分享:
分享到微信朋友圈
X
技术研究
三维高分辨压缩感知对比增强全身MRA技术的可行性研究
李景阳 富青 刘小明 吴佳威 金腾 于群 雷子乔 孔祥闯

Cite this article as: LI J Y, FU Q, LIU X M, et al. Feasibility study of 3D high-resolution compressed sensing contrast-enhanced whole-body MRA imaging technology[J]. Chin J Magn Reson Imaging, 2024, 15(11): 123-129.本文引用格式:李景阳, 富青, 刘小明, 等. 三维高分辨压缩感知对比增强全身MRA技术的可行性研究[J]. 磁共振成像, 2024, 15(11): 123-129. DOI:10.12015/issn.1674-8034.2024.11.019.


[摘要] 目的 研究单次注射、半剂量对比剂三维高分辨压缩感知对比增强全身磁共振血管造影(3D high-resolution compressed sensing contrast-enhanced whole-body magnetic resonance angiography, 3D-CS-CE-MRA)的可行性和临床应用价值。材料与方法 前瞻性收集83例临床怀疑全身动脉疾病前来行磁共振全身血管检查的患者,随机分为A、B两组,A组44例,B组39例。A组行3D-CS-CE-MRA扫描,对比剂总量为0.15 mmol/kg,单次注射,流速为2 mL/s;B组行常规三维高分辨对比增强全身磁共振血管造影(3D high-resolution contrast-enhanced whole-body magnetic resonance angiography, 3D-CE-MRA)扫描,对比剂总量为0.3 mmol/kg,分两次注射,每次用量均为0.15 mmol/kg,流速为2 mL/s。两名经验丰富的放射科医生独立分析两组25个动脉节段的图像质量,并对各段动脉进行评分;选取颈总动脉、腹主动脉、股动脉、腘动脉测量信噪比(signal-to-noise ratio, SNR)和对比噪声比(contrast-to-noise ratio, CNR),进行定量分析。采用独立样本t检验、Wilcox秩和检验进行统计分析。结果 SNR颈总、SNR股动脉、SNR腘动脉、CNR颈总、CNR股动脉、CNR腘动脉总体测量结果差异无统计学意义(P均<0.05),SNR腹主、CNR腹主总体测量结果差异有统计学意义(P均<0.05),B组均值高于A组。3D-CS-CE-MRA平均得分优于3D-CE-MRA[(3.41±0.52)vs.(3.18±0.48),P<0.05];腹主动脉、肾动脉、肠系膜上动脉、腹腔干、髂总动脉、髂外动脉、胫前动脉、腓动脉血管显示评分差异无统计学意义(P均>0.05),A、B两组图像质量差异无统计学意义;颈总动脉、头臂干、椎动脉、胸主动脉、股动脉、腘动脉、胫后动脉血管显示评分差异有统计学意义(P均<0.05),A组评分高于B组。结论 与常规3D-CE-MRA相比,3D-CS-CE-MRA只需一次对比剂注射且无需减影即可快速完成磁共振全身血管检查,在保证图像质量的前提下,减少了对比剂用量,缩短了检查时间,提高患者耐受程度,具有较好的临床应用价值。
[Abstract] Objective To investigate the feasibility and clinical value of 3D high-resolution compressed sensing contrast-enhanced whole-body magnetic resonance angiography (3D-CS-CE-MRA) with single injection and half-dose contrast agent.Materials and Methods A total of 83 patients who were suspected of systemic arterial disease and underwent MRI systemic vascular examination were prospectively enrolled and randomly divided into two groups, A and B, with 44 patients in group A and 39 patients in group B. Group A underwent 3D-CS-CE-MRA scanning with a total contrast agent volume of 0.15 mmol/kg injected at a flow rate of 2 mL/s, while group B underwent conventional 3D high-resolution contrast-enhanced whole-body magnetic resonance angiography (3D-CE-MRA) scanning with a total contrast agent volume of 0.3 mmol/kg administered in two doses of 0.15 mmol/kg each at a flow rate of 2 mL/s. Two experienced radiologists independently scored the qualitative image quality of 25 arterial segments,and signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were measured for the common carotid artery, abdominal aorta, femoral artery, and popliteal artery. The independent samples t-test and the Wilcoxon rank sum test was used for statistical analysis.Results There was no significant difference in SNRs and CNRs for the common carotid artery,the femoral artery and the popliteal artery (all P>0.05), but there was significant difference in SNRs and CNRs for the abdominal aorta (all P<0.05), the average value of group B was higher than that of group A. The average score of 3D-CS-CE-MRA was better than 3D-CE-MRA [(3.41±0.52) vs. (3.18±0.48), P<0.05]. There was no significant difference in scores of abdominal aorta, renal artery, superior mesenteric artery, celiac trunk, common iliac artery, external iliac artery, anterior tibial artery, and peroneal artery (all P>0.05); there was no difference in image quality between group A and B. There was significant difference in scores of common carotid artery, brachiocephalic trunk, vertebral artery, thoracic aorta, femoral artery, popliteal artery, and posterior tibial artery (all P<0.05), the scores of group A was higher than group B.Conclusions Compared with conventional 3D-CE-MRA, 3D-CS-CE-MRA can quickly complete magnetic resonance whole-body vascular examination with only one contrast agent injection and subtractionless, under the premise of ensuring image quality, it reduces the amount of contrast agent, shortens the examination time, and improves patient tolerance, and has great clinical application value.
[关键词] 磁共振成像;压缩感知;多回波Dixon;对比增强;全身磁共振血管造影
[Keywords] magnetic resonance imaging;compressed sensing;multi-echo Dixon;contrast-enhanced;whole-body magnetic resonance angiography

