分享:
分享到微信朋友圈
X
技术研究
MAGiC序列应用于青年志愿者骶髂关节扫描的可行性研究
占颖莺 姜云萍 张珂 李文娟 潘洁琳 方义杰 李绍林 洪国斌

Cite this article as: Zhan YY, Jiang YP, Zhang K, et al. Feasibility study on application of MAGiC sequence in sacroiliac joint of young volunteers. Chin J Magn Reson Imaging, 2020, 11(7): 568-572.本文引用格式:占颖莺,姜云萍,张珂,等. MAGiC序列应用于青年志愿者骶髂关节扫描的可行性研究.磁共振成像, 2020, 11(7): 568-572. DOI:10.12015/issn.1674-8034.2020.07.018.


[摘要] 目的 探讨MR集合序列(magnetic resonance image compilation,MAGiC)用于青年志愿者骶髂关节扫描的可行性。材料与方法 30名青年志愿者分别行骶髂关节MRI常规序列和MAGiC序列扫描。首先由2名研究者分别对常规序列(T1WI-FSE、T2WI-FSE、STIR)和MAGiC序列(T1WI、T2WI、STIR)图像质量进行评分,采用Kappa检验评价一致性。然后测量信号噪声比(signal noise ratio,SNR)和对比噪声比(carrier to noise ratio,CNR);最后利用MAGiC T2-mapping序列测量骶髂关节面下骨髓T2值。结果 60侧骶髂关节纳入本研究。常规序列和MAGiC序列图像质量评分有显著差异(P<0.05);观察者间一致性检验Kappa> 0.60。MAGiC序列图像的SNR和CNR均优于常规序列(P<0.05)。MAGiC T2-mapping测得骶髂关节面下骨质的T2值(83±8) ms,且髂侧稍高于骶侧(P<0.05)。结论 青年志愿者骶髂关节MAGiC序列的图像质量评分较常规序列稍低,但SNR和CNR较优且能获得多组定量参数,可用于骶髂关节MR扫描。
[Abstract] Objective: To explore the feasibility of applying magnetic resonance image compilation (MAGiC) sequence to MR examination of sacroiliac joint in young volunteers.Materials and Methods: All participants (14 male, 16 female) underwent conventional sequences and MAGiC sequence scanning. Image quality, atifacts, the signal to noise ratio (SNR) and the carrier to noise ratio (CNR) were assessed independently by two observers (a radiologist and a radiologic technologist) on both conventional sequences (T1WI-FSE, T2WI-FSE, STIR) and MAGiC sequences (T1WI, T2WI, STIR). Kappa (K) statistic was used to evaluate the consistency of observers. T2 value of subchondral bone were measured on MAGiC T2-mapping images.Results: Thirty young volunteers were performed MR examination (60 sacroiliac joints included). The image quality scores of conventional sequences T1WI, T2WI and STIR were 5 (0), 5 (0) and 5 (0), which of MAGiC reconstruction sequences were 5 (1), 5 (1) and 3 (2), with statistically significant differences (P<0.05). The consistency check between observers showed K> 0.61. The SNR of T1WI sequence and MAGiC T1WI sequence were 12.61±0.37, 13.96±0.52 (t=-2.63, P< 0.05), the CNR of two sequences were 1.06±0.04, 1.18±0.28 (Z=-2.373, P<0.05). The SNR of T2WI sequence and MAGiC T2WI sequence were 12.84±0.48, 112.63±0.45 (t=0.447, P>0.05), the CNR of two sequences were 1.57±0.05, 1.54 (0.03)(Z=-1.486, P>0.05). The SNR of STIR sequence and MAGiC STIR sequence were 8.36 (3.77), 5.73±0.52 (Z=-3.19, P<0.05), the CNR of two sequences were-1.9 (10.29), 0.95±0.08 (Z=-1.46, P>0.05). There was no statistical difference of artifact score between conventional sequences and MAGiC sequences [conventional sequences: 50 cases (83.3%, grade 1), 8 cases (13.3%, grade 2), 2 cases (3.3%, grade 3). MAGiC sequences: 34 cases (56.6%, grade 1), 18 cases (30%, grade 2), 8 cases (13.3%, grade 3), Z=-4.443, P<0.001]. The mean T2 value was (83±8) ms, and the iliac side was slightly higher than sacral side (P<0.05).Conclusions: Though the subjective image quality score of MAGiC sequence images are slightly lower than conventional sequences images, but with higher SNR and CNR, shorter scanning time, more quantitative parameters, MAGiC sequence can be used for MR examination of sacroiliac joint in young volunteers.
[关键词] 磁共振成像;骶髂关节;诊断技术和方法
[Keywords] magnetic resonance imaging;sacroiliac joint;diagnostic techniques and procedures

