Automatic segmentation of abdominal organs and adipose tissue compartments in water-fat MRI: Application to weight-loss in obesity

Stand-alone MRI tool for semiautomatic volumetry of abdominal adipose compartments in patients with obesity

Abdominal adipose tissue (AT) amounts are increasingly considered as potential biomarkers for a variety of diseases and clinical questions, for instance, in diabetology, oncology or cardiovascular medicine. Despite the emergence of automated deep-learning methods for tissue quantification, interactive (supervised) segmentation tools will typically be used for model training. In comparison with CT-based approaches, MRI segmentation tools are more complex and less common. This work aims to validate a novel MRI-based tissue volumetry against a reference method in patients with (pre-) obesity. The new tool (segfatMR) was developed under a Matlab-based, open-source software framework and combines fast automatic pre-segmentation followed by manual (expert) corrections where needed. Analyses were performed retrospectively on a subset of clinical research MRI datasets (1.5 T Achieva XR, Philips Healthcare) and involved the segmentation of datasets from 20 patients (10 women/men) aged 25.1-63.1 (mean 48.5) years with BMIs between 28.3 and 58.8 (mean 36.8) kg/m2. Two independent expert readers analyzed the abdominopelvic data (30-40 slices, mean 35.8) with segfatMR and a widely used commercial tool (sliceOmatic). Coefficients of determination (R2), bias and limits of agreement (Bland Altman) were determined. Segmentation performance (R2 between methods) was excellent for both readers for SAT (> 0.99) and very high for VAT (around 0.90). The novel method was almost twice as fast as the reference standard - 25 and 19 s/slice (R1 and R2) vs. 40 and 34 s/slice. The presented semiautomatic segmentation tool enables a fast and accurate quantification of whole abdominopelvic adipose tissue volume in obesity studies. Use, adjustments and extensions of the MRI volumetry tool are facilitated by the open-source design on a standard PC.

Adiposity associates with lower plasma resolvin E1 (Rve1): a population study

BACKGROUND

Inadequate inflammation resolution may contribute to persistent low-grade inflammation that accompanies many chronic conditions. Resolution of inflammation is an active process driven by Specialized Pro-resolving Mediators (SPM) that derive from long chain n-3 and n-6 fatty acids. This study examined plasma SPM in relation to sex differences, lifestyle and a broad range cardiovascular disease (CVD) risk factors in 978, 27-year olds from the Australian Raine Study.

METHODS

Plasma SPM pathway intermediates (18-HEPE, 17-HDHA and 14-HDHA), and SPM (E- and D-series resolvins, PD1, MaR1) and LTB4 were measured by liquid chromatography-tandem mass spectrometry (LCMSMS). Pearson correlations and multiple regression analyses assessed relationships between SPM and CVD risk factors. Unpaired t-tests or ANOVA assessed the effect of sex, smoking, unhealthy alcohol consumption and obesity on SPM.

RESULTS

Women had higher 17-HDHA (p = 0.01) and lower RvE1 (p < 0.0001) and RvD1 (p = 0.05) levels compared with men. In univariate analysis, obesity associated with lower RvE1 (p = 0.002), whereas smoking (p < 0.001) and higher alcohol consumption (p < 0.001) associated with increased RvE1. In multiple regression analysis, plasma RvE1 was negatively associated with a range of measures of adiposity including BMI, waist circumference, waist-to-height ratio, abdominal subcutaneous fat volume, and skinfold thicknesses in both men and women.

CONCLUSION

This population study suggests that a deficiency in plasma RvE1 may occur in response to increasing adiposity. This observation could be relevant to ongoing inflammation that associates with CVD and other chronic diseases.

Abdominal fat quantification using convolutional networks

OBJECTIVES

To present software for automated adipose tissue quantification of abdominal magnetic resonance imaging (MRI) data using fully convolutional networks (FCN) and to evaluate its overall performance-accuracy, reliability, processing effort, and time-in comparison with an interactive reference method.

MATERIALS AND METHODS

Single-center data of patients with obesity were analyzed retrospectively with institutional review board approval. Ground truth for subcutaneous (SAT) and visceral adipose tissue (VAT) segmentation was provided by semiautomated region-of-interest (ROI) histogram thresholding of 331 full abdominal image series. Automated analyses were implemented using UNet-based FCN architectures and data augmentation techniques. Cross-validation was performed on hold-out data using standard similarity and error measures.

