。

[1]
Feng RM, Zong YN, Cao SM, et al. Current cancer situation in China: good or bad news from the 2018 Global Cancer Statistics?[J]. Cancer Commun (Lond), 2019, 39(1): 1-12. DOI: 10.1186/s40880-019-0368-6.
[2]
Fahad UM. Breast cancer: current perspectives on the disease status[J]. Adv Exp Med Biol, 2019, 1152: 51-64. DOI: 10.1007/978-3-030-20301-6_4.
[3]
Spak DA, Plaxco JS, Santiago L, et al. BI-RADS ® fifth edition: a summary of changes[J]. Diagn Interv Imaging, 2017, 98(3): 179-190. DOI: 10.1016/j.diii.2017.01.001.
[4]
Rahbar H, Partridge SC, Demartini WB, et al. In vivo assessment of ductal carcinoma in situ grade: a model incorporating dynamic contrast-enhanced and diffusion-weighted breast MR imaging parameters[J]. Radiology, 2012, 263(2): 374-382. DOI: 10.1148/radiol.12111368.
[5]
Iima M, Kataoka M, Kanao S, et al. Intravoxel incoherent motion and quantitative non-Gaussian diffusion MR imaging: evaluation of the diagnostic and prognostic value of several markers of malignant and benign breast lesions[J]. Radiology, 2018, 287(2): 432-441. DOI: 10.1148/radiol.2017162853.
[6]
Hagiwara A, Warntjes M, Hori M, et al. SyMRI of the brain: rapid quantification of relaxation rates and proton density, with synthetic MRI, automatic brain segmentation, and myelin measurement[J]. Invest Radiol, 2017, 52(10): 647-657. DOI: 10.1097/RLI.0000000000000365.
[7]
李春媚, 陈敏. 积极推进合成MRI的临床应用 提升MRI诊断效能[J]. 中华放射学杂志, 2021, 55(6): 577-579. DOI: 10.3760/cma.j.cn112149-20210415-00370.
Li CM, Chen M. Promoting actively the clinical application of synthetic MRI, improving the diagnostic efficiency of MRI[J]. Chin J Radiol, 2021, 55(6): 577-579. DOI: 10.3760/cma.j.cn112149-20210415-00370.
[8]
Granberg T, Uppman M, Hashim F, et al. Clinical feasibility of synthetic MRI in multiple sclerosis: a diagnostic and volumetric validation study[J]. AJNR Am J Neuroradiol, 2016, 37(6): 1023-1029. DOI: 10.3174/ajnr.A4665.
[9]
Kumar NM, Fritz B, Stern SE, et al. Synthetic MRI of the knee: phantom validation and comparison with conventional MRI[J]. Radiology, 2018, 289(2): 465-477. DOI: 10.1148/radiol.2018173007.
[10]
Yi J, Lee YH, Song HT, et al. Clinical feasibility of synthetic magnetic resonance imaging in the diagnosis of internal derangements of the knee[J]. Korean J Radiol, 2018, 19(2): 311-319. DOI: 10.3348/kjr.2018.19.2.311.
[11]
Krauss W, Gunnarsson M, Andersson T, et al. Accuracy and reproducibility of a quantitative magnetic resonance imaging method for concurrent measurements of tissue relaxation times and proton density[J]. Magn Reson Imaging, 2015, 33(5): 584-591. DOI: 10.1016/j.mri.2015.02.013.
[12]
Lee SM, Choi YH, Cheon JE, et al. Image quality at synthetic brain magnetic resonance imaging in children[J]. Pediatr Radiol, 2017, 47(12): 1638-1647. DOI: 10.1007/s00247-017-3913-y.
[13]
Arita Y, Takahara T, Yoshida S, et al. Quantitative assessment of bone metastasis in prostate cancer using synthetic magnetic resonance imaging[J]. Invest Radiol, 2019, 54(10): 638-644. DOI: 10.1097/RLI.0000000000000579.
[14]
Roux M, Hilbert T, Hussami M, et al. MRI T2 mapping of the knee providing synthetic morphologic images: comparison to conventional Turbo spin-echo MRI[J]. Radiology, 2019, 293(3): 620-630. DOI: 10.1148/radiol.2019182843.
[15]
Jung Y, Gho SM, Back SN, et al. The feasibility of synthetic MRI in breast cancer patients: comparison of T2 relaxation time with multiecho spin echo T2 mapping method[J/OL]. Br J Radiol, 2018, 92 [2022-01-20]. https://www.birpublications.org/doi/10.1259/bjr.20180479. DOI: 10.1259/bjr.20180479.
[16]
Zhao L, Liang M, Wu PY, et al. A preliminary study of synthetic magnetic resonance imaging in rectal cancer: imaging quality and preoperative assessment[J]. Insights Imaging, 2021, 12(1): 120. DOI: 10.1186/s13244-021-01063-w.
[17]
Zhao L, Liang M, Wang SC, et al. Preoperative evaluation of extramural venous invasion in rectal cancer using radiomics analysis of relaxation maps from synthetic MRI[J]. Abdom Radiol (NY), 2021, 46(8): 3815-3825. DOI: 10.1007/s00261-021-03021-y.
