Dzyubak, Bogdan

BackgroundNonalcoholic fatty liver disease (NAFLD) affects 25% of the global population. The standard of diagnosis, biopsy, is invasive and affected by sampling error and inter‐reader variability. We hypothesized that widely available rapid MRI techniques could be used to predict nonalcoholic steatohepatitis (NASH) noninvasively by measuring liver stiffness, with magnetic resonance elastography (MRE), and liver fat, with chemical shift‐encoded (CSE) MRI. Besides, we validate an automated image analysis technique to maximize the utility of these methods.PurposeTo implement and test an automated system for analyzing CSE‐MRI and MRE data coupled with model‐based prediction of NASH.Study TypeProspective.SubjectsEighty‐three patients with suspected NAFLD.Field Strength/SequenceA 1.5 T using a flow‐compensated motion‐encoded gradient echo MRE sequence and a multiecho CSE‐MRI sequence.AssessmentsThe MRE and CSE‐MRI data were analyzed by two readers (5+ and 1 years of experience) and an automated algorithm. A logistic regression model to predict pathology‐diagnosed NASH was trained based on stiffness and proton density fat fraction, and the area under the receiver operating characteristic curve (AUROC) was calculated using 10‐fold cross validation for models based on both automated and manual measurements. A separate model was trained to predict the NASH severity score (NAS).Statistical TestsPearson's correlation, Bland–Altman, AUROC, C‐statistic.ResultsThe agreement between automated measurements and the more experienced reader (R2 = 0.87 for stiffness and R2 = 0.99 for proton density fat fraction [PDFF]) was slightly better than the agreement between readers (R2 = 0.85 and 0.98). The model for predicting biopsy‐diagnosed NASH had an AUROC of 0.87. The NAS‐prediction model had a C‐statistic of 0.85.Data ConclusionWe demonstrated a workflow that used a limited MRI acquisition protocol and fully automated analysis to predict NASH with high accuracy. These methods show promise to provide a reliable noninvasive alternative to biopsy for NASH‐screening in populations with NAFLD.Level of Evidence2Technical Efficacy Stage2

PurposeTo assess the success rate, image quality, and the ability to stage liver fibrosis of a standard 2D gradient‐recalled echo (GRE) and four different spin‐echo (SE) magnetic resonance elastography (MRE) sequences in patients with different liver iron concentrations.Materials and MethodsA total of 332 patients who underwent 3T MRE examinations that included liver fat and iron quantification were enrolled, including 136 patients with all five MRE techniques. Thirty‐four patients had biopsy results for fibrosis staging. The liver stiffness, region of interest area, image quality, and success rate of the five sequences were compared in 115/136 patients. The area under the receiver operating characteristic curves (AUCs) and the accuracies for diagnosing early‐stage fibrosis and advanced fibrosis were compared. The effect of BMI (body mass index), the R2* relaxation time, and fat fraction on the image quality and liver stiffness measurements were analyzed.ResultsThe success rates were significantly higher in the four SE sequences (99.1–100%) compared with GRE MRE (85.3%) (all P < 0.001). There were significant differences of the mean ROI area between every pair of sequences (all P < 0.0001). There were no significant differences in the AUC of the five MRE sequences for discriminating advanced fibrosis (10 P‐values ranging from 0.2410–0.9171). R2* had a significant effect on the success rate and image quality for the noniron 2D echo‐planar imaging (EPI), 3D EPI and 2D GRE (all P < 0.001) sequences. BMI had a significant effect on the iron 2D EPI (P = 0.0230) and iron 2D SE (P = 0.0040) sequences.ConclusionAll five techniques showed good diagnostic performance in staging liver fibrosis. The SE MRE sequences had higher success rates and better image quality than GRE MRE in 3T clinical hepatic imaging.Level of Evidence: 3Technical Efficacy: Stage 5J. Magn. Reson. Imaging 2018;47:976–987.

AbstractBackground and Aim:Liver stiffness (LS) measured by magnetic resonance elastography (MRE) is emerging as an important biomarker in chronic liver diseases. We examined the diagnostic performance of MRE, factors associated with an increased LS and the prognostic value of LS as measured by MRE among patients with primary sclerosing cholangitis (PSC).Methods:We performed a retrospective review of 266 patients with PSC to examine whether LS was associated with the primary endpoint of hepatic decompensation (ascites, variceal hemorrhage and hepatic encephalopathy). The ability of MRE to differentiate stages of fibrosis was examined in a subset of patients who underwent a liver biopsy (n = 20).Results:An LS of 4.93 kPa was the optimal point to detected F4 fibrosis (sensitivity, 1.00; 95% confidence interval (CI), 0.40–1.00; specificity, 0.94; 95%CI, 0.68–1.00). While a serum alkaline phosphatase <1.5 times the upper limit of normal excluded the presence of advanced LS, it was not associated with the primary endpoint (hazard ratio, 0.26; 95%CI, 0.01–1.33). However, LS was associated with the development of decompensated liver disease (hazard ratio, 1.55; 95%CI, 1.41–1.70). The optimal LS thresholds that stratified patients at a low, medium and high risk for hepatic decompensation were <4.5, 4.5–6.0 and >6.0 kPa (respectively).Conclusion:Magnetic resonance elastography is able to detect cirrhosis with high specificity and an alkaline phosphatase <1.5 times the upper limit of normal makes the presence of advanced LS unlikely. Moreover, LS obtained by MRE is predictive of hepatic decompensation in PSC.

