Automatic Segmentation of Abdominal Adipose Tissue in MRI. In: Heyden A, Kahl F, editors. Image Analysis. Berlin: Springer Berlin Heidelberg; 2011. pp. 501-11. [DOI: 10.1007/978-3-642-21227-7_47]

Automated Deep Learning-Based Segmentation of Abdominal Adipose Tissue on Dixon MRI in Adolescents: A Prospective Population-Based Study

BACKGROUND. The prevalence of childhood obesity has increased significantly worldwide, highlighting a need for accurate noninvasive quantification of body fat distribution in children. OBJECTIVE. The purpose of this study was to develop and test an automated deep learning method for subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) segmentation using Dixon MRI acquisitions in adolescents. METHODS. This study was embedded within the Generation R Study, a prospective population-based cohort study in Rotterdam, The Netherlands. The current study included 2989 children (1432 boys, 1557 girls; mean age, 13.5 years) who underwent investigational whole-body Dixon MRI after reaching the age of 13 years during the follow-up phase of the Generation R Study. A 2D competitive dense fully convolutional neural network model (2D-CDFNet) was trained from scratch to segment abdominal SAT and VAT using Dixon MRI-based images. The model underwent training, validation, and testing in 62, eight, and 15 children, respectively, who were selected by stratified random sampling, with manual segmentations used as reference. Segmentation performance was assessed using the Dice similarity coefficient and volumetric similarity. Two observers independently performed subjective visual assessments of automated segmentations in 504 children, selected by stratified random sampling, with undersegmentation and oversegmentation scored on a scale of 0-3 (with a score of 3 denoting nearly perfect segmentation). For 2820 children for whom complete data were available, Spearman correlation coefficients were computed among MRI measurements and BMI and dual-energy x-ray absorptiometry (DEXA)-based measurements. The model used (gitlab.com/radiology/msk/genr/abdomen/cdfnet) is publicly available. RESULTS. In the test dataset, the mean Dice similarity coefficient and mean volu-metric similarity, respectively, were 0.94 ± 0.03 [SD] and 0.98 ± 0.01 [SD] for SAT and 0.85 ± 0.05 and 0.92 ± 0.04 for VAT. The two observers assigned a score of 3 for SAT in 94% and 93% for the undersegmentation proportion and in 99% and 99% for the oversegmentation proportion, and they assigned a score of 3 for VAT in 99% and 99% for the undersegmentation proportion and in 95% and 97% for the oversegmentation proportion. Correlations with SAT and VAT were 0.808 and 0.698 for BMI and 0.941 and 0.801 for DEXA-derived fat mass. CONCLUSION. We trained and evaluated the 2D-CDFNet model on Dixon MRI in adolescents. Quantitative and qualitative measures of automated SAT and VAT segmentations indicated strong model performance. CLINICAL IMPACT. The automated model may facilitate large-scale studies investigating abdominal fat distribution on MRI among adolescents as well as associations of fat distribution with clinical outcomes.

Whole-body oxidative stress reduction during testosterone therapy in aging men: A randomized placebo-controlled trial

BACKGROUND

Testosterone replacement therapy in aging men increases lean body mass and decreases whole-body fat. The safety of testosterone replacement therapy concerning cardiovascular disease is unresolved and assessment of whole-body oxidative stress may contribute to future decision making.

OBJECTIVES

To determine whole-body oxidative stress during testosterone replacement therapy and placebo in aging men and evaluate if a change in oxidative stress was mediated by changed body composition.

MATERIALS AND METHODS

This was a double-blinded, randomized, placebo-controlled study for 24 weeks in 38 men aged 60-78 years with bioavailable testosterone <7.3 nmol/L and waist circumference ≥94 cm who were randomized to testosterone replacement therapy (testosterone gel) (N = 20) or placebo (N = 18). At baseline and after 24 weeks, whole-body oxidative stress was assessed by oxidized derivatives of nucleic acids, 8-oxoguanosine and 8-oxo-2'-deoxyguanosine in 24-h urine samples by ultra-performance liquid chromatography tandem mass spectrometry. Lean body mass and whole-body fat were measured by dual X-ray absorptiometry. Subcutaneous and visceral adipose tissue were estimated by magnetic resonance imaging. Testosterone replacement therapy versus placebo was compared by Mann-Whitney tests on ∆-values (24-0 weeks).

