分享:
分享到微信朋友圈
X
技术研究
3.0 T磁共振脑部快速动脉自旋标记成像的可行性研究
杨永贵 陈芳 吴秀芬 林惠芳 延根

Cite this article as: YANG Y G, CHEN F, WU X F, et al. Feasibility study of fast arterial spin labeling imaging in brain on 3.0 T MRI scanner[J]. Chin J Magn Reson Imaging, 2023, 14(1): 116-123.本文引用格式:杨永贵, 陈芳, 吴秀芬, 等. 3.0 T磁共振脑部快速动脉自旋标记成像的可行性研究[J]. 磁共振成像, 2023, 14(1): 116-123. DOI:10.12015/issn.1674-8034.2023.01.021.


[摘要] 目的 探讨3.0 T磁共振脑部快速动脉自选标记(fast arterial spin labeling, Fast ASL)成像的可行性。材料与方法 2021年12月至2022年8月在我院前瞻性采用对比分组分析法和随机分组分析法两种方法进行受检者分组、数据收集与分析。采用3.0 T临床科研型磁共振仪,16通道头部相控阵线圈进行受检者的Fast ASL/ASL成像和常规磁共振序列[T1WI、T2WI、T2压脂液体抑制反转恢复(T2 fat saturation and fluid attenuated inversion recovery, T2 fs FLAIR)]、功能与血供序列[弥散加权成像(diffusion weighted imaging, DWI)、增强灌注加权成像(perfusion weighted imaging, PWI)、磁共振血管成像(magnetic resonance angiography, MRA)]的数据采集。对比分组分析法组25例受检者分别进行两种采集方案(Fast ASL和ASL)的图像采集并进行后处理重建分析。随机分组分析法组200例受检者,按照检查时间顺序交替进行随机分组各100例,进行单种采集方案(Fast ASL或ASL)的图像采集并进行后处理重建分析。采用GE AW 4.6后处理工作站functool软件包的3D ASL进行后处理,并对半定量脑血流量(cerebral blood flow, CBF)图进行图像质量评分及结果评价,采用SPSS 26.0进行统计学分析。结果 对比分组分析法组中,Fast ASL组和ASL组的CBF图像质量得分分别为(4.32±0.55)、(4.72±0.54)分;两组数据配对存在相关性,配对样本t检验显示差异具有统计学意义(P<0.001);两组的CBF结果与相应PWI结果均存在相关性(P均<0.001),差异均不具有统计学意义(P=1.00、0.57);Fast ASL组无灌注异常者和灌注异常者、反映灌注情况的敏感度和特异度分别为8例(88.2%)和17例(75.0%),ASL组分别为7例(94.1%)和18例(75.0%),两组间差异不具有统计学意义(P=0.84)。随机分组分析法组中,Fast ASL组和ASL组的CBF图像质量得分分别为(4.30±0.50)、(4.75±0.46)分;两组数据方差不齐(P=0.04),独立样本t检验显示差异具有统计学意义(P<0.001);两组的CBF结果与相应PWI结果均存在相关性(P均<0.001),差异均不具有统计学意义(P=0.44、0.21);Fast ASL组无灌注异常者和灌注异常者、反映灌注情况的敏感度和特异度分别为34例(90.5%)和66例(75.7%),ASL组分别为23例(91.4%)和77例(84.2%),两组间差异不具有统计学意义(P=0.48)。无论是哪一分组方法的结果,Fast ASL组CBF的图像质量得分均略低于ASL组的,反映灌注情况的敏感度和特异度也略有差异,但均能显示并区分灌注异常灶。结论 优化参数后的Fast ASL成像,可获得脑部半定量的CBF图,可以明确区分脑组织正常与异常灌注区域,为临床诊疗工作提供准确的脑部灌注信息,对于疾病的防治具有重要意义。尤其是采集时间可以缩短一半,对于无法耐受的受检者,可以获得相对满意的CBF效果,达到图像质量控制与采集时间的平衡,具有一定的可行性。
[Abstract] Objective To explore the feasibility of Fast Arterial Spin Labeling (Fast ASL) imaging in brain on 3.0 T MRI scanner.Materials and Methods From December 2021 to August 2022, we adopted two grouping methods which are comparative grouping analysis and random grouping analysis, were used for grouping the subjects, data collection and perform prospective analysis in our hospital. The 3.0 T clinical research type MRI scanner and 16 channel head phased array coil were used to conduct fast ASL/ASL imaging of subjects and data acquisition of routine magnetic resonance sequences [T1WI, T2WI, T2 fat saturation and fluid attenuated inversion recovery (T2 fs FLAIR)], function and blood supply sequences [diffusion weighted imaging (DWI), perfusion weighted imaging (PWI), magnetic resonance angiography (MRA)]. Comparative grouping analysis group: the 25 subjects in comparative grouping analysis group were collected with two acquisition schemes (fast ASL and ASL) and analyzed with post-processing reconstruction. A total of 200 subjects in randomized grouping analysis group were randomly divided into 100 cases in turn according to the time sequence of examination, and image acquisition with single acquisition scheme (fast ASL or ASL) and analyzed with post-processing reconstruction. The 3D ASL of functool software package was used for post-processing on the GE AW 4.6 post-processing workstation, and the CBF image was scored and result evaluated for image quality. The statistical analysis was carried out using SPSS 26.0.Results In the comparative grouping analysis group, the CBF image quality scores of Fast ASL group and ASL group were (4.32±0.55) and (4.72±0.54) respectively. The paired data of the two groups were correlated, and the paired sample t-test showed that the difference was statistically significant (P<0.001). The CBF results of the two groups were correlated with the corresponding PWI results (P<0.001), and the differences were not statistically significant (P=1.00, 0.57). In the Fast ASL group, there were 8 cases without abnormal perfusion, 17 cases with abnormal perfusion, and the sensitivity and specificity of reflecting perfusion were 88.2% and 75.0% respectively. In the ASL group, there were 7 cases (94.1%) and 18 cases (75.0%), respectively. There was no significant difference between the two groups (P=0.84). In the randomized grouping analysis group, the CBF image quality scores of Fast ASL group and ASL group were (4.30±0.50) and (4.75±0.46) respectively. The variance of the two groups of data was not uniform (P=0.04), and the independent sample t-test showed that the difference was statistically significant (P<0.001). The CBF results of the two groups were correlated with the corresponding PWI results (P<0.001), and the differences were not statistically significant (P=0.44,0.21). In the Fast ASL group, there were 34 cases without abnormal perfusion, 66 cases with abnormal perfusion, and the sensitivity and specificity of reflecting perfusion were 90.5% and 75.7% respectively, while in the ASL group, there were 91.4% and 84.2%. There was no significant difference between the two groups (P=0.48). No matter which grouping method, the image quality score of CBF in the Fast ASL group was slightly lower than that in the ASL group, and the sensitivity and specificity of reflecting the perfusion situation were also slightly different, but the abnormal perfusion focus could be displayed and differentiated.Conclusions After optimizing the parameters, Fast ASL imaging can obtain a semi quantitative CBF map of the brain, which can clearly distinguish the normal and abnormal brain perfusion areas, provide accurate brain perfusion information for clinical diagnosis and treatment, and is of great significance for disease prevention and treatment. In particular, the acquisition time can be shortened by half. For subjects who cannot tolerate it, Fast ASL can obtain relatively satisfactory CBF effect, which is feasible.
[关键词] 脑梗死;脑部肿瘤;脑血流量;灌注;磁共振成像;动脉自选标记
[Keywords] cerebral infarction;brain tumor;cerebral blood flow;perfusion;magnetic resonance imaging;arterial spin labeling