RESULTS

The FCN models reached Dice coefficients of up to 0.954 for SAT and 0.889 for VAT segmentation during cross-validation. Volumetric SAT (VAT) assessment resulted in a Pearson correlation coefficient of 0.999 (0.997), relative bias of 0.7% (0.8%), and standard deviation of 1.2% (3.1%). Intraclass correlation (coefficient of variation) within the same cohort was 0.999 (1.4%) for SAT and 0.996 (3.1%) for VAT.

CONCLUSION

The presented methods for automated adipose-tissue quantification showed substantial improvements over common semiautomated approaches (no reader dependence, less effort) and thus provide a promising option for adipose tissue quantification.

CLINICAL RELEVANCE STATEMENT

Deep learning techniques will likely enable image-based body composition analyses on a routine basis. The presented fully convolutional network models are well suited for full abdominopelvic adipose tissue quantification in patients with obesity.

KEY POINTS

• This work compared the performance of different deep-learning approaches for adipose tissue quantification in patients with obesity. • Supervised deep learning-based methods using fully convolutional networks  were suited best. • Measures of accuracy were equal to or better than the operator-driven approach.

Automated volume measurement of abdominal adipose tissue from entire abdominal cavity in Dixon MR images using deep learning

The purpose of this study was to realize an automated volume measurement of abdominal adipose tissue from the entire abdominal cavity in Dixon magnetic resonance (MR) images using deep learning. Our algorithm involves a combination of extraction of the abdominal cavity and body trunk regions using deep learning and extraction of a fat region based on automatic thresholding. To evaluate the proposed method, we calculated the Dice coefficient (DC) between the extracted regions using deep learning and labeled images. We also compared the visceral adipose tissue (VAT) and subcutaneous adipose tissue volumes calculated by employing the proposed method with those calculated from computed tomography (CT) images scanned on the same day using the automatic calculation method previously developed by our group. We implemented our method as a plug-in in a web-based medical image processing platform. The DCs of the abdominal cavity and body trunk regions were 0.952 ± 0.014 and 0.995 ± 0.002, respectively. The VAT volume measured from MR images using the proposed method was almost equivalent to that measured from CT images. The time required for our plug-in to process the test set was 118.9 ± 28.0 s. Using our proposed method, the VAT volume measured from MR images can be an alternative to that measured from CT images.

Measurement and Modeling of the Optical Properties of Adipose Tissue in the Terahertz Range: Aspects of Disease Diagnosis

In this paper, the measurement and modeling of optical properties in the terahertz (THz) range of adipose tissue and its components with temperature changes were performed. Spectral measurements were made in the frequency range 0.25-1 THz. The structural models of main triglycerides of fatty acids are constructed using the B3LYP/6-31G(d) method and the Gaussian03, Revision B.03 program. The optical density (OD) of adipose tissue samples decreases as temperature increases, which can be associated mostly with the dehydration of the sample. Some inclusion of THz wave scattering suppression into the OD decrease can also be expected due to refractive index matching provided by free fatty acids released from adipocytes at thermally induced cell lipolysis. It was shown that the difference between the THz absorption spectra of water and fat makes it possible to estimate the water content in adipose tissue. The proposed model was verified on the basis of molecular modeling and a comparison with experimental data for terahertz spectra of adipose tissue during its heating. Knowing the exact percentage of free and bound water in adipose tissue can help diagnose and monitor diseases, such as diabetes, obesity, and cancer.

Towards a better understanding of annotation tools for medical imaging: a survey

Medical imaging refers to several different technologies that are used to view the human body to diagnose, monitor, or treat medical conditions. It requires significant expertise to efficiently and correctly interpret the images generated by each of these technologies, which among others include radiography, ultrasound, and magnetic resonance imaging. Deep learning and machine learning techniques provide different solutions for medical image interpretation including those associated with detection and diagnosis. Despite the huge success of deep learning algorithms in image analysis, training algorithms to reach human-level performance in these tasks depends on the availability of large amounts of high-quality training data, including high-quality annotations to serve as ground-truth. Different annotation tools have been developed to assist with the annotation process. In this survey, we present the currently available annotation tools for medical imaging, including descriptions of graphical user interfaces (GUI) and supporting instruments. The main contribution of this study is to provide an intensive review of the popular annotation tools and show their successful usage in annotating medical imaging dataset to guide researchers in this area.

