Wave-CAIPI susceptibility-weighted imaging achieves diagnostic performance comparable to conventional susceptibility-weighted imaging in half the scan time
Wave-CAIPI susceptibility-weighted imaging achieves diagnostic performance comparable to conventional susceptibility-weighted imaging in half the scan time
复制标题
DOI：
10.1007/s00330-019-06574-1
复制
发表时间：
2020-01-17
影响因子：
5.9
通讯作者：
Byun, Jun Soo
中科院分区：
文献类型：
--
作者：
Chung, Mi Sun;Lee, Eun Jung;Byun, Jun Soo
目的评价常规磁共振加权成像（SWI）和使用波控混叠并行成像（CAIPI）加速的快速SWI检测脑微出血（CMB）的一致性。我们还仔细检查了颅内病变的诊断一致性，并比较了两种序列之间的图像质量。方法本回顾性研究获得了机构审查委员会的批准，并放弃了知情同意。我们纳入了2017年9月至2017年11月期间使用传统SWI（扫描时间，251 s）和wave-CAIPI SWI（扫描时间，113 s）进行脑MRI的181例连续患者。由两名放射科医生独立审查所有图像，使用微出血解剖评定量表（MARS）检测和计数CMB。一名神经放射科医生诊断了颅内病变，并使用视觉分析对图像质量进行评分。计算CMB和颅内病变检测的一致性，并使用kappa和组内相关分析观察者间一致性。结果对于CMB的检测，常规和波形CAIPI SWI均显示出100%的CMB存在，而MARS则显示出94.5%的CMB存在。当按解剖位置划分时，Wave-CAIPI SWI实现了97%以上的MARS一致性，具有极好的一致性。观察者之间的协议也很好。颅内病变的诊断（28例患者的33处病变）显示了100%的一致性。两个序列的图像质量无显著差异（p = 0.20）。结论Wave-CAIPI SWI对颅内病变的检出和诊断与常规SWI相比具有较高的一致性，扫描时间缩短一半。
Objectives We aimed to evaluate the agreement in the detection of cerebral microbleeds (CMBs) between conventional susceptibility-weighted imaging (SWI) and fast SWI using wave-controlled aliasing in parallel imaging (CAIPI) acceleration. We also scrutinized the diagnostic agreement for intracranial lesions and compared the image quality between both sequences. Methods Institutional review board approval was obtained and informed consent was waived for this retrospective study. We included 181 consecutive patients who had undergone brain MRI with both conventional SWI (scan time, 251 s) and wave-CAIPI SWI (scan time, 113 s) from September 2017 to November 2017. All images were independently reviewed by two radiologists for the detection and counting of CMBs using the Microbleed Anatomical Rating Scale (MARS). One neuroradiologist diagnosed intracranial lesions and scored image quality using visual analysis. The agreement for detection of CMBs and intracranial lesions was calculated, and interobserver agreements were analyzed by using kappa and intraclass correlation. Results For detection of CMBs, both the conventional and wave-CAIPI SWI showed significantly high agreement of 100% for the presence of CMBs, and 94.5% using MARS. Wave-CAIPI SWI achieved more than 97% agreement of MARS when divided by anatomical locations, with excellent agreement. Interobserver agreements were also excellent. The diagnosis for intracranial lesions (33 lesions in 28 patients) demonstrated 100% agreement. The image quality of both sequences is not significantly different (p = 0.20). Conclusions Wave-CAIPI SWI achieved high agreement for CMB detection and diagnosis of intracranial lesions compared to conventional SWI within half of the scan time.