杨永贵 *   陈芳    吴秀芬    林惠芳    延根   

厦门医学院附属第二医院放射影像一科,厦门 361021

通信作者:杨永贵,E-mail:yangyonggui125@sina.cn

作者贡献声明:杨永贵设计本研究的方案,起草和撰写稿件,获取、分析或解释本研究的数据;陈芳、吴秀芬、林惠芳、延根获取、分析或解释本研究的数据,对稿件重要的智力内容进行了修改;杨永贵获得福建省自然科学基金面上项目基金(编号:2022J011384)和福建省卫生健康中青年骨干人才培养项目基金(2020GGB067)的资助;全体作者都同意最后的修改稿发表,同意对工作的所有方面负责,确保本研究的准确性和诚信。


基金项目: 福建省自然科学基金面上项目 2022J011384 福建省卫生健康中青年骨干人才培养项目 2020GGB067
收稿日期:2022-09-05
接受日期:2022-12-30
中图分类号:R445.2 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2023.01.021
本文引用格式:杨永贵, 陈芳, 吴秀芬, 等. 3.0 T磁共振脑部快速动脉自旋标记成像的可行性研究[J]. 磁共振成像, 2023, 14(1): 116-123. DOI:10.12015/issn.1674-8034.2023.01.021.

0 前言

       临床上,脑梗死、脑部肿瘤等诸多脑部疾病是当前发病率和致死率较高的病种,都有伴随着脑组织血流灌注的改变[1, 2, 3],如何更早更快速地检测脑灌注信息,将为临床防治及预后判定提供更精准的依据,具有重大的临床应用价值。在磁共振的检查中,除了通过有创的、需注射外源性对比剂的灌注加权成像(perfusion weighted imaging, PWI)获得相关指标[2],如何无创、精准地检测灌注相应的指标,是亟须解决且具有重要意义的技术难题。作为临床与科研近年来的热点,动脉自旋标记(arterial spin labeling, ASL)技术可无创性、无需外源性对比剂实现脑组织血流灌注成像,已经在临床诊疗工作中得到广泛应用[1, 2, 3, 4, 5, 6]。现有研究所推荐的ASL采集方案[2,5,7]时间都相对较长,一般一次采集时间在4~6 min,对于需要快速评估脑灌注情况但又无法耐受采集时间长或急性脑梗死的受检者,将无法完成这一检查而得不到相应的精准信息。本研究通过采集方案的优化,进行不同采集参数的优劣势分析,制订快速ASL(Fast ASL)最佳方案,探讨3.0 T磁共振脑部Fast ASL成像的可行性,以达到实现临床无创、快速、准确的脑灌注指标的检测,为脑梗死、脑部肿瘤等诸多脑部疾病提供更早更快速检测脑灌注信息的磁共振技术手段,为预防、诊疗及预后判定工作提供精准的依据,具有重要且实际的临床意义。