Postmenopausal Chinese-Singaporean Women Have a Higher Ratio of Visceral to Subcutaneous Adipose Tissue Volume than Caucasian Women of the Same Age and BMI

Central fat accumulation is a significant determinant of cardio-metabolic health risk, known to differ between ethnically distinct human populations. Despite evidence for preferential central adiposity in Asian populations, the proportional distribution between the subcutaneous and visceral compartments in Chinese postmenopausal women has not been thoroughly investigated. For this analysis, volumetrically quantified subcutaneous and visceral adipose tissue (SAT, VAT) in the pelvic and abdominal regions of postmenopausal Asian (Chinese-Singaporean) and Caucasian (German) women matched for age and Body Mass Index (BMI) was undertaken, to examine such differences between the two groups. Volumes were calculated from segmentations of magnetic resonance imaging datasets of the abdomen and pelvis. Despite SAT, VAT, and the corresponding total adipose tissue (TAT) being similar between the groups, VAT/SAT and VAT/TAT were higher in the Asian group (by 24.5% and 18.2%, respectively, each p = 0.02). Further, VAT/SAT and VAT/TAT were positively correlated with BMI in the Caucasian group only (p = 0.02 and p = 0.01, respectively). We concluded that VAT is proportionally higher in the non-obese Asian women, compared to the Caucasian women of matched age and BMI. This conclusion is in agreement with existing literature showing higher abdominal adiposity in Asian populations. Additionally, in the Asian group, BMI did not correlate with visceral adiposity on a significant level. Further analysis is required to examine the extent to which this increased VAT may impact cardio-metabolic health. There is, however, a need to emphasize healthy lifestyle behaviors in non-obese post-menopausal women of Chinese ancestry.

An automated liver segmentation in liver iron concentration map using fuzzy c-means clustering combined with anatomical landmark data

Background

To estimate median liver iron concentration (LIC) calculated from magnetic resonance imaging, excluded vessels of the liver parenchyma region were defined manually. Previous works proposed the automated method for excluding vessels from the liver region. However, only user-defined liver region remained a manual process. Therefore, this work aimed to develop an automated liver region segmentation technique to automate the whole process of median LIC calculation.

Methods

553 MR examinations from 471 thalassemia major patients were used in this study. LIC maps (in mg/g dry weight) were calculated and used as the input of segmentation procedures. Anatomical landmark data were detected and used to restrict ROI. After that, the liver region was segmented using fuzzy c-means clustering and reduced segmentation errors by morphological processes. According to the clinical application, erosion with a suitable size of the structuring element was applied to reduce the segmented liver region to avoid uncertainty around the edge of the liver. The segmentation results were evaluated by comparing with manual segmentation performed by a board-certified radiologist.

Results

The proposed method was able to produce a good grade output in approximately 81% of all data. Approximately 11% of all data required an easy modification step. The rest of the output, approximately 8%, was an unsuccessful grade and required manual intervention by a user. For the evaluation matrices, percent dice similarity coefficient (%DSC) was in the range 86–92, percent Jaccard index (%JC) was 78–86, and Hausdorff distance (H) was 14–28 mm, respectively. In this study, percent false positive (%FP) and percent false negative (%FN) were applied to evaluate under- and over-segmentation that other evaluation matrices could not handle. The average of operation times could be reduced from 10 s per case using traditional method, to 1.5 s per case using our proposed method.

Conclusion

The experimental results showed that the proposed method provided an effective automated liver segmentation technique, which can be applied clinically for automated median LIC calculation in thalassemia major patients.

Automatic segmentation of whole-body adipose tissue from magnetic resonance fat fraction images based on machine learning

OBJECTIVE

To propose a fully automated algorithm, which is implemented to segment subcutaneous adipose tissue (SAT) and internal adipose tissue (IAT) from the total adipose tissue for whole-body fat distribution analysis using proton density fat fraction (PDFF) magnetic resonance images.

MATERIALS AND METHODS

Adipose tissue segmentation was implemented using the U-Net deep neural network model. All datasets were collected using a 3.0 T magnetic resonance imaging (MRI) scanner for whole-body scan of 20 volunteers covering from neck to knee with about 160 images for each volunteer. PDFF images were reconstructed based on chemical-shift-encoded fat-water imaging. After selecting the representative PDFF images (total 906 images), the manual labeling of the SAT area was used for model training (504 images), validation (168 images), and testing (234 images).