李景阳 1, 2   富青 1, 2   刘小明 1, 2   吴佳威 1, 2   金腾 1, 2   于群 1, 2   雷子乔 1, 2   孔祥闯 1, 2*  

1 华中科技大学同济医学院附属协和医院放射科,武汉 430022

2 分子影像湖北省重点实验室,武汉 430022

通信作者:孔祥闯,E-mail: hongke80@163.com

作者贡献声明:孔祥闯设计本研究的方案,对稿件重要内容进行了修改;李景阳起草和撰写稿件,获取、分析和解释本研究的数据;富青、刘小明、吴佳威、金腾、于群、雷子乔获取、分析或解释本研究的数据,对稿件重要内容进行了修改;全体作者都同意发表最后的修改稿,同意对本研究的所有方面负责,确保本研究的准确性和诚信。


收稿日期:2024-07-01
接受日期:2024-11-08
中图分类号:R445.2  R543.5 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.11.019
本文引用格式:李景阳, 富青, 刘小明, 等. 三维高分辨压缩感知对比增强全身MRA技术的可行性研究[J]. 磁共振成像, 2024, 15(11): 123-129. DOI:10.12015/issn.1674-8034.2024.11.019.

0 引言

       许多动脉病变,如动脉粥样硬化、血管炎及心血管疾病等均会累及全身动脉[1, 2]。数字减影血管造影(digital subtraction angiography, DSA)虽是诊断血管病变的“金标准”,但由于其是有创性检查、涉及X线电离辐射危害、碘对比剂过敏风险、术后多种并发症等缺点而极少运用于疾病诊断[3, 4]。超声检查无创,价廉,并能半定量测定血流,但它有操作者依赖性,难以提供临床医师感兴趣的血管全貌图[5, 6, 7]。CT血管造影术(computed tomography angiography, CTA)也有电离辐射并须使用含碘对比剂,发生不良反应概率偏高,且具有肾毒性[8]

       对比增强磁共振血管造影(contrast-enhanced magnetic resonance angiography, CE-MRA)目前已成为血管病变的主要影像检查技术,几乎可用于全身各级血管进行诊断评估[1, 9]。无创性、无电离辐射及钆对比剂的安全性使CE-MRA成为比DSA和CTA更理想的选择[10, 11, 12]。MRI全身血管检查时,目前的技术通常需要两次注射对比剂,两次扫描经过减影后才可完成[13],大大增加了患者的肾功能负荷和检查时间。多回波Dixon(multi-echo Dixon, mDixon)[14, 15]是一种新的水脂分离技术,与传统减影技术相比,具有更好的信噪比(signal-to-noise ratio, SNR),更高的空间分辨率以及更短的扫描时间,已经成功应用于下肢血管成像[16]。压缩感知(compressed sensing, CS)是近年来出现的新型快速MRI技术,其原理是在非线性迭代重建中,利用图像结构的内在稀疏性和欠采样K空间数据重建获得逼近全采样的图像质量,从而缩短了采集时间[17, 18, 19]。由于CE-MRA的数据正好符合稀疏性,K空间中心数据主要决定图像的对比,因此CS加速技术适用于CE-MRA成像[20]。目前,国外已有CS在MRI中的报道,如心脏电影成像[21]、时间飞跃MRA(time of flight MRA, TOF-MRA)[22]等,与传统的并行采集(parallel imaging, PI)加速技术相比,无论是成像速度还是图像质量,CS均优于PI[23, 24, 25]。笔者尚未见国内外报道CS技术应用于全身血管成像的先例,因此尝试利用CS和mDixon技术优化全身血管成像技术,实现仅一次注射常规剂量1/2的对比剂,一次顺序扫描完成且无需减影,旨在探讨其在全身血管成像中缩短检查时间和减少对比剂用量等方面的临床应用价值,希望为临床提供一种更快速且实用的磁共振全身血管扫描方案,助力于提高患者耐受程度,减少运动伪影,提高工作效率。

1 材料与方法

1.1 研究对象

       本研究遵守《赫尔辛基宣言》,经华中科技大学同济医学院附属协和医院伦理委员会的批准[批准文号:(2021)伦审字(0122)号],全体受试者均签署了知情同意书,于2022年3月至11月在华中科技大学同济医学院附属协和医院按照就诊时间顺序前瞻性收集83例临床怀疑全身动脉疾病来行MRI全身血管检查的患者,随机分为A、B两组,A组(实验组)44例,男21例,女23例,年龄(54.8±17.2)岁,行3D-CS-CE-MRA扫描;B组(对照组)39例,男18例,女21例,年龄(48.7±16.2)岁,行常规3D-CE-MRA扫描。样本量估算使用两样本均数比较样本量计算公式:N1=N2=2[(µαβσ/δ]2,N1,N2分别表示实验组和对照组的样本量,设检验水准α=0.05,β=0.1,通过查表可知µα=1.96,µβ=1.28,σ表示两样本标准差的估计值,计算中可用两样本标准差代替;δ表示两样本均数的差值,σδ查阅相关原始资料得出δ/σ=0.83,带入计算公式可得N1=N2=30;再考虑20%的样本流失率,实验组和对照组各样本量最少为37,因此,本研究对照组和实验组分别选取约40例的病例。