1 材料与方法

1.1 一般资料

       2021年12月至2022年8月在我院前瞻性采用对比分组分析法和随机分组分析法两种分组方法,进行受检者分组、数据收集与分析。

       对比分组分析法组:受检者分别进行两种采集方案(Fast ASL和ASL)的图像采集并进行后处理重建分析,共计25例(男11例,女14例),年龄24~75(51.8±16.29)岁。所有受检者均行磁共振常规平扫和/或增强检查,根据其诊断结果有健康者8例、转移瘤11例、颅咽管瘤1例、星形细胞瘤1例、脑梗死4例,无灌注异常者8例、灌注异常者17例。

       随机分组分析法组:受检者200例,按照检查时间顺序交替进行随机分组各100例,进行单种采集方案(Fast ASL或ASL)的图像采集并进行后处理重建分析。

       Fast ASL组:100例(男51例,女49例),年龄14~75(49.21±15.78)岁。所有受检者均行磁共振常规平扫和/或增强检查,根据其诊断结果有健康者37例、转移瘤31例、颅咽管瘤5例、胶质瘤3例、脑梗死24例,无灌注异常者37例、灌注异常者63例。

       ASL组:100例(男52例,女48例),年龄24~75(51.8±15.76)岁。所有受检者均行磁共振常规平扫和/或增强检查,根据其诊断结果有健康者19例、转移瘤43例、颅咽管瘤3例、脑膜瘤4例、脑梗死31例,无灌注异常者19例、灌注异常者81例。

       本研究遵守《赫尔辛基宣言》,得到厦门医学院附属第二医院伦理学委员会批准(批注文号:2021046),所有受检者均在检查前了解了检查内容并签署了知情同意书。

1.2 纳入标准和排除标准

       纳入标准:(1)临床及影像学资料完整;(2)病情允许,并能配合项目所需的磁共振数据采集;(3)所有受检者均被告知项目研究内容,患者或监护人签署知情同意书。排除标准:(1)因头部移动大造成图像质量差影响数据分析;(2)病情等原因无法耐受长时间的数据采集,或明确表示无法配合;(3)磁共振禁忌证者。

1.3 成像方法

       采用3.0 T临床科研型磁共振仪(Discovery MR750w; GE Healthcare, Milwaukee, WI),16通道头部相控阵线圈进行受检者的Fast ASL/ASL成像和常规磁共振序列[T1WI、T2WI、T2压脂液体抑制反转恢复序列(T2 fat saturation and fluid attenuated inversion recovery, T2 fs FLAIR)]、功能与血供序列[弥散加权成像(diffusion weighted imaging, DWI)、增强灌注加权成像(perfusion weighted imaging, PWI)、磁共振血管成像(magnetic resonance angiography, MRA)]的数据采集。

       Fast ASL采集方案:三维(three-dimensional, 3D)ASL序列,标记后延迟时间(post label delay, PLD)1525.0 ms、2025.0 ms、2525.0 ms,激励次数(number of excitations, NEX)2,TR 4466.0 ms、4678.0 ms、5161.0 ms,TE 9.9 ms,带宽 62.50 kHz,FOV 24 cm×24 cm,矩阵 512×512,点数(Points)512,螺旋臂数(Arms)6,层厚6 mm,间距0 mm,层数24层,采集时间2 min 23 s、2 min 30 s、2 min 45 s。

       ASL采集方案:3D ASL序列,PLD 1525.0 ms、2025.0 ms、2525.0 ms,NEX 3,TR 4852.0 ms、4852.0 ms、5335.0 ms,TE 10.7 ms,带宽 62.50 kHz,FOV 24 cm×24 cm,矩阵512×512,Points 512,Arms 8,层厚4 mm,间距 0 mm,层数 36层,采集时间 4 min 29 s、4 min 42 s、5 min 10 s。

1.4 图像质量评分

       参照图像质量评价标准的绝对评价尺度的妨碍尺度[8]制订本组数据评分标准。5分:信噪比(signal noise ratio, SNR)优,对比度优,病灶显示好;4分:SNR良,对比度良,病灶显示良;3分:SNR一般,对比度一般,但不妨碍病灶显示;2分:SNR差,对比度差,略有妨碍病灶显示,但不影响观察;1分:SNR极差,对比度极差,严重妨碍病灶显示。

1.5 数据处理及分析

       所得数据采用GE AW 4.6后处理工作站functool软件包的3D ASL进行后处理,并对所得的半定量脑血流量(cerebral blood flow, CBF)图进行评价分析。采用SPSS 26.0分别进行配对样本t检验和配对相关性分析、独立样本t检验和方差分析、受试者工作特征(receiver operating characteristic, ROC)曲线分析等统计学分析。P<0.05为差异具有统计学意义。

2 结果

       对比分组分析法组中,Fast ASL组和ASL组的CBF图像质量得分分别为(4.32±0.55)、(4.72±0.54)分。该组数据中的两种采集方案所得的CBF的图像质量得分配对存在相关性(P<0.001),配对样本t检验显示差异具有统计学意义(P<0.001),见表1

       Fast ASL组和ASL组CBF结果分别与相应PWI结果配对均存在相关性(P均<0.001),且配对样本t检验显示分别与PWI的差异均不具有统计学意义(P=1.00、0.57);Fast ASL组无灌注异常者和灌注异常者、反映灌注情况的敏感度和特异度分别为8例(88.2%)和17例(75.0%),ASL组分别为7例(94.1%)和18例(75.0%),两组间差异不具有统计学意义(P=0.84),见表2图1