RESULTS

The automatic segmentation model was validated through three indices using the validation and test sets. The dice similarity coefficient, precision rate, and recall rate were 0.976 ± 0.048, 0.978 ± 0.048, and 0.978 ± 0.050, respectively, in both validation and test sets.

CONCLUSION

The proposed algorithm can reliably and automatically segment SAT and IAT from whole-body MRI PDFF images. The proposed method provides a simple and automatic tool for whole-body fat distribution analysis to explore the relationship between fat deposition and metabolic-related chronic diseases.

Deep Learning-Based Automated Abdominal Organ Segmentation in the UK Biobank and German National Cohort Magnetic Resonance Imaging Studies

PURPOSE

The aims of this study were to train and evaluate deep learning models for automated segmentation of abdominal organs in whole-body magnetic resonance (MR) images from the UK Biobank (UKBB) and German National Cohort (GNC) MR imaging studies and to make these models available to the scientific community for analysis of these data sets.

METHODS

A total of 200 T1-weighted MR image data sets of healthy volunteers each from UKBB and GNC (400 data sets in total) were available in this study. Liver, spleen, left and right kidney, and pancreas were segmented manually on all 400 data sets, providing labeled ground truth data for training of a previously described U-Net-based deep learning framework for automated medical image segmentation (nnU-Net). The trained models were tested on all data sets using a 4-fold cross-validation scheme. Qualitative analysis of automated segmentation results was performed visually; performance metrics between automated and manual segmentation results were computed for quantitative analysis. In addition, interobserver segmentation variability between 2 human readers was assessed on a subset of the data.

RESULTS

Automated abdominal organ segmentation was performed with high qualitative and quantitative accuracy on UKBB and GNC data. In more than 90% of data sets, no or only minor visually detectable qualitative segmentation errors occurred. Mean Dice scores of automated segmentations compared with manual reference segmentations were well higher than 0.9 for the liver, spleen, and kidneys on UKBB and GNC data and around 0.82 and 0.89 for the pancreas on UKBB and GNC data, respectively. Mean average symmetric surface distance was between 0.3 and 1.5 mm for the liver, spleen, and kidneys and between 2 and 2.2 mm for pancreas segmentation. The quantitative accuracy of automated segmentation was comparable with the agreement between 2 human readers for all organs on UKBB and GNC data.

CONCLUSION

Automated segmentation of abdominal organs is possible with high qualitative and quantitative accuracy on whole-body MR imaging data acquired as part of UKBB and GNC. The results obtained and deep learning models trained in this study can be used as a foundation for automated analysis of thousands of MR data sets of UKBB and GNC and thus contribute to tackling topical and original scientific questions.

Artificial intelligence: a critical review of current applications in pancreatic imaging

The applications of artificial intelligence (AI), including machine learning and deep learning, in the field of pancreatic disease imaging are rapidly expanding. AI can be used for the detection of pancreatic ductal adenocarcinoma and other pancreatic tumors but also for pancreatic lesion characterization. In this review, the basic of radiomics, recent developments and current results of AI in the field of pancreatic tumors are presented. Limitations and future perspectives of AI are discussed.

Artificial intelligence for early detection of pancreatic adenocarcinoma: The future is promising

Pancreatic ductal adenocarcinoma (PDAC) is a worldwide public health concern. Despite extensive research efforts toward improving diagnosis and treatment, the 5-year survival rate at best is approximately 15%. This dismal figure can be attributed to a variety of factors including lack of adequate screening methods, late symptom onset, and treatment resistance. Pancreatic ductal adenocarcinoma remains a grim diagnosis with a high mortality rate and a significant psy-chological burden for patients and their families. In recent years artificial intelligence (AI) has permeated the medical field at an accelerated pace, bringing potential new tools that carry the promise of improving diagnosis and treatment of a variety of diseases. In this review we will summarize the landscape of AI in diagnosis and treatment of PDAC.

Noninvasive, longitudinal imaging-based analysis of body adipose tissue and water composition in a melanoma mouse model and in immune checkpoint inhibitor-treated metastatic melanoma patients

Background

As cancer cachexia (CC) is associated with cancer progression, early identification would be beneficial. The aim of this study was to establish a workflow for automated MRI-based segmentation of visceral (VAT) and subcutaneous adipose tissue (SCAT) and lean tissue water (LTW) in a B16 melanoma animal model, monitor diseases progression and transfer the protocol to human melanoma patients for therapy assessment.