       纳入标准:(1)可疑全身血管疾病的患者;(2)年龄18~80岁。排除标准:(1)植入心脏起搏器的患者;(2)处于怀孕或哺乳期的妇女;(3)其他不能兼容磁共振的金属植入物患者;(4)钆对比剂过敏患者。

1.2 检查方法

       应用3.0 T MR(Ingenia CX, Philips Healthcare, Best, the Netherlands),具有80 mT/m超高梯度场强和200 mT/m/ms的超高梯度切换率。线圈采用16通道头颈联合线圈,2片32通道腹部线圈和脊柱线圈。受试者扫描过程中采取仰卧位,身体位于床面中央,正中矢状面与床面垂直,身高大于160 cm分5段采集,小于160 cm分4段采集,扫描范围从颅顶到足底,1~5段依次覆盖头颈动脉、胸腹部动脉、盆腔动脉、大腿和小腿全部动脉及其分支,每段扫描视野为450 mm×450 mm,相邻两段重叠90~120 mm(根据患者身高灵活调整),便于后处理拼接,均采集冠状位,无需改变患者体位。记录A、B两组的平均扫描时间及图像重建时间。

       A组扫描参数见表1。对比剂注射方案:对比剂用量为0.15 mmol/kg的钆喷酸葡胺(Gd-DTPA,佳迪显,中国恒瑞),a-b-a-b式单次四相注射,即a筒(对比剂)0.1 mmol/kg,b筒(盐水)20 mL,a筒(对比剂)0.05 mmol/kg,b筒(盐水)20 mL,流速均为2 mL/s。采用透视触发技术,首先启动超快速二维梯度回波序列,将肺动脉作为目标血管进行监控,当对比剂到达上腔静脉,让受检者执行吸气憋气口令,当对比剂达到肺动脉峰值时,马上切换到mDixon序列并启动扫描,依次采集第1~5段,第1段设置K空间中心部分最后填充,扫描时间9.8 s,第2段设置K空间中心部分优先填充,扫描时间11.2 s,屏气一次扫描第1段和第2段共21 s,两段K空间中心填充时间差约为4~5 s。

       B组扫描参数见表2。分两次注射对比剂,对比剂总量为0.3 mmol/kg的钆喷酸葡胺(Gd-DTPA,佳迪显,中国恒瑞),每次用量均为0.15 mmol/kg,两次对比剂注射完毕后均追加20 mL生理盐水,流速均为2 mL/s。先行1~5段的常规定位和血管定位像的扫描,然后在每段血管定位像上确定相应的扫描范围。同样采用透视触发技术,首先启动超快速二维梯度回波序列,将肺动脉作为目标血管进行监控,当对比剂进入目标血管,立刻让受试者执行吸气憋气口令,然后马上切换到3D T1-FFE序列并启动扫描,依次采集第2~5段,扫描完成后间隔10 min进行第二次扫描,即扫描第1段,监测颈动脉,当对比剂到达颈总动脉时启动3D T1-FFE序列扫描。每段在增强前后均扫描1次,通过减影自动去除背景信号,得到纯血管像。

表1  3D-CS-CE-MRA扫描参数
Tab. 1  Parameters of 3D-CS-CE-MRA
表2  3D-CE-MRA扫描参数
Tab. 2  Parameters of 3D-CE-MRA

1.3 图像分析

       扫描获得的原始图像,在后处理工作站上用最大密度投影法(maximum intensity projection, MIP)进行重建。然后用图像合成软件(MobiView, R5.71 version, Philips Healthcare, Best, the Netherlands)把1~5段的MIP图像融合成动脉全貌图。

       定量分析:由一名有20年工作经验的副主任技师测量血管的SNR和对比噪声比(contrast-to-noise ratio, CNR)。选取左颈总动脉(颈动脉分叉层面)、腹主动脉(平腹腔干层面)、左侧股动脉(髂外动脉和股动脉移行层面)、左侧腘动脉(股动脉和腘动脉移行层面)勾画圆形感兴趣区(region of interest, ROI),同层背景区肌肉组织勾画圆形ROI,颈总动脉、股动脉、腘动脉ROI像素面积均值0.2 mm2,腹主动脉ROI像素面积均值2 mm2,每个ROI测量两次,取平均值,记录血管平均信号强度,背景平均信号强度,背景噪声标准差(standard deviation, SD),分别计算SNR和CNR,见式(1)~(2)。

