Proc. SPIE. 10134, Medical Imaging 2017: Computer-Aided Diagnosis
KEYWORDS: Signal to noise ratio, Liver, Tissues, Magnetic resonance imaging, Image segmentation, Wave propagation, Elastography, Algorithm development, 3D image processing, Magnetic resonance elastography
Magnetic Resonance Elastography (MRE) is a phase-contrast MRI technique which calculates quantitative stiffness images, called elastograms, by imaging the propagation of acoustic waves in tissues. It is used clinically to diagnose liver fibrosis. Automated analysis of MRE is difficult as the corresponding MRI magnitude images (which contain anatomical information) are affected by intensity inhomogeneity, motion artifact, and poor tissue- and edge-contrast. Additionally, areas with low wave amplitude must be excluded. An automated algorithm has already been successfully developed and validated for clinical 2D MRE. 3D MRE acquires substantially more data and, due to accelerated acquisition, has exacerbated image artifacts. Also, the current 3D MRE processing does not yield a confidence map to indicate MRE wave quality and guide ROI selection, as is the case in 2D. In this study, extension of the 2D automated method, with a simple wave-amplitude metric, was developed and validated against an expert reader in a set of 57 patient exams with both 2D and 3D MRE. The stiffness discrepancy with the expert for 3D MRE was -0.8% ± 9.45% and was better than discrepancy with the same reader for 2D MRE (-3.2% ± 10.43%), and better than the inter-reader discrepancy observed in previous studies. There were no automated processing failures in this dataset. Thus, the automated liver elasticity calculation (ALEC) algorithm is able to calculate stiffness from 3D MRE data with minimal bias and good precision, while enabling stiffness measurements to be fully reproducible and to be easily performed on the large 3D MRE datasets.
Magnetic Resonance Elastography (MRE) is an MRI-based technique that is used for the clinical diagnosis and staging
of liver fibrosis by quantitatively measuring the stiffness of the liver. Due to the complexity of the signal characteristics
and the presence of artifacts both in the acquired images and in the resulting stiffness images, the selection of the ROI for the stiffness measurement is currently performed manually, which may lead to significant inter- and intrareader
variability. An algorithm has been developed to fully automate this analysis for liver MRE images. Automated segmentation of liver MRE images is challenging due to signal inhomogeneity, low contrast, and variability in patient anatomy. An initial liver contour is found by fitting Gaussian peaks to the image histogram and selecting the
peak that comprises intensities in the expected range and produces a mask near the expected location of the liver. After correction to reduce intensity inhomogeneity, an active contour based on intensity, with morphology used to implicitly enforce smoothness, is used to segment liver tissue while avoiding blood vessels. The resulting mask is used to initialize another segmentation which splits the region of the elastogram belonging to the liver into homogeneous liver tissue and areas with inclusions, partial volume effects, and artifacts. In a set of 88 cases the algorithm had a -6.0 ± 14.2% stiffness difference from an experienced reader, which was superior to the 6.8 ± 22.8% difference between two readers. The segmentation was run on an additional 200 cases and the final ROIs were subjectively rated by a radiologist. The ROIs in 98% of cases received an average rating of “good” or “acceptable.”