Search

Purpose: An approach for the automated segmentation of visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) in multicenter water-fat MRI scans of the abdomen was investigated, using 2 different neural network architectures., Methods: The 2 fully convolutional network architectures U-Net and V-Net were trained, evaluated, and compared using the water-fat MRI data. Data of the study Tellus with 90 scans from a single center was used for a 10-fold cross-validation in which the most successful configuration for both networks was determined. These configurations were then tested on 20 scans of the multicenter study beta-cell function in JUvenile Diabetes and Obesity (BetaJudo), which involved a different study population and scanning device., Results: The U-Net outperformed the used implementation of the V-Net in both cross-validation and testing. In cross-validation, the U-Net reached average dice scores of 0.988 (VAT) and 0.992 (SAT). The average of the absolute quantification errors amount to 0.67% (VAT) and 0.39% (SAT). On the multicenter test data, the U-Net performs only slightly worse, with average dice scores of 0.970 (VAT) and 0.987 (SAT) and quantification errors of 2.80% (VAT) and 1.65% (SAT)., Conclusion: The segmentations generated by the U-Net allow for reliable quantification and could therefore be viable for high-quality automated measurements of VAT and SAT in large-scale studies with minimal need for human intervention. The high performance on the multicenter test data furthermore shows the robustness of this approach for data of different patient demographics and imaging centers, as long as a consistent imaging protocol is used., (© 2018 International Society for Magnetic Resonance in Medicine.)

Alzheimer's disease is characterized by hippocampal atrophy. Other factors also influence the hippocampal volume, but their interactive effect has not been investigated before in cognitively healthy individuals. The aim of this study is to evaluate the interactive effect of key demographic and clinical factors on hippocampal volume, in contrast to previous studies frequently investigating these factors in a separate manner. Also, to investigate how comparable the control groups from ADNI, AIBL, and AddNeuroMed are with five population-based cohorts. In this study, 1958 participants were included (100 AddNeuroMed, 226 ADNI, 155 AIBL, 59 BRC, 295 GENIC, 279 BioFiNDER, 398 PIVUS, and 446 SNAC-K). ANOVA and random forest were used for testing between-cohort differences in demographic-clinical variables. Multiple regression was used to study the influence of demographic-clinical variables on hippocampal volume. ANCOVA was used to analyze whether between-cohort differences in demographic-clinical variables explained between-cohort differences in hippocampal volume. Age and global brain atrophy were the most important variables in explaining variability in hippocampal volume. These variables were not only important themselves but also in interaction with gender, education, MMSE, and total intracranial volume. AddNeuroMed, ADNI, and AIBL differed from the population-based cohorts in several demographic-clinical variables that had a significant effect on hippocampal volume. Variability in hippocampal volume in individuals with normal cognition is high. Differences that previously tended to be related to disease mechanisms could also be partly explained by demographic and clinical factors independent from the disease. Furthermore, cognitively normal individuals especially from ADNI and AIBL are not representative of the general population. These findings may have important implications for future research and clinical trials, translating imaging biomarkers to the general population, and validating current diagnostic criteria for Alzheimer's disease and predementia stages., (© 2017 Wiley Periodicals, Inc.)

The two-point Dixon method is a proton chemical shift imaging technique that produces separated water-only and fat-only images from a dual-echo acquisition. It is shown how this can be achieved without the usual constraints on the echo times. A signal model considering spectral broadening of the fat peak is proposed for improved water/fat separation. Phase errors, mostly due to static field inhomogeneity, must be removed prior to least-squares estimation of water and fat. To resolve ambiguity of the phase errors, a corresponding global optimization problem is formulated and solved using a message-passing algorithm. It is shown that the noise in the water and fat estimates matches the Cramér-Rao bounds, and feasibility is demonstrated for in vivo abdominal breath-hold imaging. The water-only images were found to offer superior fat suppression compared with conventional spectrally fat suppressed images., (Copyright © 2010 Wiley-Liss, Inc.)

Accurate knowledge of the relationship between contrast agent concentration and tissue relaxation is a critical requirement for quantitative assessment of tissue perfusion using contrast-enhanced MRI. In the present study, using a pig model, the relationship between steady-state blood concentration levels of an iron oxide nanoparticle with a hydrated diameter of 12 nm (NC100150 Injection) and changes in the transverse and longitudinal relaxation rates (1/T2* and 1/T1, respectively) in blood, muscle, and renal cortex was investigated at 1.5 T. Ex vivo measurements of 1/T2* and 1/T1 were additionally performed in whole pig blood spiked with different concentrations of the iron oxide nanoparticle. In renal cortex and muscle, 1/T2* increased linearly with contrast agent concentration with slopes of 101 +/-22 s(-1)mM(-1) and 6.5 +/-0.9 s(-1)mM(-1) (mean +/- SD), respectively. In blood, 1/T2* increased as a quadratic function of contrast agent concentration, with different quadratic terms in the ex vivo vs. the in vivo experiments. In vivo, 1/T1 in blood increased linearly with contrast agent concentration, with a slope (T1-relaxivity) of 13.9 +/- 0.9 s(-1)mM(-1). The achievable increase in 1/T1 in renal cortex and muscle was limited by the rate of water exchange between the intra- and extravascular compartments and the 1/T1-curves were well described by a two-compartment water exchange limited relaxation model., (Copyright 2002 Wiley-Liss, Inc.)

Quantitative perfusion measurements require accurate knowledge of the correlation between first-pass signal changes and the corresponding tracer concentration in tissue. In the present study, a detailed analysis of first-pass renal cortical changes in T(1) and T(*)(2) following bolus injection of the iron oxide nanoparticle NC100150 Injection was investigated in a pig model using a double-echo gradient-echo sequence. The estimated change in 1/T(*)(2) during first pass calculated from single-echo sequences was compared to the true double-echo-derived 1/T(*)(2) curves. Using a single-echo (TE = 6 ms) spoiled gradient-echo sequence, the first-pass 1/T(*)(2) response following a bolus injection of 1 mg Fe/kg of NC100150 Injection was significantly underestimated due to counteracting T(1) effects. Signal response simulations showed that the relative error in the first-pass response decreased with increasing TE and contrast agent dose. However, both the maximum TE and the maximum dose are limited by excessive cortical signal loss, and the maximum TE is further limited by high temporal resolution requirements. The problem of T(1) contamination can effectively be overcome by using a double-echo gradient-echo sequence. This yields a first-pass response that truly reflects the tissue tracer concentration, which is a critical requirement for quantitative renal perfusion assessment., (Copyright 2002 Wiley-Liss, Inc.)