       其中,S血管为血管的平均信号强度,S背景为背景的平均信号强度,SD背景为背景噪声标准差。

       定性分析:两名有10年诊断经验的放射科主治医师分别对A、B两组25段动脉进行评分:胸主动脉、腹主动脉、颈总动脉(双侧)、头臂干、椎动脉(双侧)、髂总动脉(双侧)、髂外动脉(双侧)、肾动脉(双侧)、腹腔干动脉、肠系膜上动脉、股动脉(双侧)、腘动脉(双侧)、胫前动脉(双侧)、胫后动脉(双侧)、腓动脉(双侧)。采用4分法[26]进行评分:4分,显示很好,血管轮廓清楚且连续,信号均匀;3分,显示较好,血管轮廓较清楚且连续,信号较均匀;2分,显示一般,血管轮廓较清楚但不连续,信号较均匀;1分,显示差,血管轮廓不清楚、不连续,信号不均匀。

1.4 统计学分析

       采用SPSS 26.0(IBM Corp.,Armonk, New York, USA)统计学分析软件。采用组内相关系数(intra-class correlation coefficient, ICC)评估两名诊断医师定性分析、动脉评分的一致性:ICC值>0.80时,表示较完美的一致性;0.60<ICC值≤0.80时,表示一致性较好;0.40<ICC值≤0.60时,表示一致性中等;0.20<ICC值≤0.40时,表示一致性较低;ICC值≤0.20时,表示一致性差。采用Shapiro-Wilkin对数据进行正态性分析,符合正态分布的计量资料用x¯±s表示,采用独立样本t检验进行比较;不符合正态分布计量资料采用MP25,P75)表示,组间比较采用Wilcoxon秩和检验。P<0.05为差异有统计学意义。

2 结果

2.1 一般资料

       83例患者均顺利完成全身血管成像检查,无任何不良反应发生,A、B两组患者身高、体质量、年龄差异均无统计学意义(P>0.05)(表3)。A组44例患者平均扫描时间为60 s,B组39例患者平均扫描时间为194 s,相较于3D-CE-MRA,3D-CS-CE-MRA缩短了近70%的扫描时间。

表3  A、B两组一般资料比较
Tab. 3  Comparison of general information of group A and B

2.2 定量分析

       A组和B组SNR颈总、SNR股动脉、SNR腘动脉、CNR颈总、CNR股动脉、CNR腘动脉总体测量结果差异无统计学意义(P均>0.05);SNR腹主、CNR腹主的平均值B组高于A组,总体测量结果差异有统计学意义(P均<0.05)(表4)。

表4  A、B两组图像质量客观评价比较
Tab. 4  Comparison of objective evaluation of image quality of group A and B

2.3 定性分析

       A组有1101段动脉接受评估,B组有975段动脉接受评估,两位医师对动脉评分的一致性较好(ICC=0.787~0.901,P<0.001)。A组6例患者,B组5例患者全身血管检查未见明显异常(图1);A组37例患者有不同程度的动脉狭窄(图2),1例患者双右肾动脉供血,B组32例患者有不同程度的动脉狭窄;B组两例患者由于扫描时间长,检查过程中抖动导致减影错位,静脉污染严重(图3),A组未有此情况发生;B组15例患者由于第一次的残留对比剂影响,导致静脉污染严重,干扰了颈部动脉的评估(图4),而A组只有3例仅轻微静脉污染。3D-CS-CE-MRA平均得分优于3D-CE-MRA[(3.41±0.52)vs.(3.18±0.48),P<0.05]。腹主动脉、肾动脉、肠系膜上动脉、腹腔干、髂总动脉、髂外动脉、胫前动脉、腓动脉血管显示评分差异无统计学意义(P均>0.05),A、B两组图像质量差异无统计学意义;颈总动脉、头臂干、椎动脉、胸主动脉、股动脉、腘动脉、胫后动脉血管显示评分差异有统计学意义(P均<0.05),A组评分高于B组(表5)。