占颖莺 中山大学附属第五医院放射科,珠海 519000

姜云萍 中山大学附属第五医院放射科,珠海 519000

张珂 中山大学附属第五医院放射科,珠海 519000

李文娟 中山大学附属第五医院放射科,珠海 519000

潘洁琳 中山大学附属第五医院放射科,珠海 519000

方义杰 中山大学附属第五医院放射科,珠海 519000

李绍林 中山大学附属第五医院放射科,珠海 519000

洪国斌* 中山大学附属第五医院放射科,珠海 519000

通信作者:洪国斌,E-mail:honggb@mail.sysu.edu.cn

利益冲突:无


收稿日期:2020-02-28
接受日期:2020-05-21
中图分类号:R445.2; R323.43 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2020.07.018
本文引用格式:占颖莺,姜云萍,张珂,等. MAGiC序列应用于青年志愿者骶髂关节扫描的可行性研究.磁共振成像, 2020, 11(7): 568-572. DOI:10.12015/issn.1674-8034.2020.07.018.

       MR集合序列(magnetic resonance image compilation,MAGiC)基于多个延迟多回波(multiple-delay multiple-echo ,MDME),在不同TR内施加4个120饱和脉冲,同时进行双回波采集,通过2个TE和4个TI的组合实现一次扫描生成8组对比图像及3组定量图像,是一种全新MR集合序列,且能任意调节TR(repetition time)、TE (echo time)、TI (inversion time)[1,2,3,4]。MAGiC序列扫描时间短、所获图像信息完整,目前在临床应用日趋增多,多集中在脑部[5,6,7]和乳腺[4]等,在肌骨系统的应用相对少见,在骶髂关节的应用则未见报道。本研究首次将MAGiC技术用于青年志愿者骶髂关节扫描,探索其应用可行性。

1 材料与方法

1.1 一般资料

       前瞻性搜集青年健康志愿者30名,其中男14名,女16名,年龄19~25 (22±1)岁,平均BMI为20.36±0.5。

       纳入标准:无慢性腰背部疼痛史、无外周关节痛病史、无风湿及类风湿病史、无相关骨髓疾病病史、无肿瘤病史、近期无腰背部及骨盆外伤、感染史,无腰背部手术史;排除标准:体重大于80 kg、幽闭恐惧症等磁共振检查禁忌证。

       本研究符合中山大学伦理学规范,所有被检者均知情同意并自愿参加。

1.2 成像序列与参数

       采用GE signa pioneer 3.0 T超导型MR扫描仪,16通道腹部相控阵线圈及脊柱线圈。受检者仰卧位,头先进,定位中心对准线圈中心及两侧髂前上棘连线中点,平行于骶1~骶3椎体背侧扫描斜冠状位T1WI-FSE、T2WI-FSE、T2WI-FSE-FS、MAGiC图像,具体扫描参数见表1