Methods

For in vivo monitoring of CC B16 melanoma-bearing and healthy mice underwent longitudinal three-point DIXON MRI (days 3, 12, 17 after subcutaneous tumor inoculation). In a prospective clinical study, 18 metastatic melanoma patients underwent MRI before, 2 and 12 weeks after onset of checkpoint inhibitor therapy (CIT; n = 16). We employed an in-house MATLAB script for automated whole-body segmentation for detection of VAT, SCAT and LTW.

Results

B16 mice exhibited a CC phenotype and developed a reduced VAT volume compared to baseline (B16 − 249.8 µl, − 25%; controls + 85.3 µl, + 10%, p = 0.003) and to healthy controls. LTW was increased in controls compared to melanoma mice. Five melanoma patients responded to CIT, 7 progressed, and 6 displayed a mixed response. Responding patients exhibited a very limited variability in VAT and SCAT in contrast to others. Interestingly, the LTW was decreased in CIT responding patients (− 3.02% ± 2.67%; p = 0.0034) but increased in patients with progressive disease (+ 1.97% ± 2.19%) and mixed response (+ 4.59% ± 3.71%).

Conclusion

MRI-based segmentation of fat and water contents adds essential additional information for monitoring the development of CC in mice and metastatic melanoma patients during CIT or other treatment approaches.

Channel width optimized neural networks for liver and vessel segmentation in liver iron quantification

INTRODUCTION

MRI T2* relaxometry protocols are often used for Liver Iron Quantification in patients with hemochromatosis. Several methods exist to semi-automatically segment parenchyma and exclude vessels for this calculation.

PURPOSE

To determine if inclusion of multiple echoes inputs to Convolutional Neural Networks (CNN) improves automated liver and vessel segmentation in MRI T2* relaxometry protocols and to determine if the resultant segmentations agree with manual segmentations for liver iron quantification analysis.

METHODS

Multi echo Gradient Recalled Echo (GRE) MRI sequence for T2* relaxometry was performed for 79 exams on 31 patients with hemochromatosis for iron quantification analysis. 275 axial liver slices were manually segmented as ground truth masks. A batch normalized U-Net with variable input width to incorporate multiple echoes is used for segmentation, using DICE as the accuracy metric. ANOVA is used to evaluate significance of channel width changes in segmentation accuracy. Linear regression is used to model the relationship of channel width on segmentation accuracy. Liver segmentations are applied to relaxometry data to calculate liver T2* yielding liver iron concentration(LIC) derived from literature based calibration curves. Manual and CNN based LIC values are compared with Pearson correlation. Bland altman plots are used to visualize differences between manual and CNN based LIC values.

RESULTS

Performance metrics are tested on 55 hold out slices. Linear regression indicates that there is a monotonic increase of DICE with increasing channel depth (p = 0.001) with a slope of 3.61e-3. ANOVA indicates a significant increase segmentation accuracy over single channel starting at 3 channels. Incorporation of all channels results in an average DICE of 0.86, an average increase of 0.07 over single channel. The calculated LIC from CNN segmented livers agrees well with manual segmentation (R = 0.998, slope = 0.914, p«0.001), with an average absolute difference 0.27 ± 0.99 mg Fe/g or 1.34 ± 4.3%.

CONCLUSION

More input echoes yields higher model accuracy until the noise floor. Echos beyond the first three echo times in GRE based T2* relaxometry do not contribute significant information for segmentation of liver for LIC calculation. Deep learning models with three channel width allow for generalization of model to protocols of more than three echoes, effectively a universal requirement for relaxometry. Deep learning segmentations achieve a good accuracy compared with manual segmentations with minimal preprocessing. Liver iron values calculated from hand segmented liver and Neural network segmented liver were not statistically different from each other.

MRI estimated changes in visceral adipose tissue and liver fat fraction in patients with obesity during a very low-calorie-ketogenic diet compared to a standard low-calorie diet

AIM

To compare the changes in visceral adipose tissue (VAT), liver fat fraction, and liver stiffness using quantitative magnetic resonance imaging (MRI) during a very-low-calorie ketogenic (VLCK) diet and a standard low-calorie diet (LC).