图1  全身各节段动脉图。1A:男,48岁,行3D-CS-CE-MRA扫描;1B:男,39岁,行3D-CE-MRA扫描。两种扫描方案均清晰显示了全身各节段动脉,未见明显异常,图像质量好,但3D-CE-MRA颈部动脉有少许静脉污染。3D-CS-CE-MRA:三维高分辨压缩感知对比增强全身磁共振血管造影;3D-CE-MRA:三维高分辨对比增强全身磁共振血管造影。
Fig. 1  Arterial diagram of the whole body. 1A: Male, 48 years old, underwent 3D-CS-CE-MRA scanning; 1B: Male, 39 years old, underwent 3D-CE-MRA scan. Both scanning schemes clearly show all segments of the body's arteries with no obvious abnormalities and good image quality, but 3D-CE-MRA neck arteries shows a little venous contamination. 3D-CS-CE-MRA: 3D high-resolution compressed sensing contrast-enhanced whole-body magnetic resonance angiography; 3D-CE-MRA: 3D high-resolution contrast-enhanced whole-body magnetic resonance angiography.
图2  男,35岁,白塞病3年余。3D-CS-CE-MRA(2A)清晰显示右侧髂外动脉远端-股动脉近端闭塞(2B,白箭),右侧胫前、胫后动脉不同程度狭窄,管腔粗细不均,腓动脉显影浅淡(2C,黄箭)。3D-CS-CE-MRA:三维高分辨压缩感知对比增强全身磁共振血管造影。
Fig. 2  Male, 35 years old, behcet disease for more than 3 years. 3D-CS-CE-MRA (2A) clearly shows occlusion of the distal part of the right external iliac artery and the proximal part of the femoral artery (2B, white arrow), different degrees of stenosis of the right anterior tibial artery and posterior tibial artery, uneven lumen, and faint visualization of the peroneal artery (2C, yellow arrow). 3D-CS-CE-MRA: 3D high-resolution compressed sensing contrast-enhanced whole-body magnetic resonance angiography.
图3  扫描对比图。3A:女,22岁,行3D-CS-CE-MRA扫描;全身各动脉节段显示良好,无运动伪影且几乎没有静脉污染,肠系膜上动脉起始处管腔狭窄致肠系膜上动脉与肠系膜下动脉之间Riolan动脉弓代偿性增粗(黄箭)。3B:男,68岁,下肢动脉硬化闭塞症,行3D-CE-MRA扫描;两次扫描时间长,两次注射对比剂以及扫描过程中患者的抖动导致减影错位,静脉污染严重,诊断难度大。3D-CS-CE-MRA:三维高分辨压缩感知对比增强全身磁共振血管造影;3D-CE-MRA:三维高分辨对比增强全身磁共振血管造影。
Fig. 3  Scan comparison diagram. 3A: A female patient, 22 years old, 3D-CS-CE-MRA. All arterial segments are well displayed with no motion artifacts and almost no venous contamination, the stenosis of the lumen at the origin of the superior mesenteric artery leads to compensatory thickening of the Riolan arch between the superior and inferior mesenteric arteries (yellow arrow). 3B: A male patient, 68 years old, arteriosclerosis obliterans of the lower extremities, 3D-CE-MRA. There are more vessel segments contaminated by veins due to longer scanning time, double contrast agent injection and misaligned subtraction because of patient motion. This is a great challenge for the accuracy of radiological diagnosis. 3D-CS-CE-MRA: 3D high-resolution compressed sensing contrast-enhanced whole-body magnetic resonance angiography; 3D-CE-MRA: 3D high-resolution contrast-enhanced whole-body magnetic resonance angiography.
图4  女,44岁,白塞病,反复口腔溃疡三年余。3D-CE-MRA全身MIP融合图像(4A),注射对比剂前后,左下肢位置的改变导致下肢血管拼接失败,出现错层(白箭);颈部MRA多方位MIP图(4B、4C)显示,前一次注射后残留的对比剂导致静脉污染严重,干扰了颈部血管的评估,右侧椎动脉起始段及双侧椎动脉V3段显影模糊,需要鉴别病变。3D-CE-MRA:三维高分辨对比增强全身磁共振血管造影;MIP:最大密度投影;MRA:全身磁共振血管造影。
Fig. 4  Female, 44 years old, Behcet disease, recurrent mouth ulcer for more than three years. 3D-CE-MRA whole body MIP fusion image (4A), before and after injection of contrast agent, the change in the position of the left lower limb led to the failure of lower limb vascular splicing and split-layer (white arrow). The MRA multidirectional MIP map (4B, 4C) of the neck shows that the residual contrast agent after the previous injection led to serious venous contamination, which interfers with the evaluation of the neck blood vessels. The development of the right vertebral artery initial segment and bilateral vertebral artery V3 segment are blurred, and the lesions need to be identified. 3D-CE-MRA:3D high-resolution contrast-enhanced whole-body magnetic resonance angiography; MIP: maximum intensity projection; MRA: magnetic resonance angiography.
表5  A、B两组25个动脉节段图像质量主观评分比较
Tab. 5  Comparison of subjective evaluation of image quality in 25 vessel segments of group A and B

3 讨论

       本研究证明了3D-CS-CE-MRA技术的可行性和实用性,该方案采用了CS加速技术及mDixon技术,只需一次注射对比剂即可快速完成全身血管成像,一次顺序扫描且无需减影,图像质量更高。与传统方案相比,扫描时间缩短了近70%,对比剂用量减少了一半。扫描时间的缩短提升了患者的耐受程度和检查成功率,降低了运动伪影的发生率;对比剂量的减少降低了钆对比剂相关风险,包括肾源性系统性纤维化、中枢神经系统钆沉积等,也降低了钆对比剂使用成本,减轻了患者的经济负担。

3.1 加速技术的对比

       传统全身血管成像多采用PI加速技术,但是PI加速因子通常仅为2或3倍,因为在较大的加速因子下噪声或混叠现象会迅速增加[27, 28],对应每段的扫描时间为13~15 s,而颈部动脉与胸腹部动脉成像的最佳时间窗仅相差4~5 s,因此单次注射对比剂完成全身血管成像很难实现。在本研究采用CS加速技术,CS是一种新的信息获取与处理理论,近年来被应用于医学图像重建算法的改进,效果非常显著[29, 30, 31]。它是通过在适当的变换域中利用潜在稀疏性来对K空间进行欠采样的方法,直接感知压缩之后的信号,通过有选择性地采集少量重要数据并采用有效的重构算法实现原始信号的重构,实现缩短信号采集所需时间,并在一定程度上保持原始信号重建质量的要求。本研究中的图像采集方案用10倍的CS加速因子,对应第一段和第二段的扫描时间分别为9.8 s和11.2 s,而本研究又采用了反向K空间填充技术,即第1段设置K空间中心部分最后填充,第2段设置K空间中心部分优先填充,两段K空间中心填充时间差约为4~5 s,因此可以使头颈动脉和胸腹部动脉完美衔接。虽然CS加速技术会一定程度上降低图像信噪比,但在我们的研究中,A组CS加速因子设置为10,B组sense加速因子设置为4时两组间颈总动脉、股动脉、腘动脉的SNR和CNR没有显著性差异,但对于腹主动脉SNR和CNR的定量分析,A组不及B组,可能是因为A组采用了反向K空间填充技术,第二段对比剂达峰时间与填充K空间中心时间未能完美契合,采集的信号自然偏低一些,而B组对比剂达峰时即为K空间中心填充时,动脉的强化更好,但这并不影响腹主动脉的主观评价。