表1  MR扫描参数表
Tab. 1  Scanning parameters

1.3 图像处理与分析

1.3.1 图像质量主观评分

       对常规序列(T1WI-FSE、T2WI-FSE、STIR)及MAGiC序列(T1WI、MAGiC T2WI、MAGiC STIR)图像质量进行评分。由2名研究者(1名放射技师,从事影像技术工作12年;1名放射医师,从事影像诊断工作5年),采用盲法按照以下标准分别进行评分(5点李克特型量表)[5],一周后采用同样标准再次评分,比较组内及组间评分一致性。

       评分标准:5分,优秀(图像清晰);4分,良好(图像较清晰,适用于诊断);3分,一般(图像质量一般,基本满足诊断要求);2分,较差(图像无法满足诊断,但在特殊情况下可用于提供参考意见);1分,差(图像不能用于诊断)。

1.3.2 图像SNR和CNR测量

       医师A在GE AW4.7后处理工作站,选择双侧骶髂关节最佳显示层面及无伪影背景的层面和同层面腰大肌,画出骨髓和肌肉的感兴趣区(region of interests,ROI) ,ROI选择圆形且面积均为100 mm2,骨髓ROI置于右侧骶椎近关节面处,SNR和CNR计算公式为:SNR骨髓=I骨髓/δ骨髓、CNR=(I骨髓-I肌肉)/δ平均。其中I为信号值,δ骨髓为ROI骨髓的标准差,δ平均为ROI骨髓与肌肉标准差的均值[8,9]

1.3.3 图像伪影评价

       对常规序列(T1WI-FSE、T2WI-FSE、STIR)及MAGiC序列(T1WI、MAGiC T2WI、MAGiC STIR)图像伪影进行评估[3],伪影分级标准:1级,无伪影;2级,有伪影但不易发现;3级,轻微伪影;4级,明显伪影;5级,伪影严重。

1.3.4 骶髂关节面下骨质T2值测量

       在GE主台MAGiC处理软件包测量骶髂关节软骨下骨髓T2值。人工手动勾画ROI ,骶髂关节分为骶侧、髂侧及上中下共12个区域,如图1所示。ROI尽量靠近但不接触软骨,避开硬化区、脂肪沉积区、血管、囊变以及伪影等,分别测量每个区域骨髓的T2值[10,11]

图1  骶髂关节分区T2值测量示意图。骶髂关节分为左右骶侧和髂侧,每侧平均分为上中下份,共12个区域
图2  各扫描序列MR图像(A~F)。A、C、E为常规序列T1WI-FSE、T2WI-FSE、STIR;B、D、F为MAGiC重建序列T1WI、MAGiC T2WI、MAGiC STIR
Fig. 1  T2 value measurement diagram of the sacroiliac joint. The sacroiliac joint was divided into both left and right sacral and iliac sides, and each side was divided into the upper middle and lower parts, totaly 12 regions.
Fig. 2  MR images of each scanning sequence (A—F). Figure A, C, E were T1WI-FSE, T2WI-FSE, STIR which belongs to the conventional sequences, and B, D, F were MAGiC T1WI, MAGiC T2WI, MAGiC STIR which belongs to the reconstruction sequences.

1.4 统计学分析

       采用SPSS 22.0统计分析软件。符合正态分布的计量资料以±s表示,偏态分布的计量资料以M (QI)表示,等级资料以中位数±四分位数间距表示。采用Friedman检验比较所有序列SNR和CNR的差异及MAGiC上中下部位骨髓定量值的差异,P<0.01有统计学意义。采用配对t检验比较T1WI-FSE和MAGiC T1WI的SNR,采用Wilcoxon秩和检验比较图像质量评分、T1WI-FSE和MAGiC T1WI的CNR、STIR和MAGiC STIR的SNR和CNR差异、MAGiC内外侧骨髓定量值的差异,P<0.05差异有统计学意义。采用kappa检验比较组间及组内主观评分的一致性,kappa> 0.81为一致性好,0.61~0.81为一致性较好,0.41~0.60为一致性中等,0.21~0.40为一致性一般,0.0~0.20为一致性差。