       随机分组分析法组中,Fast ASL组和ASL组的CBF图像质量得分分别为(4.30±0.50)、(4.75±0.46)分。该组数据中的两种采集方案所得的CBF的图像质量得分方差不齐(Levene's Test,F=4.07,P=0.04),独立样本t检验显示差异具有统计学意义(P<0.001),见表3

       Fast ASL组和ASL组CBF结果分别与相应PWI结果配对均存在相关性(P均<0.001),且配对样本t检验显示分别与PWI的差异均不具有统计学意义(P=0.44、0.21)。

       Fast ASL组无灌注异常者和灌注异常者、反映灌注情况的敏感度和特异度分别为34例(90.5%)和66例(75.7%),ASL组分别为23例(91.4%)和77例(84.2%),两组间差异不具有统计学意义(P=0.48),见表4图1

       无论是哪一分组方法的结果,Fast ASL组CBF的图像质量得分均略低于ASL组,反映灌注情况的敏感度和特异度也略有差异,但均能显示并区分灌注异常灶。

图1  受试者工作特征曲线分析。1A:对比分组分析法组动脉自选标记(ASL)组;1B:对比分组分析法组快速动脉自选标记(Fast ASL)组;1C:随机分组分析法组ASL组;1D:随机分组分析法组Fast ASL组。
Fig. 1  Analysis of receiver operating characteristic curves. 1A: comparative grouping analysis group arterial spin labeling (ASL) group; 1B: comparative grouping analysis group Fast ASL group; 1C: randomized grouping analysis group ASL group; 1D: randomized grouping analysis group Fast ASL group.
表1  对比分组分析法组图像质量分值配对样本t检验
Tab. 1  Paired sample t-test for image quality score of the comparative grouping analysis group
表2  对比分组分析法组ROC曲线分析
Tab. 2  Analysis of ROC curve of the comparative grouping analysis group
表3  随机分组分析法组图像质量分值独立样本t检验
Tab. 3  Independent sample t-test for image quality score of the randomized grouping analysis group
表4  随机分组分析法组ROC曲线分析
Tab. 4  Analysis of ROC curve of the randomized grouping analysis group

3 讨论

       本研究创新性地采用对比分组分析法和随机分组分析法两种分组方法,进行受检者分组、数据收集与前瞻性分析,探讨在3.0 T 磁共振扫描仪Fast ASL成像方案的可行性。结果显示,相对于有创的PWI和采集时间长的ASL成像,Fast ASL成像图像质量评分略低,但可获得脑部半定量的CBF图,无创、快速、准确地脑灌注指标检测,解决了ASL采集时间长受检者无法耐受而失败的难题,为脑梗死、脑肿瘤等诸多脑部疾病提供更早更快速地检测脑灌注信息的磁共振技术手段[1,3,7],为预防与诊疗工作提供精准的依据,具有重要且实际的临床意义。

3.1 ASL技术

       ASL利用所标记的动脉中的水分子作为可随意扩散的示踪剂[7, 9],通过标记水和组织进行交换来定量灌注,纵向(T1)弛豫提供一个可测量的衰减率,且所采用的内源性示踪剂限于T1弛豫时间,仅为1~2 s,能更好地反映组织水平的灌注情况。因其无需外源性示踪剂、无辐射、无创反映组织灌注情况,而在科研与临床工作中得到广泛应用。

       ASL是对在成像平面的上游血液进行标记使其自旋弛豫状态改变,待被标记的血流对组织灌注后进行成像。获取成像区包含静态组织和流入组织的标记血液信息的标记像和同样包括静态背景组织和非标记血液信息的非标记像两组对比图像,再通过标记像减去非标记像最终获得灌注像[7,9, 10]

       ASL技术定量测量灌注情况,主要依赖于T1弛豫率、标记方式、成像质量等多种因素[11]。自20世纪90年代以来,ASL技术不断发展,现有的ASL根据流动血液标记方式不同分为连续式ASL(continuous ASL, CASL)和脉冲式ASL(pulsed ASL, PASL),还有准连续式脉冲ASL(pseudo-continuous ASL, PCASL)和速率选择性ASL(velocity selective Arterial Spin Labeling, VSASL)等技术[12]。CASL是连续标记相应层面近端的动脉血液,被标记的血液连续流入组织。PASL则是使用选择性的射频脉冲,脉冲式地标记成像层面近端的一个厚块中的血液,等一段时间使标记的血液与组织充分混合后再进行成像。根据标记脉冲的对称与否,PASL又分为对称式和非对称式,对称式包括血流敏感性的交替反转恢复(flow-sensitive alternating inversion recovery, FAIR)、非血流敏感性的FAIR、带有额外射频脉冲的FAIR、免分离T1测定的FAIR等;非对称式包括信号靶向交替射频、非共振效应控制的近端反转、非敏感转换标记技术、单次减法灌注成像、带有bOth标记的近端标记的双反转和控制iMAges等[13]

       本研究是采用全脑3D PCASL技术[12],采用1.5 s内大于1000次的准连续式脉冲激发对流入动脉血液进行连续标记,待标记血液流入脑组织后,进行全脑3D快速对比非标记成像,测量全脑血流量变化。大大增加自旋标记的效果,并进行背景抑制优化,消除静脉干扰,突出血流量信息,且有良好的电磁吸收率值控制。ASL所标记的动脉中的水分子是可随意扩散的示踪剂,能更好地反映组织水平的灌注情况。3D数据采集可实现全脑非增强灌注扫描,扫描范围广图像SNR高;基于3D FSE序列,对金属及出血性病变的影响不敏感,磁敏感等伪影小;螺旋式K空间填充,扫描速度快,使扫描时间大大缩短且可减轻运动伪影;操作简单易行,重复性好。无论在哪一种采集方案的成像效果,都能体现所采用3D PCASL的优势。