[18]
高微波, 杨全新, 陈欣, 等. 合成MRI对乳腺良恶性病变鉴别诊断的价值[J]. 中华放射学杂志, 2021, 55(6): 605-608. DOI: 10.3760/cma.j.cn112149-20200831-01043.
Gao WB, Yang QX, Chen X, et al. The value of synthetic MRI in the differential diagnosis of benign and malignant breast lesions[J]. Chin J Radiol, 2021, 55(6): 605-608. DOI: 10.3760/cma.j.cn112149-20200831-01043.
[19]
Matsuda M, Tsuda T, Ebihara R, et al. Triple-negative breast cancer on contrast-enhanced MRI and synthetic MRI: a comparison with non-triple-negative breast carcinoma[J/OL]. Eur J Radiol, 2021, 142 [2022-01-20]. https://www.ejradiology.com/article/S0720-048X(21)00319-3/fulltext. DOI: 10.1016/j.ejrad.2021.109838.
[20]
Liu L, Yin B, Geng DY, et al. Changes of T2 relaxation time from neoadjuvant chemotherapy in breast cancer lesions[J/OL]. Iran J Radiol, 2016, 13(3)[2022-01-20]. https://brieflands.com/articles/iranjradiol-18060.html. DOI: 10.5812/iranjradiol.24014.
[21]
徐晶, 赵世华, 陆敏杰. 心脏T2定量成像技术及其临床应用研究进展[J]. 中华放射学杂志, 2020(11): 1132-1136.
Xu J, Zhao SH, Lu MJ. Research advances in imaging techniques and clinical applications of myocardial T2 mapping[J]. Chin J Radiol, 2020(11): 1132-1136.
[22]
Chaland B, Mariette F, Marchal P, et al. 1H nuclear magnetic resonance relaxometric characterization of fat and water states in soft and hard cheese[J]. J Dairy Res, 2000, 67(4): 609-618. DOI: 10.1017/s0022029900004398.
[23]
Liu L, Yin B, Shek K, et al. Role of quantitative analysis of T2 relaxation time in differentiating benign from malignant breast lesions[J]. J Int Med Res, 2018, 46(5): 1928-1935. DOI: 10.1177/0300060517721071.
[24]
车树楠, 李静, 薛梅, 等. 集成磁共振成像对乳腺良恶性病变的鉴别诊断价值[J]. 中华肿瘤杂志, 2021, 43(8): 872-877. DOI: 10.3760/cma.j.cn112152-20210322-00254.
Che SN, Li J, Xue M, et al. The value of synthetic MRI in differential diagnosis of benign and malignant breast lesions[J]. Chin J Oncol, 2021, 43(8): 872-877. DOI: 10.3760/cma.j.cn112152-20210322-00254.
[25]
Matsuda M, Tsuda T, Ebihara R, et al. Enhanced masses on contrast-enhanced breast: differentiation using a combination of dynamic contrast-enhanced MRI and quantitative evaluation with synthetic MRI[J]. J Magn Reson Imaging, 2021, 53(2): 381-391. DOI: 10.1002/jmri.27362.
[26]
Gao WB, Zhang SQ, Guo JX, et al. Investigation of synthetic relaxometry and diffusion measures in the differentiation of benign and malignant breast lesions as compared to BI-RADS[J]. J Magn Reson Imaging, 2021, 53(4): 1118-1127. DOI: 10.1002/jmri.27435.
[27]
卢冬梅, 王佳美, 刘瑜琳, 等. 乳腺病变DWI不同指数模型的研究进展[J]. 磁共振成像, 2018, 9(4): 308-311. DOI: 10.12015/issn.1674-8034.2018.04.014.
Lu DM, Wang JM, Liu YL, et al. Progress in the application of different DWI exponential models for breast lesions[J]. Chin J Magn Reson Imaging, 2018, 9(4): 308-311. DOI: 10.12015/issn.1674-8034.2018.04.014.
[28]
孙诗昀, 李卓琳, 聂莉莎, 等. 合成MRI联合扩散加权成像对乳腺良恶性病变的鉴别诊断价值[J]. 中华放射学杂志, 2021, 55(6): 597-604. DOI: 10.3760/cma.j.cn112149-20200717-00927.
Sun SY, Li ZL, Nie LS, et al. The value of synthetic MRI combined with diffusion weighted imaging in differential diagnosis of benign and malignant breast lesions[J]. Chin J Radiol, 2021, 55(6): 597-604. DOI: 10.3760/cma.j.cn112149-20200717-00927.
[29]
Lu SS, Kim SJ, Kim N, et al. Histogram analysis of apparent diffusion coefficient maps for differentiating primary CNS lymphomas from tumefactive demyelinating lesions[J]. AJR Am J Roentgenol, 2015, 204(4): 827-834. DOI: 10.2214/AJR.14.12677.
[30]
徐良洲, 徐霖, 贺梦吟, 等. 集成MR序列T1、T2弛豫定量的可重复性研究[J]. 放射学实践, 2019, 34(11): 1178-1181. DOI: 10.13609/j.cnki.1000-0313.2019.11.001.
Xu LZ, Xu L, He MY, et al. Reproducibility of quantitative relaxation study of synthetic MRI[J]. Radiol Pract, 2019, 34(11): 1178-1181. DOI: 10.13609/j.cnki.1000-0313.2019.11.001.

上一篇 MRI多序列模型融合影像组学预测局部晚期鼻咽癌患者同步放化疗疗效的价值
下一篇 T2 mapping预测直肠癌脉管侵犯的初步研究
  
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