Background Overweight and obesity are major worldwide health concerns characterized by an abnormal accumulation of fat in adipose tissue (AT) and liver. Purpose To evaluate the volume and the fatty acid (FA) composition of the subcutaneous adipose tissue (SAT) and the visceral adipose tissue (VAT) and the fat content in the liver from 3D chemical‐shift‐encoded (CSE)‐MRI acquisition, before and after a 31‐day overfeeding protocol. Study Type Prospective and longitudinal study. Subjects Twenty‐one nonobese healthy male volunteers. Field Strength/Sequence A 3D spoiled‐gradient multiple echo sequence and STEAM sequence were performed at 3T. Assessment AT volume was automatically segmented on CSE‐MRI between L2 to L4 lumbar vertebrae and compared to the dual‐energy X‐ray absorptiometry (DEXA) measurement. CSE‐MRI and MR spectroscopy (MRS) data were analyzed to assess the proton density fat fraction (PDFF) in the liver and the FA composition in SAT and VAT. Gas chromatography‐mass spectrometry (GC‐MS) analyses were performed on 13 SAT samples as a FA composition countermeasure. Statistical Tests Paired t‐test, Pearson's correlation coefficient, and Bland–Altman plots were used to compare measurements. Results SAT and VAT volumes significantly increased (P < 0.001). CSE‐MRI and DEXA measurements were strongly correlated (r = 0.98, P < 0.001). PDFF significantly increased in the liver (+1.35, P = 0.002 for CSE‐MRI, + 1.74, P = 0.002 for MRS). FA composition of SAT and VAT appeared to be consistent between localized‐MRS and CSE‐MRI (on whole segmented volume) measurements. A significant difference between SAT and VAT FA composition was found (P < 0.001 for CSE‐MRI, P = 0.001 for MRS). MRS and CSE‐MRI measurements of the FA composition were correlated with the GC‐MS results (for ndb: rMRS/GC‐MS = 0.83 P < 0.001, rCSE‐MRI/GC‐MS = 0.84, P = 0.001; for nmidb: rMRS/GC‐MS = 0.74, P = 0.006, rCSE‐MRI/GC‐MS = 0.66, P = 0.020) Data Conclusion The follow‐up of liver PDFF, volume, and FA composition of AT during an overfeeding diet was demonstrated through different methods. The CSE‐MRI sequence associated with a dedicated postprocessing was found reliable for such quantification. Level of Evidence: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:1587–1599.
In recent years, deep learning has been successfully applied to the analysis and processing of ultrasound images. To date, most of this research has focused on segmentation and view recognition. This paper benchmarks different convolutional neural network algorithms for motion estimation in ultrasound imaging. We evaluated and compared several networks derived from FlowNet2, one of the most efficient architectures in computer vision. The networks were tested with and without transfer learning and the best configuration was compared against the particle-imaging-velocimetry method, a popular state-of-the-art block-matching algorithm. Rotations are known to be difficult to track from ultrasound images due to a significant speckle decorrelation. We thus focused on images of rotating disks, that could be tracked through speckle features only. Our database consisted of synthetic and in-vitro B-mode images after log-compression, and covered a large range of rotational speeds. One of the FlowNet2 sub-networks, FlowNet2SD, produced competitive results with a motion field error smaller than 1 pixel on real data after transfer learning based on simulated data. These errors remains small for a large velocity range without the need for hyper-parameter tuning, which indicates the high potential and adaptability of deep learning solutions to motion estimation in ultrasound imaging.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.