3.2 Fast ASL采集方案的优劣势

       多个研究团队[14, 15, 16]都进行了ASL与PWI的对比研究,认为ASL所得半定量CBF与PWI存在相关性,本实验数据中,无论是哪一分组方法的任意一组数据的结果,证实了这一观点。PWI可提供半定量CBF、脑血容量、平均首过时间、首过达峰时间等,反映高浓度通过血管时的磁敏感效应,但必须使用EPI序列,磁敏感伪影重,而且大分子造影剂无法通过血脑屏障,需要高压注射器团注造影剂,操作复杂,不利于长期多次随访。正因为ASL技术有着比PWI更多的优势,在临床已经应用于缺血、梗死、血管畸形等脑血管疾病,肿瘤性疾病恶性肿瘤分级,癫痫、阿尔茨海默病、帕金森病等慢性退行性改变,感染或炎症性疾病的诊断与随访[17, 18, 19, 20, 21, 22, 23],如图24

       NEX影响SNR和采集时间;Arms是Spiral采集的螺旋臂数(相当于Cartesian采集的相位编码数),数目越多分辨率越高,但分辨率不能过高,否则将导致信噪比过低;层厚的增加一定程度上降低了分辨率,还减少固定范围内的采集层数,进而较少采集时间。本实验中所采用的Fast 3D ASL是以不影响影像诊断效果为前提,采集范围不变,NEX、Arms、层厚为主要优化参数,自动选择使TR、TE、层数也随之变化,致使采集时间大大缩短至原参数的一半,可以尽量减少因为采集时间过长受检者无法耐受而导致的扫描失败。

       无论是哪一分组方法的结果,本研究发现Fast ASL组CBF的图像质量得分均略低于ASL组的,反映灌注情况的敏感度和特异度也略有差异,但均能显示并区分灌注异常灶,有好的敏感度和特异度。

       针对所得到的CBF图的质量评分,通过结合原始图来分析其得分低的原因。对比分组分析法组中,为避免先后顺序带来的数据误差,特意根据入组顺序以“先ASL后Fast ASL”或“先Fast ASL后ASL”交替采集。入选的受检者大多数能很好配合,但该组每个受检者均需同时采集两种方案的ASL因而时间较长,导致无法耐受而出现伪影的比较少,分别有2例,均占8%,但这4例均是以“先ASL后Fast ASL”采集,可见ASL采集时间长确实容易导致失败。随机分组分析法组中,因受检者病痛和采集时间长无法耐受而出现原始图运动伪影的Fast ASL组6例(6%)、ASL组9例(9%)。因无法耐受而导致失败的Fast ASL组也相对较少,可以达到图像质量控制与采集时间等方面的平衡,因此Fast ASL在临床实际工作中应用是可行的。