MATERIALS AND METHODS

The study involved secondary analysis of prospective collected clinical data. Patients undergoing weight loss interventions were randomised to either a LC or a VLCK diet. VAT, liver fat fraction, and stiffness were measured at baseline and after 2 months.

RESULTS

Forty-six patients were included; 39 patients were evaluated at baseline and at 2 months follow-up. Mean weight loss was -9.7±3.8 kg (interquartile range [IQR]: -12.3; -7 kg) in the VLCK group and -1.67±2.2 kg (IQR: -3.3, -0.1 kg) in the LC group (p<0.0001). Mean VAT reductions were -39.3±40 cm² (IQR: -52, -10 cm²) and -12.5±38.3 cm² (IQR: -29, 5 cm²; p=0.0398), and mean liver proton density fat fraction (PDFF) reductions were -4.77±4.2% (IQR: -7.3, -1.7%) and -0.79±1.7%, (IQR: -1.8, -0.4%; p<0.005) in the VLCK group and in the LC group, respectively. No significant changes in liver stiffness occurred from baseline to follow-up.

CONCLUSION

A VLCK diet resulted in greater weight loss than a standard low-calorie diet and in significantly greater reduction in liver PDFF. As anthropometric measurements may not correlate with liver fat changes, it may be advantageous to include quantitative MRI to the monitoring strategies of patients undergoing weight-loss programmes.

Magnetic Resonance Imaging and Bioelectrical Impedance Analysis to Assess Visceral and Abdominal Adipose Tissue

OBJECTIVE

This study aimed to compare a state-of-the-art bioelectrical impedance analysis (BIA) device with two-point Dixon magnetic resonance imaging (MRI) for the quantification of visceral adipose tissue (VAT) as a health-related risk factor.

METHODS

A total of 63 male participants were measured using a 3-T MRI scanner and a segmental, multifrequency BIA device. MRI generated fat fraction (FF) maps, in which VAT volume, total abdominal adipose tissue volume, and FF of visceral and total abdominal compartments were quantified. BIA estimated body fat mass and VAT area.

RESULTS

Coefficients of determination between abdominal (r²  = 0.75) and visceral compartments (r²  = 0.78) were similar for both groups, but slopes differed by a factor of two. The ratio of visceral to total abdominal FF was increased in older men compared with younger men. This difference was not detected with BIA. MRI and BIA measurements of the total abdominal volume correlated moderately (r²  = 0.31-0.56), and visceral measurements correlated poorly (r²  = 0.13-0.44).

CONCLUSIONS

Visceral BIA measurements agreed better with MRI measurements of the total abdomen than of the visceral compartment, indicating that BIA visceral fat area assessment cannot differentiate adipose tissue between visceral and abdominal compartments in young and older participants.

Efficacy of a 2-Month Very Low-Calorie Ketogenic Diet (VLCKD) Compared to a Standard Low-Calorie Diet in Reducing Visceral and Liver Fat Accumulation in Patients With Obesity

Background: Currently the treatment of non-alcoholic fatty liver disease (NAFLD) is based on weight loss through lifestyle changes, such as exercise combined with calorie-restricted dieting. Objectives: To assess the effects of a commercially available weight loss program based on a very low-calorie ketogenic diet (VLCKD) on visceral adipose tissue (VAT) and liver fat content compared to a standard low-calorie (LC) diet. As a secondary aim, we evaluated the effect on liver stiffness measurements. Methods: Open, randomized controlled, prospective pilot study. Patients were randomized and treated either with an LC or a VLCKD and received orientation and encouragement to physical activity equally for both groups. VAT, liver fat fraction, and liver stiffness were measured at baseline and after 2 months of treatment using magnetic resonance imaging. Paired t-tests were used for comparison of continuous variables between visits and unpaired test between groups. Categorical variables were compared using the χ²-test. Pearson correlation was used to assess the association between VAT, anthropometric measures, and hepatic fat fraction. A significance level of the results was established at p < 0.05. Results: Thirty-nine patients (20 with VLCKD and 19 with LC) were evaluated at baseline and 2 months of intervention. Relative weight loss at 2 months was -9.59 ± 2.87% in the VLCKD group and -1.87 ± 2.4% in the LC group (p < 0.001). Mean reductions in VAT were -32.0 cm² for VLCKD group and -12.58 cm² for LC group (p < 0.05). Reductions in liver fat fraction were significantly more pronounced in the VLCKD group than in the LC group (4.77 vs. 0.79%; p < 0.005). Conclusion: Patients undergoing a VLCKD achieved superior weight loss, with significant VAT and liver fat fraction reductions when compared to the standard LC diet. The weight loss and rapid mobilization of liver fat demonstrated with VLCKD could serve as an effective alternative for the treatment of NAFLD. Clinical Trial Registration: ClinicalTrials.gov, identifier: NCT04322110.