RESULTS

Baseline age was 67 (64-72) years (median [interquartile range]), body mass index 29.8 (26.6-33.3) kg/m2 , waist 107 (99-117) cm, and bioavailable testosterone 4.7 (3.7-5.9) nmol/L. During testosterone replacement therapy, 8-oxoguanosine in 24-h urine samples decreased from 21.6 (19.8; 27.7) nm to 15.0 (12.2; 18.8) nm (p = 0.038 vs. placebo), lean body mass increased (p < 0.01) and whole-body fat (p = 0.02) and subcutaneous adipose tissue (p < 0.01) decreased. 8-Oxoguanosine in 24-h urine samples was inversely associated with Δ-lean body mass (ρ = -0.38, p = 0.03), which remained significant after adjusting for Δ-total testosterone. 8-Oxo-2'-deoxyguanosine in 24-h urine samples was unchanged (p = 0.06) during testosterone replacement therapy and Δ-8-oxo-2'-deoxyguanosine in 24-h urine samples was associated with Δ-whole-body fat (kg) (ρ = 0.47, p < 0.01). Δ-Values of oxidative stress biomarkers were not associated with Δ-fasting insulin or Δ-homeostatic model assessment of insulin resistance.

DISCUSSION

Oxidative stress decreased during testosterone replacement therapy compared to placebo, which could be mediated by changed body composition.

CONCLUSION

Whole-body oxidative stress decreased during 24 weeks of testosterone replacement therapy in aging men.

LVNet: Lightweight Model for Left Ventricle Segmentation for Short Axis Views in Echocardiographic Imaging

Lightweight segmentation models are becoming more popular for fast diagnosis on small and low cost medical imaging devices. This study focuses on the segmentation of the left ventricle (LV) in cardiac ultrasound (US) images. A new lightweight model [LV network (LVNet)] is proposed for segmentation, which gives the benefits of requiring fewer parameters but with improved segmentation performance in terms of Dice score (DS). The proposed model is compared with state-of-the-art methods, such as UNet, MiniNetV2, and fully convolutional dense dilated network (FCdDN). The model proposed comes with a post-processing pipeline that further enhances the segmentation results. In general, the training is done directly using the segmentation mask as the output and the US image as the input of the model. A new strategy for segmentation is also introduced in addition to the direct training method used. Compared with the UNet model, an improvement in DS performance as high as 5% for segmentation with papillary (WP) muscles was found, while showcasing an improvement of 18.5% when the papillary muscles are excluded. The model proposed requires only 5% of the memory required by a UNet model. LVNet achieves a better trade-off between the number of parameters and its segmentation performance as compared with other conventional models. The developed codes are available at https://github.com/navchetanawasthi/Left_Ventricle_Segmentation.

Automated Segmentation of Visceral, Deep Subcutaneous, and Superficial Subcutaneous Adipose Tissue Volumes in MRI of Neonates and Young Children

Purpose

To develop and evaluate an automated segmentation method for accurate quantification of abdominal adipose tissue (AAT) depots (superficial subcutaneous adipose tissue [SSAT], deep subcutaneous adipose tissue [DSAT], and visceral adipose tissue [VAT]) in neonates and young children.

Materials and Methods

This was a secondary analysis of prospectively collected data, which used abdominal MRI data from Growing Up in Singapore Towards healthy Outcomes, or GUSTO, a longitudinal mother-offspring cohort, to train and evaluate a convolutional neural network for volumetric AAT segmentation. The data comprised imaging volumes of 333 neonates obtained at early infancy (age ≤2 weeks, 180 male neonates) and 755 children aged either 4.5 years (n = 316, 150 male children) or 6 years (n = 439, 219 male children). The network was trained on images of 761 randomly selected volumes (neonates and children combined) and evaluated on 100 neonatal volumes and 227 child volumes by using 10-fold validation. Automated segmentations were compared with expert-generated manual segmentation. Segmentation performance was assessed using Dice scores.

Results

When the model was tested on the test datasets across the 10 folds, the model had strong agreement with the ground truth for all testing sets, with mean Dice similarity scores for SSAT, DSAT, and VAT, respectively, of 0.960, 0.909, and 0.872 in neonates and 0.944, 0.851, and 0.960 in children. The model generalized well to different body sizes and ages and to all abdominal levels.

