Purpose:To improve the quality of images obtained via dynamic contrast-enhanced MRI (DCE-MRI) that include motion artifacts and blurring using a deep learning approach.Methods: A multi-channel convolutional neural network (MARC) based method is proposed for reducing the motion artifacts and blurring caused by respiratory motion in images obtained via DCE-MRI of the liver. The training datasets for the neural network included images with and without respiration-induced motion artifacts or blurring, and the distortions were generated by simulating the phase error in k-space. Patient studies were conducted using a multi-phase T1-weighted spoiled gradient echo sequence for the liver containing breath-hold failures during data acquisition. The trained network was applied to the acquired images to analyze the filtering performance, and the intensities and contrast ratios before and after denoising were compared via Bland-Altman plots.
Results:The proposed network was found to significantly reduce the magnitude of the artifacts and blurring induced by respiratory motion, and the contrast ratios of the images after processing via the network were consistent with those of the unprocessed images.
Purpose: We evaluated the use of magnetic resonance (MR) elastography (MRE) for staging liverˆbrosis in patients with chronic hepatitis C and compared the ability of MRE and serumˆbrosis markers for discriminating each stage ofˆbrosis.Methods: We evaluated 114 patients with chronic hepatitis C in whom the pathological brosis stage was determined (ˆbrosis stage 0 [F0], 3; F1, 15; F2, 28; F3, 25; and F4, 43). All patients underwent MRE using a 1.5-tesla MR system and pneumatic driver system. We measured stiŠness values (kPa) of the liver in a circular region of interest placed on elastograms. We determined the optimal cutoŠ value and diagnostic ability for discriminating each stage ofˆbrosis using receiver operating characteristic (ROC) curve analysis and compared the discriminative ability of MRE with that of serumˆbrosis markers. Conclusion: MRE is a reliable technique for staging liverˆbrosis and discriminating liver brosis stages in patients with chronic hepatitis C.
Purpose To investigate the potential of diffusion magnetic resonance (MR) imaging to provide quantitative estimates of tissue stiffness without using mechanical vibrations in patients with chronic liver diseases and to generate a new elasticity-driven intravoxel incoherent motion (IVIM) contrast. Materials and Methods This retrospective study, conducted from January to April 2016, was approved by an institutional review board that waived the requirement for informed consent. Fifteen subjects were included (13 men and two women; mean age ± standard deviation, 73 years ± 8). MR elastography and diffusion MR imaging were performed at 3 T. A search for an empirical relationship between MR elastographic shear modulus, µ, and a shifted apparent diffusion coefficient (sADC) was performed. The sADC was then inverted to estimate patient liver shear modulus directly from diffusion MR imaging signals. Results A significant correlation (r = 0.90, P = 1 · 10) was observed between µ and sADC calculated by using diffusion MR imaging signals acquired with b values of 200 (S) and 1500 (S ) sec/mm (sACD). On the basis of the relationship between the µ and sADC, a diffusion-based shear modulus, µ, could be estimated with the following equation: µ = (-9.8 ± 0.8) ln(S/S) + (14.0 ± 0.9). IVIM virtual elastograms also could be generated to reveal new contrast features in lesions, depending on pseudovibration frequency and amplitude. Conclusion Diffusion MR imaging, through a calibration of sADC with standard MR elastography, can be converted quantitatively into shear modulus without using mechanical vibrations to provide information on the degree of liver fibrosis; these virtual elastograms require only two b values to be acquired and processed. Propagating shear wave can also be emulated, leading to a new elasticity-driven IVIM contrast with ranges of virtual vibration frequencies and amplitudes not limited by MR elastography or MR imaging hardware capacities. RSNA, 2017 Online supplemental material is available for this article.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.