2 结果

2.1 图像质量主观评分和观察者间一致性检验结果

       2名研究者分别对常规序列和MAGiC序列图像(图2)的主观评分结果如表2所示。常规图像主观评分较MAGiC序列图像稍高(P<0.001);2名观察者的组内和组间一致性均较好(P>0.61)(图3)。

图3  2名研究者图像质量主观评分一致性检验结果。图3A、B为2名观察者组内一致性结果,图3C、D为2名观察者两次评分组间一致性结果
Fig. 3  The intergroup and intra-group comparisons of observer A and B about image quality. Figure 3A and B are the consistent results within the two observer groups, while Figure 3C and D are the consistent results between the two observer groups.
表2  2名观察者对图像质量评分结果[M (上下四分位数间距)]
Tab. 2  Results of two observers (A, B) image quality scores (Median±IQR)

2.2 图像SNR和CNR测量结果

       常规序列T1WI和MAGiC序列T1WI的SNR分别为12.61±0.37、13.96±0.52 (t=-2.63 ,P <0.05) ,CNR分别为1.06±0.04、1.18±0.28 (Z=-2.373,P<0.05)。常规序列T2WI和MAGiC序列T2WI的SNR分别为12.84±0.48、112.63±0.45 (t=0.447,P>0.05),CNR分别为1.57±0.05、1.54 (0.03)(Z=-1.486,P>0.05)。常规序列STIR和MAGiC序列STIR的SNR分别为8.36(3.77)、5.73±0.52 (Z=-3.19,P<0.05);CNR分别为-1.9 (10.29)、0.95±0.08 (Z=-1.46,P>0.05)。对比常规序列与常规序列图像SNR和CNR,如图4所示。

图4  常规序列和MAGiC重建序列SNR和CNR对比。图A、B分别为T1WI-FSE和MAGiC T1WI的SNR和CNR对比;图C、D分别为T2WI-FSE和MAGiC T2WI的SNR和CNR对比;图E、F分别为STIR和MAGiC STIR图像的SNR、CNR对比
图5  骶髂关节面下骨髓T2值测量结果。图A为骶髂关节骶、髂侧MAGiC T2值对比;图B为骶髂关节上、中、下部分MAGiC T2值对比
Fig. 4  Comparison between normal sequence and MAGiC reconstruction sequence Figure A and B were the comparison of SNR and CNR of T1WI-fse and MAGiC T1WI. Figure C and D showed the SNR and CNR comparisons of T2WI-FSE and MAGiC T2WI. Figure E and F were the SNR and CNR comparisons of STIR and MAGiC STIR images.
Fig. 5  Measurement results of bone marrow T2 under the sacroiliac joint surface. Figure 5A, Comparison of the sacral and iliac sides MAGiC T2 values of the sacroiliac joint. 5B, comparison of the upper, middle and lower parts of the sacroiliac joint MAGiC T2 values.

2.3 图像伪影评价结果

       2名研究者对共60例次图像伪影进行分级,常规序列1级50例次(83.3%)、2级8例次(13.3%)、3级2例次(3.3%);MAGiC重建序列1级34例次(56.6%)、2级18例次(30%)、3级8例次(13.3%),差异有统计学意义(Z= -4.443,P<0.001),常规序列和MAGiC重组序列均无伪影分级为4、5级图像。

2.4 骶髂关节面下骨髓T2值测量结果

       采用MAGiC T2-mapping序列测得骶髂关节面下骨髓T2值为(83±8) ms,其中髂侧(83±7) ms,骶侧(82±9) ms,髂侧高于骶侧(P<0.001)。骶髂关节面下上、中、下3个区域的T2值分别为(81.9±0.56) ms ,(83±8) ms和(82±8.5) ms(图5),差异无统计学意义(P=0.076)。

