基于T1W-3D序列探讨光梭成像技术的临床应用价值  

The clinical application value of constellation shuttling imaging technology based on T1W-3D sequence

作　　者：马媛媛[1] 李彦[1] 张雪坤 严福华[1] 宋琦[1] MA Yuanyuan;LI Yan;ZHANG Xuekun;YAN Fuhua;SONG Qi(Department of Radiology,Ruijin Hospital,Shanghai Jiaotong University of Medicine,Shanghai 200025,China)

机构地区：[1]上海交通大学医学院附属瑞金医院放射科,上海200025

出　　处：《磁共振成像》2021年第11期52-56,共5页Chinese Journal of Magnetic Resonance Imaging

基　　金：上海科学技术委员会科研计划项目(编号:18DZ1930103)。

摘　　要：目的探讨光梭成像技术在头颅磁共振扫描中的临床应用价值。材料与方法2020年7月至8月,共招募无磁共振扫描禁忌证且健康成年志愿者20名,其中男10名,女10名,年龄22~69(40.2±13.3)岁。采用联影uMR7803.0 T超导磁共振扫描,24通道相控阵线圈,对志愿者进行脑部数据采集。具有主治医师水平以上的2名放射医生在不知道图像序列信息的情况下进行主观图像质量评估:基于视觉评估、综合信噪比、伪影、灰白质边界的清晰度等进行Likert 5级评分,数据结果进行Kappa检验。在解剖结构相应的灰质(gray matter,GM)和白质(white matter,WM)以及背景噪声区域放置感兴趣区(region of interest,ROI),对灰质、白质信号及噪声信号(noise signal)强度重复三次测量取平均值,计算信噪比(signal to noise ratio,SNR)和对比噪声比(contrast to noise ratio,CNR)。以上均采用SPSS 23.0版统计软件进行配对t检验,均以P<0.05为差异有统计学意义。结果纳入研究图像质量Likert 5级评分结果为4~5分,检验评分结果:Kappa值是0.801,P值为<0.001,认为2名医生图像质量评估结果在统计学意义上具有显著一致性。对于SNR和CNR进行t检验比较两组数据,结果两者之间信噪比差异有统计学意义,光梭成像技术组高于并行采集技术组。运用基于体素的形态计量学CAT-12软件分别得出脑实质总体积,灰质总体积,白质总体积及脑脊液总体积,进行配对t检验,结果P值分别为0.98,0.25,0.50,0.11,均大于0.05,两组不同扫描方式数据基于体素的形态计量学结果差异无统计学意义。光梭成像技术组扫描时间减少116 s,减少为40%。结论较传统的并行采集技术,光梭成像技术在头颅T1W-3D序列中具有更优的图像信噪比并可以明显缩短扫描时间,具有重要的临床应用价值。Objective:To explore the clinical application value of constellation shuttling imaging technology in cranial magnetic resonance scanning.Materials and Methods:Brain imaging of 20 healthy adult volunteers on a United Imaging uMR7803.0 T superconducting magnetic resonance,with 24-channel phased array coils,from July to August 2020,this study recruited without contraindications to magnetic resonance scanning,including 10 males and 10 females,aged 22—69(40.2±13.3)year old.Two radiologists with an attending physician level or above perform subjective image quality assessment without knowing the image sequence information:based on visual assessment,comprehensive signal-to-noise ratio,artifacts,and the definition of gray matter borders,they perform Likert 5-level scoring.The data results were tested by kappa.Place the region of interest(ROI)in the gray matter(GM,)and white matter(WM)and background noise regions corresponding to the anatomical structure,and repeat the intensity of the gray matter,white matter signal and noise signal(noise signal),take the average of the three measurements and calculate the SNR and CNR.Using the voxel-based morphometrics CAT-12 software to obtain the total brain parenchymal volume,total gray matter volume,total white matter volume,and total cerebrospinal fluid volume.All of the above uses the statistical software version SPSS 23.0 to perform Paired t test,and the difference is statistically significant with P<0.05.Results:The image quality of the included study Likert grade 5 score(4—5 points),the test score result:Kappa value is 0.801,P value is less than 0.001,it is considered that the image quality evaluation results of the two doctors are statistically consistent.For SNR and CNR,a t test was performed to compare the two sets of data.The results showed that the signal-to-noise ratio between the two was statistically different,and the light shuttle imaging technology group was higher than the parallel acquisition technology group.Using the voxel-based morphometrics CAT-12 software

关 键 词：磁共振成像 光梭成像技术 压缩感知技术 并行采集技术 全脑三维T1加权序列 

分 类 号：R445.2[医药卫生—影像医学与核医学] R651.l[医药卫生—诊断学]