3.2 减影技术与mDixon技术的对比

       传统全身血管成像需要在注射对比剂前扫描蒙片,注射对比剂后,重复扫描一次,通过减影技术去除背景组织,才能获得纯血管像,不仅效率慢,而且一旦患者位置移动,减影后的血管像就会产生运动伪影;其次,减影技术还会使血管的信噪比降低2倍[32],而且第二次注射对比剂势必造成在采集第1段时,该段动脉内已经有第一次注射后残留的对比剂,虽然用减影技术可去除残留对比剂的信号,但也造成了第1段动脉信号对比下降[16]。本方案采用了mDixon技术,mDixon的7峰脂肪重建模型保证了精准的水脂分离,其水相图无需减影即可获得极佳的压脂效果[33],这样不仅避免了减影带来的伪影、噪声、重复检查,而且血管与背景的对比度更高[15, 34]。本研究结果也显示,B组15例患者由于第一次的残留对比剂影响,导致静脉污染严重,干扰了颈部动脉的评估,A组只有3例仅轻微静脉污染,可能是由于触发扫描时间稍晚致颈静脉早期显影,但并不影响颈部动脉的评估,无论是SNR和CNR的测量结果,还是动脉节段的评分结果,3D-CS-CE-MRA均优于3D-CE-MRA。对于股动脉等下肢动脉,3D-CS-CE-MRA的定性评分、SNR及CNR均高于3D-CE-MRA,这与WEISS等[16]的研究结果基本一致,但其只评估了SNR和CNR,未进行定性评分,样本量相对较小(10例),且未评估颈部动脉及胸腹部动脉。

3.3 本研究的局限性

       本研究存在以下不足:(1)后处理速度慢,本研究中每例患者完成检查后的图像重建时间都在5~10 min,大大降低了工作效率,可通过高性能计算机系统或更好的重建算法来解决。(2)由于MRI设备的局限,每段扫描的最大视野只有450 mm,而且相邻两段之间重叠范围在100 mm左右,导致身高较高的患者需要采集5段方可完成全身血管检查,相比于500 mm大视野采集4段即可完成,5段采集无疑增加了采集时间,加大了成像的难度,可通过硬件的更新,开发更大视野高性能MRI设备解决。(3)未对颅脑动脉,冠状动脉以及双侧上肢动脉进行研究,原因如下,①1.1×1.2×1.2 mm3的体素不足以评估颅内动脉的病变,不打对比剂的颅脑TOF-MRA技术已经十分成熟;②冠状动脉MRA成像仍十分具有挑战性;③腹部线圈尺寸的限制导致左右方向的视野不宜过大;④未与CTA及金标准DSA进行对比研究,未来的研究方向应该是与CTA和DSA为标准,评估其诊断疾病的敏感度和特异度。

4 结论

       综上所述,与常规3D-CE-MRA相比,3D-CS-CE-MRA只需一次对比剂注射且无需减影即可快速完成磁共振全身血管检查,在保证图像质量的前提下,减少了对比剂用量,缩短了检查时间,提高患者耐受程度,具有较好的临床应用价值。