Conclusion

The proposed segmentation approach provided accurate automated volumetric assessment of AAT compartments on MR images of neonates and children.Keywords Pediatrics, Deep Learning, Convolutional Neural Networks, Water-Fat MRI, Image Segmentation, Deep and Superficial Subcutaneous Adipose Tissue, Visceral Adipose TissueClinical trial registration no. NCT01174875 Supplemental material is available for this article. © RSNA, 2021.

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.

FatSegNet: A fully automated deep learning pipeline for adipose tissue segmentation on abdominal dixon MRI

PURPOSE

Introduce and validate a novel, fast, and fully automated deep learning pipeline (FatSegNet) to accurately identify, segment, and quantify visceral and subcutaneous adipose tissue (VAT and SAT) within a consistent, anatomically defined abdominal region on Dixon MRI scans.

METHODS

FatSegNet is composed of three stages: (a) Consistent localization of the abdominal region using two 2D-Competitive Dense Fully Convolutional Networks (CDFNet), (b) Segmentation of adipose tissue on three views by independent CDFNets, and (c) View aggregation. FatSegNet is validated by: (1) comparison of segmentation accuracy (sixfold cross-validation), (2) test-retest reliability, (3) generalizability to randomly selected manually re-edited cases, and (4) replication of age and sex effects in the Rhineland Study-a large prospective population cohort.

RESULTS

The CDFNet demonstrates increased accuracy and robustness compared to traditional deep learning networks. FatSegNet Dice score outperforms manual raters on VAT (0.850 vs. 0.788) and produces comparable results on SAT (0.975 vs. 0.982). The pipeline has excellent agreement for both test-retest (ICC VAT 0.998 and SAT 0.996) and manual re-editing (ICC VAT 0.999 and SAT 0.999).

CONCLUSIONS

FatSegNet generalizes well to different body shapes, sensitively replicates known VAT and SAT volume effects in a large cohort study and permits localized analysis of fat compartments. Furthermore, it can reliably analyze a 3D Dixon MRI in ∼1 minute, providing an efficient and validated pipeline for abdominal adipose tissue analysis in the Rhineland Study.

Anthropometry, DXA, and leptin reflect subcutaneous but not visceral abdominal adipose tissue on MRI in 197 healthy adolescents

BackgroundAbdominal fat distribution is associated with the development of cardio-metabolic disease independently of body mass index (BMI). We assessed anthropometry, serum adipokines, and DXA as markers of abdominal subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) using magnetic resonance imaging (MRI).MethodsWe performed a cross-sectional study that included 197 healthy adolescents (114 boys) aged 10-15 years nested within a longitudinal population-based cohort. Clinical examination, blood sampling, DXA, and abdominal MRI were performed. SAT% and VAT% were adjusted to total abdominal volume.ResultsGirls had a higher SAT% than did boys in early and late puberty (16 vs. 13%, P<0.01 and 20 vs. 15%, P=0.001, respectively), whereas VAT% was comparable (7% in both genders, independently of puberty). DXA android fat% (standard deviation score (SDS)), suprailiac skinfold thickness (SDS), leptin, BMI (SDS), waist-to-height ratio (WHtR), and waist circumference (SDS) correlated strongly with SAT% (descending order: r=0.90-0.55, all P<0.001) but weakly with VAT% (r=0.49-0.06). Suprailiac skinfold was the best anthropometric marker of SAT% (girls: R²=48.6%, boys: R²=65%, P<0.001) and VAT% in boys (R²=16.4%, P<0.001). WHtR was the best marker of VAT% in girls (R²=7.6%, P=0.007).ConclusionsHealthy girls have a higher SAT% than do boys, whereas VAT% is comparable, independently of puberty. Anthropometry and circulating leptin are valid markers of SAT%, but not of VAT%.