3 讨论

3.1 MAGiC序列

       MAGiC序列作为一种全新MR集合序列,近年来在临床使用日趋增多,然而在肌骨系统的应用相对少见,尚未见骶髂关节的应用报道。本研究首次报道MAGiC技术应用于青年志愿者骶髂关节扫描,初步结果显示:MAGiC序列用于青年志愿者骶髂关节扫描的图像质量主观评分较常规序列稍低,但SNR和CNR优于常规序列,且总扫描时间短,更能同时获得多组定量参数,值得临床进一步深入研究。

3.2 图像主观与客观评分

       MAGiC序列的图像主观评分低于常规序列,我们分析可能的原因有:MAGiC应用于骶髂关节的扫描参数有待进一步优化,样本量相对偏少,同时可能与研究者的个体偏倚有一定关系。本研究用MAGiC STIR与常规STIR进行图像质量对比,MAGiC STIR序列的CNR与常规STIR没有统计学差异,但其图像主观评分和SNR均不如常规STIR序列,这可能是由于MAGiC合成STIR图像中脂肪抑制效果较差[12,13]。STIR序列对磁场不均匀性不敏感[14]

3.3 MAGiC T2-mapping定量图

       本研究MAGiC序列扫描时间8 min 8 s,一次扫描除了获得8组对比图像,更获得3组定量图像用于进一步量化分析,扫描时间显著低于获得同样信息的常规序列和常规T2-mapping之和(21 min 24 s)。MAGiC序列由两个数据采集阶段完成,(1)饱和阶段:层面选择饱和脉冲使一个层面饱和;(2)采集阶段:由另一层面的层面选择自旋回波进行采集。测量这些采集的多回波得到横向弛豫时间T2,由饱和脉冲定量得到纵向弛豫时间T1,计算出射频场大小B1。通过T1、T2、B1计算出磁化矢量值M0及PD (proton density,质子密度)定量图。通过MAGiC后处理软件,获得任意TR、TE、TI时间的合成图像[7,15,16,17]。本研究利用MAGiC T2-mapping图对骶髂关节面下骨髓T2定量值做了初步研究,发现髂侧T2值稍高于骶侧(P<0.05);上、中、下份对比没有统计学差异(P>0.05)。姚晓龙等[18]通过T2 mapping序列对早期骶髂关节炎患者骶侧和髂侧T2定量值,结论与本研究相似。究其原因,可能与髂侧骨质密度较大有关。

       本研究存在如下不足:样本量较少,且均为青年志愿者,年龄层较局限。其次,MAGiC缺少合成T2WI-FSE-FS的模板,有待新版本改进。再次,我们仅对MAGiC序列进行图像质量研究,下一步将评价MAGiC序列在AS患者的诊断效能。

       综上所述,MAGiC序列用于青年志愿者骶髂关节扫描的图像质量主观评分较常规序列稍低,但SNR和CNR优于常规序列,且总扫描时间短,更能同时获得多组定量图用于进一步量化分析。