[1]
WEIR-MCCALL J R, WHITE R D, RAMKUMAR P G, et al. Follow-up of atheroma burden with sequential whole body contrast enhanced MR angiography: a feasibility study[J]. Int J Cardiovasc Imaging, 2016, 32(5): 825-832. DOI: 10.1007/s10554-016-0842-z.
[2]
GANDY S J, LAMBERT M, BELCH J J, et al. Technical assessment of whole body angiography and cardiac function within a single MRI examination[J]. Clin Radiol, 2015, 70(6): 595-603. DOI: 10.1016/j.crad.2015.02.003.
[3]
GALANAKIS N, MARIS T G, KONTOPODIS N, et al. Perfusion imaging techniques in lower extremity peripheral arterial disease[J/OL]. Br J Radiol, 2022, 95(1135): 20211203 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/35522774/. DOI: 10.1259/bjr.20211203.
[4]
AL-LAMI B S, DLSHAD B, AL-TAWIL Y N, et al. Comparative diagnostic efficacy of cranial CT, CTA, and DSA in subarachnoid hemorrhage management: a systematic review and meta-analysis[J/OL]. J Med Imaging Radiat Sci, 2024, 55(3): 101427 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/38772769/. DOI: 10.1016/j.jmir.2024.04.020.
[5]
ANDERSEN S B, SØRENSEN C M, JENSEN J A, et al. Microvascular imaging with super-resolution ultrasound[J]. Ultraschall Med, 2022, 43(6): 543-547. DOI: 10.1055/a-1937-6868.
[6]
GARCÍA-JIMÉNEZ R, ARROYO E, BORRERO C, et al. Evaluation of placental micro-vascularization by superb micro-vascular imaging Doppler in cases of intra-uterine growth restriction: a first step[J]. Ultrasound Med Biol, 2021, 47(6): 1631-1636. DOI: 10.1016/j.ultrasmedbio.2021.01.029.
[7]
罗浩柔, 尹立雪. 超声微血管成像与彩色多普勒血流成像对甲状腺结节诊断价值的Meta分析[J]. 中华医学超声杂志(电子版), 2021, 18(6): 554-563. DOI: 10.3877/cma.j.issn.1672-6448.2021.06.004.
LUO H R, YIN L X. Diagnostic value of superbmicrovascular imaging and color Doppler flow imaging in thyroid nodules: a Meta-analysis[J]. Chin J Med Ultrasound Electron Ed, 2021, 18(6): 554-563. DOI: 10.3877/cma.j.issn.1672-6448.2021.06.004.
[8]
中华医学会放射学分会质量控制与安全管理专业委员会. 肾病患者静脉注射碘对比剂应用专家共识[J]. 中华放射学杂志, 2021, 55(6): 580-590. DOI: 10.3760/cma.j.cn112149-20201111-01226.
Quality Control and Safety Management Committee of Chinese Society of Radiology Chinese Medical. Expert consensus of iodinated contrast agent use in patients with renal diseases[J]. Chin J Radiol, 2021, 55(6): 580-590. DOI: 10.3760/cma.j.cn112149-20201111-01226.
[9]
LAMBERT M A, WEIR-MCCALL J R, SALSANO M, et al. Prevalence and distribution of atherosclerosis in a low- to intermediate-risk population: assessment with whole-body MR angiography[J]. Radiology, 2018, 287(3): 795-804. DOI: 10.1148/radiol.2018171609.
[10]
KNOBLOCH G, LAUFF M T, HANKE M, et al. Non-contrast-enhanced MR-angiography (MRA) of lower extremity peripheral arterial disease at 3tesla: examination time and diagnostic performance of 2quiescent-interval single-shot MRA vsD. 3D fast spin-Echo MRA[J/OL]. Magn Reson Imaging, 2021, 76: 17-25 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/33157187/. DOI: 10.1016/j.mri.2020.10.016.
[11]
ZHANG J, DING S H, ZHAO H L, et al. Evaluation of chronic carotid artery occlusion by non-contrast 3D-MERGE MR vessel wall imaging: comparison with 3D-TOF-MRA, contrast-enhanced MRA, and DSA[J]. Eur Radiol, 2020, 30(11): 5805-5814. DOI: 10.1007/s00330-020-06989-1.
[12]
NAGPAL P, GRIST T M. MR angiography: contrast-enhanced acquisition techniques[J]. Magn Reson Imaging Clin N Am, 2023, 31(3): 493-501. DOI: 10.1016/j.mric.2023.04.007.
[13]
FENCHEL M, REQUARDT M, TOMASCHKO K, et al. Whole-body MR angiography using a novel 32-receiving-channel MR system with surface coil technology: first clinical experience[J]. J Magn Reson Imaging, 2005, 21(5): 596-603. DOI: 10.1002/jmri.20303.
[14]
STINSON E G, TRZASKO J D, FLETCHER J G, et al. Dual echo Dixon imaging with a constrained phase signal model and graph cuts reconstruction[J]. Magn Reson Med, 2017, 78(6): 2203-2215. DOI: 10.1002/mrm.26620.
[15]
ZHANG T, CHEN Y X, BAO S S, et al. Resolving phase ambiguity in dual-echo Dixon imaging using a projected power method[J/OL]. Magn Reson Med, 2017, 77(5): 2066-2076 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/27221766/. DOI: 10.1002/mrm.26287.
[16]
WEISS K J, EGGERS H, STEHNING C, et al. Feasibility and robustness of 3T magnetic resonance angiography using modified Dixon fat suppression in patients with known or suspected peripheral artery disease[J/OL]. Front Cardiovasc Med, 2020, 7: 549392 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/33195449/. DOI: 10.3389/fcvm.2020.549392.
[17]
ZHANG X, CAO Y Z, MU X H, et al. Highly accelerated compressed sensing time-of-flight magnetic resonance angiography may be reliable for diagnosing head and neck arterial steno-occlusive disease: a comparative study with digital subtraction angiography[J]. Eur Radiol, 2020, 30(6): 3059-3065. DOI: 10.1007/s00330-020-06682-3.