Segmentation and quantification of adipose tissue by magnetic resonance imaging

In this brief review, introductory concepts in animal and human adipose tissue segmentation using proton magnetic resonance imaging (MRI) and computed tomography are summarized in the context of obesity research. Adipose tissue segmentation and quantification using spin relaxation-based (e.g., T1-weighted, T2-weighted), relaxometry-based (e.g., T1-, T2-, T2*-mapping), chemical-shift selective, and chemical-shift encoded water-fat MRI pulse sequences are briefly discussed. The continuing interest to classify subcutaneous and visceral adipose tissue depots into smaller sub-depot compartments is mentioned. The use of a single slice, a stack of slices across a limited anatomical region, or a whole body protocol is considered. Common image post-processing steps and emerging atlas-based automated segmentation techniques are noted. Finally, the article identifies some directions of future research, including a discussion on the growing topic of brown adipose tissue and related segmentation considerations.

Strength training and testosterone treatment have opposing effects on migration inhibitor factor levels in ageing men

Background. The beneficial effects of testosterone treatment (TT) are debated. Methods. Double-blinded, placebo-controlled study of six months TT (gel) in 54 men aged 60–78 with bioavailable testosterone (BioT) <7.3 nmol/L and waist >94 cm randomized to TT (50–100 mg/day, n = 20), placebo (n = 18), or strength training (ST) (n = 16) for 24 weeks. Moreover, the ST group was randomized to TT (n = 7) or placebo (n = 9) after 12 weeks. Outcomes. Chemokines (MIF, MCP-1, and MIP-1α) and lean body mass (LBM), total, central, extremity, visceral, and subcutaneous (SAT) fat mass established by DXA and MRI. Results. From 0 to 24 weeks, MIF and SAT decreased during ST + placebo versus placebo, whereas BioT and LBM were unchanged. TT decreased fat mass (total, central, extremity, and SAT) and increased BioT and LBM versus placebo. MIF levels increased during TT versus ST + placebo. ST + TT decreased fat mass (total, central, and extremity) and increased BioT and LBM versus placebo. From 12 to 24 weeks, MCP-1 levels increased during TT versus placebo and MCP-1 levels decreased during ST + placebo versus placebo. Conclusion. ST + placebo was associated with decreased MIF levels suggesting decreased inflammatory activity. TT may be associated with increased inflammatory activity. This trial is registered with ClinicalTrials.gov NCT00700024.

Exercise training favors increased insulin-stimulated glucose uptake in skeletal muscle in contrast to adipose tissue: a randomized study using FDG PET imaging

Physical exercise increases peripheral insulin sensitivity, but regional differences are poorly elucidated in humans. We investigated the effect of aerobic exercise training on insulin-stimulated glucose uptake in five individual femoral muscle groups and four different adipose tissue regions, using dynamic (femoral region) and static (abdominal region) 2-deoxy-2-[¹⁸F]fluoro-d-glucose (FDG) PET/CT methodology during steady-state insulin infusion (40 mU·m⁻²·min⁻¹). Body composition was measured by dual X-ray absorptiometry and MRI. Sixty-one healthy, sedentary [V(O2max) 36(5) ml·kg⁻¹·min⁻¹; mean(SD)], moderately overweight [BMI 28.1(1.8) kg/m²], young [age: 30(6) yr] men were randomized to sedentary living (CON; n = 17 completers) or moderate (MOD; 300 kcal/day, n = 18) or high (HIGH; 600 kcal/day, n = 18) dose physical exercise for 11 wk. At baseline, insulin-stimulated glucose uptake was highest in femoral skeletal muscle followed by intraperitoneal visceral adipose tissue (VAT), retroperitoneal VAT, abdominal (anterior + posterior) subcutaneous adipose tissue (SAT), and femoral SAT (P < 0.0001 between tissues). Metabolic rate of glucose increased similarly (~30%) in the two exercise groups in femoral skeletal muscle (MOD 24[9, 39] μmol·kg⁻¹·min⁻¹, P = 0.004; HIGH 22[9, 35] μmol·kg⁻¹·min⁻¹, P = 0.003) (mean[95% CI]) and in five individual femoral muscle groups but not in femoral SAT. Standardized uptake value of FDG decreased ~24% in anterior abdominal SAT and ~20% in posterior abdominal SAT compared with CON but not in either intra- or retroperitoneal VAT. Total adipose tissue mass decreased in both exercise groups, and the decrease was distributed equally among subcutaneous and intra-abdominal depots. In conclusion, aerobic exercise training increases insulin-stimulated glucose uptake in skeletal muscle but not in adipose tissue, which demonstrates some interregional differences.