Automated pancreas segmentation from computed tomography and magnetic resonance images: A systematic review

The pancreas is a highly variable organ, the size, shape, and position of which are affected by age, sex, adiposity, the presence of diseases affecting the pancreas (e.g., diabetes, pancreatic cancer, pancreatitis) and other factors. Accurate automated segmentation of the pancreas has the potential to facilitate timely diagnosing and managing of diseases of the endocrine and exocrine pancreas. The aim was to systematically review studies reporting on automated pancreas segmentation algorithms derived from computed tomography (CT) or magnetic resonance (MR) images. The MEDLINE database and three patent databases were searched. Data on the performance of algorithms were meta-analysed, when possible. The algorithms were classified into one of four groups: multiorgan atlas-based, landmark-based, shape model-based, and neural network-based. A total of 13 cohorts suitable for meta-analysis were pooled to determine the performance of pancreas segmentation algorithms altogether using the Dice coefficient. These cohorts, comprising 1110 individuals, yielded a weighted mean Dice coefficient of 74.4%. Eight cohorts suitable for meta-analysis were pooled to determine the performance of pancreas segmentation algorithms altogether using the Jaccard index. These cohorts, comprising 636 individuals, yielded a weighted mean Jaccard index of 63.7%. Multiorgan atlas-based algorithms had a weighted mean Dice coefficient of 70.1% and a weighted mean Jaccard index of 59.8%. Neural network-based algorithms had a weighted mean Dice coefficient of 82.3% and a weighted mean Jaccard index of 70.1%. Studies using the other two types of algorithms were not meta-analysable. The above findings indicate that the automation of pancreas segmentation represents a considerable challenge as the performance of current automated pancreas segmentation algorithms is suboptimal. Adopting standardised reporting on performance of pancreas segmentation algorithms and encouraging the use of benchmark pancreas segmentation datasets will allow future algorithms to be tested and compared more easily and fairly.

Computer-aided pancreas segmentation based on 3D GRE Dixon MRI: a feasibility study

Background

Pancreas segmentation is of great significance for pancreatic cancer radiotherapy positioning, pancreatic structure, and function evaluation.

Purpose

To investigate the feasibility of computer-aided pancreas segmentation based on optimized three-dimensional (3D) Dixon magnetic resonance imaging (MRI).

Material and Methods

Seventeen healthy volunteers (13 men, 4 women; mean age = 53.4 ± 13.2 years; age range = 28–76 years) underwent routine and optimized 3D gradient echo (GRE) Dixon MRI at 3.0 T. The computer-aided segmentation of the pancreas was executed by the Medical Imaging Interaction ToolKit (MITK) with the traditional segmentation algorithm pipeline (a threshold method and a morphological method) on the opposed-phase and water images of Dixon. The performances of our proposed computer segmentation method were evaluated by Dice coefficients and two-dimensional (2D)/3D visualization figures, which were compared for the opposed-phase and water images of routine and optimized Dixon sequences.

Results

The dice coefficients of the computer-aided pancreas segmentation were 0.633 ± 0.080 and 0.716 ± 0.033 for opposed-phase and water images of routine Dixon MRI, respectively, while they were 0.415 ± 0.143 and 0.779 ± 0.048 for the optimized Dixon MRI, respectively. The Dice index was significantly higher based on the water images of optimized Dixon than those in the other three groups (all P values < 0.001), including water images of routine Dixon MRI and both of the opposed-phase images of routine and optimized Dixon sequences.

Conclusion

Computer-aided pancreas segmentation based on Dixon MRI is feasible. The water images of optimized Dixon obtained the best similarity with a good stability.

Fully convolutional networks for automated segmentation of abdominal adipose tissue depots in multicenter water-fat MRI

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.