图2  对比分组分析法组占位性病变图像示例。2A:弥散加权成像;2B:T1WI;2C:T2WI;2D:T2压脂液体抑制反转恢复序列(T2 fs FLAIR);2E:ADC;2F:Sag T2 fs FLAIR;2G:SWAN;2H:3D TOF磁共振血管成像(MRA);2I:快速动脉自选标记(Fast ASL)的0~80标度尺的脑血流量(CBF);2J:动脉自选标记(ASL)的0~80标度尺的CBF;2K:Fast ASL的0~240标度尺的CBF;2L:ASL的0~240标度尺的CBF。ADC为表观弥散系数;Sag为矢状位;SWAN为磁敏感加权血管成像;TOF为时间飞跃法。
Fig. 2  Legend of space occupying lesion in the comparative grouping analysis group. 2A: diffusion weighted imaging; 2B: T1WI; 2C: T2WI; 2D: T2 fat saturation and fluid attenuated inversion recovery (T2 fs FLAIR); 2E: ADC; 2F: Sag T2 fs FLAIR; 2G: SWAN; 2H: 3D TOF magnetic resonance angiography (MRA); 2I: cerebral blood flow (CBF) of 0~80 scale of Fast arterial spin labeling (ASL); 2J: CBF of 0~80 scale of ASL; 2K: CBF of 0-240 scale of Fast ASL; 2L: CBF of the 0-240 scale of ASL. ADC: apparent diffusion coefficient; Sag: sagittal; SWAN: susceptibility weighted angiography; TOF: time-of-flight.
图3  对比分组分析法组血管性病变图像示例。3A:弥散加权成像;3B:T1WI;3C:T2WI;3D:T2压脂液体抑制反转恢复序列(T2 fs FLAIR);3E:ADC;3F:Sag T2 fs FLAIR;3G:Cor T2 fs FLAIR;3H:3D TOF磁共振血管成像(MRA);3I:动脉自选标记(ASL)标记后延迟时间(PLD)=1.5 s的脑血流量(CBF);3J:快速动脉自选标记(Fast ASL)PLD=1.5 s的CBF;3K:ASL PLD=2.5 s的CBF;3L:Fast ASL PLD=2.5 s的CBF。ADC为表观弥散系数;Sag为矢状位;Cor为冠状位;TOF为时间飞跃法。
Fig. 3  Legend of vascular lesions in the comparative grouping analysis group. 3A: Diffusion weighted imaging; 3B: T1WI; 3C: T2WI; 3D: T2 fat saturation and fluid attenuated inversion recovery (T2 fs FLAIR); 3E: ADC; 3F: Sag T2 fs FLAIR; 3G: Cor T2 fs FLAIR; 3H: 3D TOF magnetic resonance angiography (MRA); 3I: arterial spin labeling (ASL) post label delay (PLD) =1.5 s (cerebral blood flow, CBF); 3J: Fast ASL PLD=1.5 s CBF; 3K: ASL PLD= 2.5 s CBF; 3L: Fast ASL PLD= 2.5 s CBF. ADC: apparent diffusion coefficient; Sag: sagittal; Cor: coronal; TOF: time-of-flight.
图4  随机分组分析法组血管性病变和占位性病变的Fast ASL图像示例。4A~4E:单发血管性病变;4A:弥散加权成像(DWI);4B:T1WI;4C:T2WI;4D:T2压脂液体抑制反转恢复序列(T2 fs FLAIR);4E:标记后延迟时间(PLD)=1.5 s 的CBF。4F~4L:多发血管性病变;4F:DWI;4G:T1WI;4H:T2WI;4I:T2 fs FLAIR;4J:SWAN;4K:PLD=1.5 s的CBF;4L:PLD=2.5 s 的CBF。4M~4T:占位性病变;4M:DWI;4N:T1WI;4O:T2WI;4P:T2 fs FLAIR;4Q:T1WI+C;4R:SWAN;4S:0~80标度尺的PLD=2.0 s的CBF;4T:0~60标度尺的PLD=2.0 s的CBF。
Fig. 4  Legend of Fast ASL images of vascular lesions and space occupying lesions in the randomized grouping analysis group. 4A: Diffusion-weighted imaging (DWI); 4B: T1WI; 4C: T2WI; 4D: T2 fat saturation and fluid attenuated inversion recovery (T2 fs FLAIR); 4E: post label delay (PLD) =1.5 s CBF. 4F~4L: multiple vascular lesions; 4F: DWI; 4G: T1WI; 4H: T2WI; 4I: T2 fs FLAIR; 4J: SWAN; 4K: PLD=1.5 s CBF; 4L: PLD=2.5 s CBF. 4M-4T: space occupying lesion; 4M: DWI; 4N: T1WI; 4O: T2WI; 4P: T2 fs FLAIR; 4Q: T1WI+C; 4R: SWAN; 4S: PLD of 0~80 scale=CBF of 2.0 s; 4T: PLD of 0 to 60 scales = CBF of 2.0 s.

3.3 数据分组的必要性

       采用对比分组分析法和随机分组分析法两种入选数据的分组方法进行数据分组后分析,结合原始图比较所得的CBF图像质量评分、敏感度和特异度,探讨实验数据的分组可行性,对比分组分析法组可以避免受检者自身因素引起的数据偏差,随机分组分析法组可以避免人为分组导致的数据偏差。主要都是尽量避免分组带来的数据分布问题。

3.4 局限性

       本研究纳入的受检者包含健康者、脑梗死、转移瘤、颅咽管瘤、星形细胞瘤、脑膜瘤等,但纳入的病变种类不多,且大多是脑梗死、转移瘤。据文献报道[5,24, 25, 26, 27, 28, 29, 30, 31],上述病变大多有灌注异常,且通过ASL可以区分病灶实质区、坏死区及周围水肿区等,而我院有大量肺癌脑肿瘤和脑梗死受检者。这是本实验数据病例纳入上述病种的原因。本研究旨在验证Fast ASL采集方案的效果,入选的均为临床确认且通过ASL或PWI证实灌注情况的案例,这会带来分析上的局限性。

4 结论

       优化参数后的Fast ASL成像,可获得脑部半定量的CBF图,可以明确区分脑组织正常与异常灌注区域,为临床诊疗工作提供准确的脑部灌注信息,对于脑部疾病的防治具有重要意义。尤其是采集时间可以缩短一半,对于无法耐受的受检者,可以获得相对满意的CBF效果,达到了图像质量控制与采集时间的平衡,具有一定的可行性,可实现临床无创、快速、准确的脑灌注指标的检测。

