T2-weighted Liver Mri at 3t Using a Blade Technique: Comparison with a Standard Rectilinear T2-weighted Sequence for Image Quality and Lesion Detection

Introduction: The quality of T2-weighted images (T2WI) of the liver in the clinical setting is often degraded by ghosting artifact from various sources of motion. This artifact can obscure the margins of the liver, intra-hepatic vessels, and focal liver lesions, lowering diagnostic utility. One study has noted that such motion within the liver is more severe at 3T compared with at 1.5T [1]. Furthermore, it is not uncommon for B1-inhomogeneity artifact at 3T to lead to prominent areas of signal void over the left lobe. As a result of these artifacts, T2WI of the liver at 3T using conventional sequences is of variable image quality. With the BLADE technique, k-space data is acquired as a series of short overlapping echo trains (blades), each comprising the lowest phase-encoding steps; these acquisitions rotate around the center of k-space in a periodic fashion, yielding central k-space data that is shared by all blades and that may be used to correct for in-plane translation and rotation. Several studies have demonstrated improved image quality for T2WI of the liver using BLADE compared with a conventional rectilinear technique, with one study showing improved accuracy for focal lesion detection [2,3,4]. However, these studies were all performed at 1.5T. Therefore, the purpose of this study was to compare T2WI of the liver at 3T using BLADE and standard rectilinear k-space schemes in terms of image quality and focal lesion detection.