[1]
Liu HM, Yin GP, Bie F, et al. Comparison of head MR image quality between set sequence and conventional sequence. Chin Med Imaging Technol, 2019, 35(2): 268-271.
刘辉明,尹国平,别非,等.对比集合序列与常规序列头部MR图像质量.中国医学影像技术, 2019, 35(2): 268-271.
[2]
Duchaussoy T, Budzik JF, Norberciak L, et al. Synthetic T2 mapping is correlated with time from stroke onset: a future tool in wake-up stroke management. Eur Radiol, 2019, 29(12): 7019-7026.
[3]
Zheng ZF, Zhang DP, Mao L, et al. Comparative study on the image quality of integrated craniocerebral MRI and conventional MRI. Magnetic resonance imaging, 2019, 10(5): 361-365.
郑作锋,张东坡,毛磊,等.颅脑集成MRI与常规MRI图像质量的对比研究.磁共振成像, 2019, 10(5): 361-365
[4]
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. Br J Radiol, 2018, 92(1093): 20180479.
[5]
Tanenbaum LN, Tsiouris AJ, Johnson AN, et al. Synthetic MRI for clinical neuroimaging: results of the magnetic resonance image compilation (MAGiC) prospective, multicenter, multireader trial. AJNR Am J Neuroradiol, 2017, 38(6): 1103-1110.
[6]
Lee SM, Choi HY, Cheon JE, et al. Image quality at synthetic brain magnetic resonanceimaging in children. Pediatr Radiol, 2017, 47(12): 1638-1647.
[7]
Hagiwara A, Hori M, Cohen-Adad J, et al. Linearity, bias, intrascanner repeatability, and interscanner reproducibility of quantitative multidynamic multiecho sequence for rapid simultaneous relaxometry at 3 T: a validation study with a standardized phantom and healthy controls. Invest Radiol, 2019, 54(1): 39-47.
[8]
Greese J, Diekhoff T, Sieper J, et al. Detection of sacroiliitis by short-tau inversion recovery and T2-weighted turbo spin echo sequences: results from the SIMACT study. J Rheumatol, 2019, 46(4): 376-383.
[9]
Di Giuliano F, Minosse S, Picchi E, et al. Comparison between synthetic and conventional magnetic resonance imaging in patients with multiple sclerosis and controls. MAGMA, 2019, 28: 804-809.
[10]
Li F. Study on the value of T2* mapping of sacroiliac joint in the evaluation of AS patients. Zhengzhou: First Affiliated Hospital of Zhengzhou University, 2017.
李飞.骶髂关节T2* mapping成像在AS患者病情评估方面的价值研究.郑州:郑州大学第一附属医院, 2017.
[11]
Albano D, Chianca V, Cuocolo R, et al. T2-mapping of the sacroiliac joints at 1.5 Tesla: a feasibility and reproducibility study. Skeletal Radiol, 2018, 47(12): 1691-1696.
[12]
Vargas MI, Drake-Pérez M, Delattre BMA. Feasibility of a Synthetic MR imaging sequence for spine imaging. AJNR Am J Neuroradiol, 2018, 39(9): 1756-1763.
[13]
Yi J, Lee YH, Song HT, et al. Double-inversion recovery with synthetic magnetic resonance: a pilot study for assessing synovitis of the knee joint compared to contrast-enhanced magnetic resonance imaging. Eur Radiol, 2019, 29(5): 2573-2580.
[14]
Sung S, Kim HS, Kwon JW. MRI assessment of sacroiliitis for the diagnosis of axial spondyloarthropathy: comparison of fat-saturated T2, STIR and contrast-enhancedsequences. Br J Radiol, 2017, 90(1078): 20170090.
[15]
Kang KM, Choi SH, Kim H, et al. The effect of varying slice thickness and interslice gap on T1 and T2 measured with the multidynamic multiecho sequence. Magn Reson Med Sci, 2019, 18(2): 126-133.
[16]
Kang N, Zu DL, Zhang HJ. Spin density p, relaxation time T1 and T2 quantitative magnetic resonance imaging. Chin Med Imaging Technol, 2004, 20 (12): 1944-1947.
康宁,俎栋林,张宏杰.自旋密度p、弛豫时间T1和T2定量磁共振成像.中国医学影像技术, 2004, 20(12): 1944-1947.
[17]
Drake-Pérez M, Delattre B, Boto J, et al. Normal values of magnetic relaxation parameters of spine components with the synthetic MRI sequence. AJNR Am J Neuroradiol, 2018, 39(4): 788-795.
[18]
Yao XL, Yao J, Guo H, et al. Application value of T2mapping sequence in sacroiliac joint lesions in early ankylosis. Chin J Osteoporosis, 2016, 22(3): 348-350, 357.
姚晓龙,姚娟,郭辉,等. T2mapping序列在早期强直性脊柱炎骶髂关节病变中的应用价值.中国骨质疏松杂志, 2016, 22(3): 348-350, 357.

上一篇 联合脑血流速和磁敏感成像研究定向视觉疲劳视区改变
下一篇 DWI联合MRS对脑胶质瘤分级诊断的价值
  
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