[1]
HERNANDEZ-GARCIA L, LAHIRI A, SCHOLLENBERGER J. Recent progress in ASL[J]. NeuroImage, 2019, 187: 3-16. DOI: 10.1016/j.neuroimage.2017.12.095.
[2]
MANNING P, DAGHIGHI S, RAJARATNAM M K, et al. Differentiation of progressive disease from pseudoprogression using 3D PCASL and DSC perfusion MRI in patients with glioblastoma[J]. J Neurooncol, 2020, 147(3): 681-690. DOI: 10.1007/s11060-020-03475-y.
[3]
PELLERIN A, KHALIFÉ M, SANSON M, et al. Simultaneously acquired PET and ASL imaging biomarkers may be helpful in differentiating progression from pseudo-progression in treated gliomas[J]. Eur Radiol, 2021, 31(10): 7395-7405. DOI: 10.1007/s00330-021-07732-0.
[4]
PANG H P, DANG X F, REN Y, et al. 3D-ASL perfusion correlates with VEGF expression and overall survival in glioma patients: comparison of quantitative perfusion and pathology on accurate spatial location-matched basis[J]. J Magn Reson Imaging, 2019, 50(1): 209-220. DOI: 10.1002/jmri.26562.
[5]
LIU J Y, LIN C, MINUTI A, et al. Arterial spin labeling compared to dynamic susceptibility contrast MR perfusion imaging for assessment of ischemic penumbra: a systematic review[J]. J Neuroimaging, 2021, 31(6): 1067-1076. DOI: 10.1111/jon.12913.
[6]
刘桑妮, 谢春明. 基于动脉自旋标记评估缺血性脑卒中患者脑侧支循环的研究进展[J]. 中华神经医学杂志, 2020, 19(9): 909-915. DOI: 10.3760/cma.j.cn115354-20200225-00116.
LIU S N, XIE C M. Recent advance in assessment of collateral circulation in patients with ischemic stroke based on arterial spin labeling magnetic resonance imaging[J]. Chin J Neuromed, 2020, 19(9): 909-915. DOI: 10.3760/cma.j.cn115354-20200225-00116.
[7]
中华医学会放射学分会质量管理与安全管理学组中华医学会放射学分会磁共振学组. 动脉自旋标记脑灌注MRI技术规范化应用专家共识[J]. 中华放射学杂志, 2016, 50(11): 817-824. DOI: 10.3760/cma.j.issn.1005-1201.2016.11.003.
Chinese Medical Association Chinese Society of Radiology Quality and Safety Group, Chinese Medical Association Chinese Society of Radiology MR Group. Expert consensus on standardized application of spin-labeled cerebral perfusion MRI technique[J]. Chin J Radiol, 2016, 50(11): 817-824. DOI: 10.3760/cma.j.issn.1005-1201.2016.11.003.
[8]
杨永贵, 郭岗. 1.5 T磁共振不同NEX脑部肿瘤酰胺质子转移成像的应用初探[J]. 磁共振成像, 2017, 8(12): 923-928. DOI: 10.12015/issn.1674-8034.2017.12.009.
YANG Y G, GUO G. The application of different NEX amide proton transfer imaging in brain tumors on 1.5 T MRI scanner[J]. Chin J Magn Reson Imaging, 2017, 8(12): 923-928. DOI: 10.12015/issn.1674-8034.2017.12.009.
[9]
NAM K W, KIM C K, KO S B, et al. Regional arterial spin labeling perfusion defect is associated with early ischemic recurrence in patients with a transient ischemic attack[J]. Stroke, 2020, 51(1): 186-192. DOI: 10.1161/STROKEAHA.119.026556.
[10]
WANG Y L, CHEN S, XIAO H F, et al. Differentiation between radiation-induced brain injury and glioma recurrence using 3D pCASL and dynamic susceptibility contrast-enhanced perfusion-weighted imaging[J]. Radiother Oncol, 2018, 129(1): 68-74. DOI: 10.1016/j.radonc.2018.01.009.
[11]
WILLIAMS D S, DETRE J A, LEIGH J S, et al. Magnetic resonance imaging of perfusion using spin inversion of arterial water[J]. Proc Natl Acad Sci USA, 1992, 89(1): 212-216. DOI: 10.1073/pnas.89.1.212.
[12]
周倩, 王倩倩, 刘新疆. 磁共振三维动脉自旋标记技术研究进展及临床应用[J]. 磁共振成像, 2019, 10(12): 955-960. DOI: 10.12015/issn.1674-8034.2019.12.019.
ZHOU Q, WANG Q Q, LIU X J. Research progress and clinical application of three-dimensional magnetic resonance arterial spin labeling[J]. Chin J Magn Reson Imaging, 2019, 10(12): 955-960. DOI: 10.12015/issn.1674-8034.2019.12.019.
[13]
GOLAY X, HENDRIKSE J, LIM T C C. Perfusion imaging using arterial spin labeling[J]. Top Magn Reson Imaging, 2004, 15(1): 10-27. DOI: 10.1097/00002142-200402000-00003.
[14]
ARACKI-TRENKIC A, LAW-YE B, RADOVANOVIC Z, et al. ASL perfusion in acute ischemic stroke: the value of CBF in outcome prediction[J]. Clin Neurol Neurosurg, 2020, 194: 105908 [2022-09-04]. https://www.sciencedirect.com/science/article/abs/pii/S0303846720302511. DOI: 10.1016/j.clineuro.2020.105908.
[15]
LIU D P, XU F, LI W B, et al. Improved velocity-selective-inversion arterial spin labeling for cerebral blood flow mapping with 3D acquisition[J]. Magn Reson Med, 2020, 84(5): 2512-2522. DOI: 10.1002/mrm.28310.
[16]
ZHANG S X, YAO Y H, ZHANG S, et al. Comparative study of DSC-PWI and 3D-ASL in ischemic stroke patients[J]. J Huazhong Univ Sci Technol [Med Sci ], 2015, 35(6): 923-927. DOI: 10.1007/s11596-015-1529-8.