[18]
KIDO T, HIRAI K, OGAWA R, et al. Comparison between conventional and compressed sensing cine cardiovascular magnetic resonance for feature tracking global circumferentialstrain assessment[J/OL]. J Cardiovasc Magn Reson, 2021, 23(1): 10 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/33618722/. DOI: 10.1186/s12968-021-00708-5.
[19]
DATTA S, PAUL J S. Adaptive continuation based smooth l0-norm approximation for compressed sensing MR image reconstruction[J/OL]. J Med Imaging, 2024, 11(3): 035003 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/38827777/. DOI: 10.1117/1.JMI.11.3.035003.
[20]
OGAWA R, KIDO T, NAKAMURA M, et al. Comparison of compressed sensing and conventional coronary magnetic resonance angiography for detection of coronary artery stenosis[J/OL]. Eur J Radiol, 2020, 129: 109124 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/32563962/. DOI: 10.1016/j.ejrad.2020.109124.
[21]
ZHANG X Y, XIE G X, LU N, et al. 3D self-gated cardiac cine imaging at 3 Tesla using stack-of-stars bSSFP with tiny golden angles and compressed sensing[J]. Magn Reson Med, 2019, 81(5): 3234-3244. DOI: 10.1002/mrm.27612.
[22]
DE BUCK M H S, JEZZARD P, HESS A T. Optimization of undersampling parameters for 3D intracranial compressed sensing MR angiography at 7 T[J]. Magn Reson Med, 2022, 88(2): 880-889. DOI: 10.1002/mrm.29236.
[23]
JURKA M, MACOVA I, WAGNEROVA M, et al. Deep-learning-based reconstruction of T2-weighted magnetic resonance imaging of the prostate accelerated by compressed sensing provides improved image quality at half the acquisition time[J]. Quant Imaging Med Surg, 2024, 14(5): 3534-3543. DOI: 10.21037/qims-23-1488.
[24]
LU S S, QI M, ZHANG X, et al. Clinical evaluation of highly accelerated compressed sensing time-of-flight MR angiography for intracranial arterial stenosis[J]. AJNR Am J Neuroradiol, 2018, 39(10): 1833-1838. DOI: 10.3174/ajnr.A5786.
[25]
DEMERATH T, BONATI L, MEKABATY A E, et al. High-resolution compressed-sensing time-of-flight MRA in a case of acute ICA/MCA dissection[J]. Neuroradiology, 2020, 62(6): 753-756. DOI: 10.1007/s00234-020-02395-y.
[26]
LEINER T, HABETS J, VERSLUIS B, et al. Subtractionless first-pass single contrast medium dose peripheral MR angiography using two-point Dixon fat suppression[J]. Eur Radiol, 2013, 23(8): 2228-2235. DOI: 10.1007/s00330-013-2833-y.
[27]
IKEDA H, OHNO Y, MURAYAMA K, et al. Compressed sensing and parallel imaging accelerated T2 FSE sequence for head and neck MR imaging: comparison of its utility in routine clinical practice[J/OL]. Eur J Radiol, 2021, 135: 109501 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/33395594/. DOI: 10.1016/j.ejrad.2020.109501.
[28]
HAMILTON J, FRANSON D, SEIBERLICH N. Recent advances in parallel imaging for MRI[J]. Prog Nucl Magn Reson Spectrosc, 2017, 101: 71-95. DOI: 10.1016/j.pnmrs.2017.04.002.
[29]
MUSSARD E, HILBERT T, FORMAN C, et al. Accelerated MP2RAGE imaging using Cartesian phyllotaxis readout and compressed sensing reconstruction[J]. Magn Reson Med, 2020, 84(4): 1881-1894. DOI: 10.1002/mrm.28244.
[30]
李贝贝, 张浩南, 方鑫, 等. 不同压缩感知加速因子对头部3D-T1WI成像质量及基于体素的形态测量定量参数的影响[J]. 中国医学影像技术, 2022, 38(11): 1730-1734. DOI: 10.13929/j.issn.1003-3289.2022.11.031.
LI B B, ZHANG H N, FANG X, et al. Impacts of compressed sensing acceleration factors on imaging quality of head 3D-T1WI and voxel-based morphometry quantitative parameters[J]. Chin J Med Imag Technol, 2022, 38(11): 1730-1734. DOI: 10.13929/j.issn.1003-3289.2022.11.031.
[31]
VIDYA SHANKAR R, HU H H, BIKKAMANE JAYADEV N, et al. 2-D magnetic resonance spectroscopic imaging of the pediatric brain using compressed sensing[J]. Pediatr Radiol, 2019, 49(13): 1798-1808. DOI: 10.1007/s00247-019-04495-1.
[32]
RIEDERER S J, STINSON E G, WEAVERS P T. Technical aspects of contrast-enhanced MR angiography: current status and new applications[J]. Magn Reson Med Sci, 2018, 17(1): 3-12. DOI: 10.2463/mrms.rev.2017-0053.
[33]
MIZOSHIRI T, YOSHIDA M, ODA S, et al. Non-contrast mDixon MR angiography of the neck: comparison with time-of-flight MR angiography in normal subjects[J/OL]. Medicine, 2021, 100(51): e28351 [2024-06-30]. https://pubmed.ncbi.nlm.nih.gov/34941146/. DOI: 10.1097/MD.0000000000028351.
[34]
GOO E H, KIM S S. Evaluating compressed SENSE (CS) MRI metal artifact reduction using pig L-spine phantom and transplant patients: focused on the CS-SEMAC (SPIR), mDixon(O-MAR) and STIR techniques[J]. Tomography, 2022, 8(5): 2298-2312. DOI: 10.3390/tomography8050192.

上一篇 基于深度学习的T2 Flair序列提升白质高信号图像质量的价值
下一篇 5.0 T磁共振呼吸触发和单次屏气两种胰胆管成像序列价值比较
  
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