From the Departments of Radiology (S.S., G.K., B.L.)
Department of Biomedical Engineering (Z.L., Y.W.), Cornell University, New York, New York.
AJNR Am J Neuroradiol. 2018 Apr;39(4):648-653. doi: 10.3174/ajnr.A5550. Epub 2018 Feb 22.
Identifying cerebral microhemorrhage burden can aid in the diagnosis and management of traumatic brain injury, stroke, hypertension, and cerebral amyloid angiopathy. MR imaging susceptibility-based methods are more sensitive than CT for detecting cerebral microhemorrhage, but methods other than quantitative susceptibility mapping provide results that vary with field strength and TE, require additional phase maps to distinguish blood from calcification, and depict cerebral microhemorrhages as bloom artifacts. Quantitative susceptibility mapping provides universal quantification of tissue magnetic property without these constraints but traditionally requires a mask generated by skull-stripping, which can pose challenges at tissue interphases. We evaluated the preconditioned quantitative susceptibility mapping MR imaging method, which does not require skull-stripping, for improved depiction of brain parenchyma and pathology.
Fifty-six subjects underwent brain MR imaging with a 3D multiecho gradient recalled echo acquisition. Mask-based quantitative susceptibility mapping images were created using a commonly used mask-based quantitative susceptibility mapping method, and preconditioned quantitative susceptibility images were made using precondition-based total field inversion. All images were reviewed by a neuroradiologist and a radiology resident.
Ten subjects (18%), all with traumatic brain injury, demonstrated blood products on 3D gradient recalled echo imaging. All lesions were visible on preconditioned quantitative susceptibility mapping, while 6 were not visible on mask-based quantitative susceptibility mapping. Thirty-one subjects (55%) demonstrated brain parenchyma and/or lesions that were visible on preconditioned quantitative susceptibility mapping but not on mask-based quantitative susceptibility mapping. Six subjects (11%) demonstrated pons artifacts on preconditioned quantitative susceptibility mapping and mask-based quantitative susceptibility mapping; they were worse on preconditioned quantitative susceptibility mapping.
Preconditioned quantitative susceptibility mapping MR imaging can bring the benefits of quantitative susceptibility mapping imaging to clinical practice without the limitations of mask-based quantitative susceptibility mapping, especially for evaluating cerebral microhemorrhage-associated pathologies, such as traumatic brain injury.
识别脑微出血负担有助于诊断和治疗创伤性脑损伤、中风、高血压和脑淀粉样血管病。基于磁共振成像磁敏感性的方法比 CT 更敏感地检测脑微出血，但除定量磁敏感图外的其他方法提供的结果因场强和 TE 而异，需要额外的相位图来区分血液和钙化，并将脑微出血描绘为晕影伪影。定量磁敏感图提供了组织磁特性的通用量化，而没有这些限制，但传统上需要通过颅骨剥离生成的蒙版，这在组织界面处可能会带来挑战。我们评估了不需要颅骨剥离的预处理定量磁敏感图磁共振成像方法，以改善脑实质和病理学的描绘。
56 例患者进行了 3D 多回波梯度回波采集的脑部磁共振成像。使用常用的基于蒙版的定量磁敏感图方法创建基于蒙版的定量磁敏感图图像，使用基于预条件的总场反转制作预处理定量磁敏感图。所有图像均由神经放射科医师和放射科住院医师进行审查。
10 例（18%）受试者均有创伤性脑损伤，3D 梯度回波成像上显示有血液产物。所有病变在预处理定量磁敏感图上均可见，而在基于蒙版的定量磁敏感图上则有 6 个不可见。31 例（55%）受试者在预处理定量磁敏感图上可见脑实质和/或病变，但在基于蒙版的定量磁敏感图上不可见。6 例（11%）受试者在预处理定量磁敏感图和基于蒙版的定量磁敏感图上均出现脑桥伪影；预处理定量磁敏感图上的伪影更严重。
预处理定量磁敏感图磁共振成像可以在不限制基于蒙版的定量磁敏感图的情况下，将定量磁敏感图成像的优势引入临床实践，特别是在评估与脑微出血相关的病变，如创伤性脑损伤。