[17]
SENERS P, OPPENHEIM C, TURC G, et al. Perfusion imaging and clinical outcome in acute ischemic stroke with large core[J]. Ann Neurol, 2021, 90(3): 417-427. DOI: 10.1002/ana.26152.
[18]
JIANG L, AI Z P, GENG W, et al. Predictive value of perfusion weighted imaging for early new lesions after stroke patients receive endovascular treatment[J]. Quant Imaging Med Surg, 2021, 11(8): 3643-3654. DOI: 10.21037/qims-21-1.
[19]
SHIN J, KIM Y S, JANG H S, et al. Perfusion recovery on TTP maps after endovascular stroke treatment might predict favorable neurological outcomes[J]. Eur Radiol, 2020, 30(12): 6421-6431. DOI: 10.1007/s00330-020-07066-3.
[20]
NAM K W, KIM C K, YOON B W, et al. Multiphase arterial spin labeling imaging to predict early recurrent ischemic lesion in acute ischemic stroke[J]. Sci Rep, 2022, 12: 1456 [2022-09-04]. https://www.nature.com/articles/s41598-022-05465-8. DOI: 10.1038/s41598-022-05465-8.
[21]
吕瑞瑞, 杨治花, 葛鑫, 等. 集成MRI联合三维动脉自旋标记成像鉴别胶质瘤复发和假性进展的初步研究[J]. 磁共振成像, 2022, 13(8): 19-23, 35. DOI: 10.12015/issn.1674-8034.2022.08.004.
LÜ R R, YANG Z H, GE X, et al. Preliminary study of synthetic MRI combined with three-dimensional arterial spin labeling imaging in differentiating recurrence and pseudoprogression of glioma[J]. Chin J Magn Reson Imaging, 2022, 13(8): 19-23, 35. DOI: 10.12015/issn.1674-8034.2022.08.004.
[22]
保莎莎, 刘一帆, 罗玥媛, 等. 脑胶质瘤治疗后假性进展与复发的影像学鉴别研究进展[J]. 磁共振成像, 2021, 12(3): 85-88. DOI: 10.12015/issn.1674-8034.2021.03.020.
BAO S S, LIU Y F, LUO Y Y, et al. Advances in imaging differentiation of pseudoprogression and recurrence of brain gliomas after treatment[J]. Chin J Magn Reson Imaging, 2021, 12(3): 85-88. DOI: 10.12015/issn.1674-8034.2021.03.020.
[23]
陈红日, 杨文蕊, 李青润, 等. 应用磁共振动脉自旋标记技术对不同运动亚型帕金森病患者脑灌注改变的研究[J]. 磁共振成像, 2021, 12(8): 1-5, 10. DOI: 10.12015/issn.1674-8034.2021.08.001.
CHEN H R, YANG W R, LI Q R, et al. Study on cerebral perfusion changes in patients with Parkinson's disease with different motor subtypes by using arterial spin labeling technique[J]. Chin J Magn Reson Imaging, 2021, 12(8): 1-5, 10. DOI: 10.12015/issn.1674-8034.2021.08.001.
[24]
JOSEPH C R. Utilizing 3D arterial spin labeling to identify cerebrovascular leak and glymphatic obstruction in neurodegenerative disease[J]. Diagnostics (Basel), 2021, 11(10): 1888 [2022-09-04]. https://www.mdpi.com/2075-4418/11/10/1888. DOI: 10.3390/diagnostics11101888.
[25]
ODUDU A, NERY F, HARTEVELD A A, et al. Arterial spin labelling MRI to measure renal perfusion: a systematic review and statement paper[J]. Nephrol Dial Transplant, 2018, 33(suppl_2): ii15-ii21 [2022-09-04]. https://academic.oup.com/ndt/article/33/suppl_2/ii15/5078404?login=true. DOI: 10.1093/ndt/gfy180.
[26]
NERY F, BUCHANAN C E, HARTEVELD A A, et al. Consensus-based technical recommendations for clinical translation of renal ASL MRI[J]. MAGMA, 2020, 33(1): 141-161. DOI: 10.1007/s10334-019-00800-z.
[27]
MCDONALD R J, MCDONALD J S, KALLMES D F, et al. Intracranial gadolinium deposition after contrast-enhanced MR imaging[J]. Radiology, 2015, 275(3): 772-782. DOI: 10.1148/radiol.15150025.
[28]
吴静, 汪文胜, 颜刘清, 等. rCBFmax值对胶质瘤术前分级的价值及与MVD表达间关系的初步研究[J]. 磁共振成像, 2018, 9(6): 417-421. DOI: 10.12015/issn.1674-8034.2018.06.004.
WU J, WANG W S, YAN L Q, et al. The study on rCBFmax value for predicting glioma grading before operation and the relationship between the MVD expression[J]. Chin J Magn Reson Imaging, 2018, 9(6): 417-421. DOI: 10.12015/issn.1674-8034.2018.06.004.
[29]
LU S S, CAO Y Z, SU C Q, et al. Hyperperfusion on arterial spin labeling MRI predicts the 90-day functional outcome after mechanical thrombectomy in ischemic stroke[J]. J Magn Reson Imaging, 2021, 53(6): 1815-1822. DOI: 10.1002/jmri.27455.
[30]
THAMM T, GUO J, ROSENBERG J, et al. Contralateral hemispheric cerebral blood flow measured with arterial spin labeling can predict outcome in acute stroke[J]. Stroke, 2019, 50(12): 3408-3415. DOI: 10.1161/STROKEAHA.119.026499.
[31]
MOROFUJI Y, HORIE N, TATEISHI Y, et al. Arterial spin labeling magnetic resonance imaging can identify the occlusion site and collateral perfusion in patients with acute ischemic stroke: comparison with digital subtraction angiography[J]. Cerebrovasc Dis, 2019, 48(1/2): 70-76. DOI: 10.1159/000503090.

上一篇 卷积神经网络单激发技术在配合不佳患者颅脑磁共振成像中的可行性
下一篇 基于多尺度残差网络的MRI脑肿瘤分类
  
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