MRI adipose tissue and muscle composition analysis-a review of automation techniques

MRI is becoming more frequently used in studies involving measurements of adipose tissue and volume and composition of skeletal muscles. The large amount of data generated by MRI calls for automated analysis methods. This review article presents a summary of automated and semi-automated techniques published between 2013 and 2017. Technical aspects and clinical applications for MRI-based adipose tissue and muscle composition analysis are discussed based on recently published studies. The conclusion is that very few clinical studies have used highly automated analysis methods, despite the rapidly increasing use of MRI for body composition analysis. Possible reasons for this are that the availability of highly automated methods has been limited for non-imaging experts, and also that there is a limited number of studies investigating the reproducibility of automated methods for MRI-based body composition analysis.

Rodent Abdominal Adipose Tissue Imaging by MR

Rodents including rats and mice are important models to study obesity, diabetes, and metabolic syndrome in a preclinical setting. Translational and longitudinal imaging of these rodents permit investigation of metabolic diseases and identification of imaging biomarkers suitable for clinical translation. Here we describe the imaging protocols for achieving quantitative abdominal imaging in small animals followed by segmentation and quantification of fat volumes.

Association of proton density fat fraction in adipose tissue with imaging-based and anthropometric obesity markers in adults

Background/Objectives:

The purpose of this study was to examine the relationship of the proton density fat fraction (PDFF), measured by magnetic resonance imaging (MRI), of supraclavicular and gluteal adipose tissue with subcutaneous and visceral adipose tissue (SAT and VAT) volumes, liver fat fraction and anthropometric obesity markers. The supraclavicular fossa was selected as a typical location where brown adipocytes may be present in humans and the gluteal region was selected as a typical location enclosing primarily white adipocytes.

Subjects/Methods:

In this cross-sectional study, 61 adults (44 women, median age 29.3 years, range 21–68 years) underwent an MRI examination of the neck and the abdomen/pelvis (3T, Ingenia, Philips Healthcare). PDFF maps of the supraclavicular and gluteal adipose tissue and the liver were generated. Volumes of SAT and VAT were calculated and supraclavicular and subcutaneous fat were segmented using custom-built post-processing algorithms. Body mass index (BMI), waist circumference and waist-to-height ratio were recorded. Statistical analysis was conducted using the Student's t-test and Pearson correlation analysis.

Results:

Mean supraclavicular PDFF was 75.3±4.7% (range 65.4–83.8%) and mean gluteal PDFF was 89.7±2.9% (range 82.2-94%), resulting in a significant difference (P<0.0001). Supraclavicular PDFF was positively correlated with VAT (r=0.76, P<0.0001), SAT (r=0.73, P<0.0001), liver PDFF (r=0.42, P=0.0008) and all measured anthropometric obesity markers. Gluteal subcutaneous PDFF also correlated with VAT (r=0.59, P<0.0001), SAT (r=0.63, P<0.0001), liver PDFF (r=0.3, P=0.02) and anthropometric obesity markers.

Conclusions:

The positive correlations between adipose tissue PDFF and imaging, as well as anthropometric obesity markers suggest that adipose tissue PDFF may be useful as a biomarker for improving the characterization of the obese phenotype, for risk stratification and for selection of appropriate treatment strategies.

Shortcuts to a functional adipose tissue: The role of small non-coding RNAs

Metabolic diseases such as type 2 diabetes are a major public health issue worldwide. These diseases are often linked to a dysfunctional adipose tissue. Fat is a large, heterogenic, pleiotropic and rather complex tissue. It is found in virtually all cavities of the human body, shows unique plasticity among tissues, and harbors many cell types in addition to its main functional unit - the adipocyte. Adipose tissue function varies depending on the localization of the fat depot, the cell composition of the tissue and the energy status of the organism. While the white adipose tissue (WAT) serves as the main site for triglyceride storage and acts as an important endocrine organ, the brown adipose tissue (BAT) is responsible for thermogenesis. Beige adipocytes can also appear in WAT depots to sustain heat production upon certain conditions, and it is becoming clear that adipose tissue depots can switch phenotypes depending on cell autonomous and non-autonomous stimuli. To maintain such degree of plasticity and respond adequately to changes in the energy balance, three basic processes need to be properly functioning in the adipose tissue: i) adipogenesis and adipocyte turnover, ii) metabolism, and iii) signaling. Here we review the fundamental role of small non-coding RNAs (sncRNAs) in these processes, with focus on microRNAs, and demonstrate their importance in adipose tissue function and whole body metabolic control in mammals.