PurposeTo evaluate with magnetic resonance elastography (MRE) whether patients with constrictive pericarditis (CP) have increased hepatic stiffness. CP results in reduced pericardial compliance, ventricular interdependence, and right heart failure. Patients with untreated CP may develop liver fibrosis and ultimately cirrhosis due to chronic venous congestion. Chronic venous congestion ± fibrosis may lead to increased liver stiffness.Materials and MethodsProspectively, patients with suspected CP underwent 2D transthoracic echocardiography, cardiac MRI, and liver MRE. An automated method was used to draw regions of interest (ROIs) on the stiffness maps to calculate the mean liver stiffness in kilopascals (kPa). A t‐test with α = 0.05 was performed between stiffness values of patients with positive and negative CP findings based on previously published echocardiography criteria.ResultsNineteen patients met inclusion criteria with a mean ± standard deviation (SD) age of 51 ± 16 years. Nine patients (47%) had CP. Mean liver stiffness trended higher in patients with CP compared to those without CP (4.04 kPa vs. 2.46; P = 0.045). Liver stiffness correlated with MRI septal bounce (P = 0.04), inferior vena cava size (P = 0.003), echo abnormal septal motion (P = 0.04), and echo mitral inflow variation >25% (P = 0.02). Only MRI septal bounce predicted CP by echocardiography (P < 0.001).ConclusionCP was associated with increased liver stiffness. The increased stiffness is most likely secondary to chronic hepatic venous congestion and/or fibrosis. MRE may be useful for noninvasive liver stiffness assessment in CP. J. Magn. Reson. Imaging 2016;44:81–88.

PurposeMR elastography (MRE) is a phase‐contrast MRI technique that is used to quantitatively assess liver stiffness for staging hepatic fibrosis. The current approach requires manual selection of a region of interest (ROI) with good wave quality from which to measure stiffness. The purpose of this work was to develop and evaluate a fully automated approach for measuring hepatic stiffness from MRE images to further reduce measurement variability.Materials and MethodsAn automated liver elasticity calculation (ALEC) algorithm was developed to address reader stiffness measurement variability. ALEC has three stages: initial tissue estimation, segmentation, and ROI cleanup. Stiffnesses measured by the algorithm were compared with technicians and an expert radiologist in a set of 121 clinical cases acquired at 1.5 Tesla. Intra‐class correlation (ICC), Bland‐Altman analysis, and a noninferiority test were performed to evaluate whether the algorithm can be used in place of manual analysis by technicians.ResultsThe stiffness measurement difference with the expert was 1.42% ± 11.17% (mean ± standard deviation) for the algorithm and 1.82% ± 13.65% for the technicians. The ICCs were 0.981 and 0.984, respectively. Both the algorithm and technicians were equivalent to the expert within a 5% significance margin (P < 0.01). The algorithm had no failures in the 119 cases that were considered analyzable by the human readers.ConclusionThe results of this study show that the newly developed automated algorithm is able to measure stiffness in clinical liver MRE exams with an accuracy that is equivalent to that of an expert radiologist. ALEC may be useful for analysis of archived data and suitable for performing multi‐center studies. J. Magn. Reson. Imaging 2016;43:1055–1063.

PurposeTo provide a fully automated algorithm for obtaining stiffness measurements from hepatic magnetic resonance elastography (MRE) images that are consistent with measurements performed by expert readers.Materials and MethodsAn initial liver contour was found using an adaptive threshold and expanded using an active contour to select a homogeneous area of the liver. The confidence map generated during the stiffness calculation was used to select a region of reliable wave propagation. The average stiffness within the automatically generated region of interest (ROI) was compared to measurements by two trained readers in a set of 88 clinical test cases ranging from healthy to severely fibrotic.ResultsThe stiffness measurements reported by the readers differed by −6.76% ± 22.8% (95% confidence) and had an intraclass correlation coefficient (ICC) of 0.972 (P < 0.05). The algorithm and the more experienced reader differed by 4.32% ± 14.9 with an ICC of 0.987.ConclusionThe automated algorithm performed reliably, even though MRE acquisitions often have motion artifacts present. The correlation between the automated measurements and those from the trained readers was superior to the correlation between the readers. J. Magn. Reson. Imaging 2013;38:371–379. © 2013 Wiley Periodicals, Inc.