Intraperitoneal, but not retroperitoneal, visceral adipose tissue is associated with diabetes mellitus: a cross-sectional, retrospective pilot analysis

A fully convolutional neural network for comprehensive compartmentalization of abdominal adipose tissue compartments in MRI

BACKGROUND

Existing literature has highlighted structural, physiological, and pathological disparities among abdominal adipose tissue (AAT) sub-depots. Accurate separation and quantification of these sub-depots are crucial for advancing our understanding of obesity and its comorbidities. However, the absence of clear boundaries between the sub-depots in medical imaging data has challenged their separation, particularly for internal adipose tissue (IAT) sub-depots. To date, the quantification of AAT sub-depots remains challenging, marked by a time-consuming, costly, and complex process.

PURPOSE

To implement and evaluate a convolutional neural network to enable granular assessment of AAT by compartmentalization of subcutaneous adipose tissue (SAT) into superficial subcutaneous (SSAT) and deep subcutaneous (DSAT) adipose tissue, and IAT into intraperitoneal (IPAT), retroperitoneal (RPAT), and paraspinal (PSAT) adipose tissue.

MATERIAL AND METHODS

MRI datasets were retrospectively collected from Singapore Preconception Study for Long-Term Maternal and Child Outcomes (S-PRESTO: 389 women aged 31.4 ± 3.9 years) and Singapore Adult Metabolism Study (SAMS: 50 men aged 28.7 ± 5.7 years). For all datasets, ground truth segmentation masks were created through manual segmentation. A Res-Net based 3D-UNet was trained and evaluated via 5-fold cross-validation on S-PRESTO data (N = 300). The model's final performance was assessed on a hold-out (N = 89) and an external test set (N = 50, SAMS).

RESULTS

The proposed method enabled reliable segmentation of individual AAT sub-depots in 3D MRI volumes with high mean Dice similarity scores of 98.3%, 97.2%, 96.5%, 96.3%, and 95.9% for SSAT, DSAT, IPAT, RPAT, and PSAT respectively.

CONCLUSION

Convolutional neural networks can accurately sub-divide abdominal SAT into SSAT and DSAT, and abdominal IAT into IPAT, RPAT, and PSAT with high accuracy. The presented method has the potential to significantly contribute to advancements in the field of obesity imaging and precision medicine.

Automatic segmentation of large-scale CT image datasets for detailed body composition analysis

BACKGROUND

Body composition (BC) is an important factor in determining the risk of type 2-diabetes and cardiovascular disease. Computed tomography (CT) is a useful imaging technique for studying BC, however manual segmentation of CT images is time-consuming and subjective. The purpose of this study is to develop and evaluate fully automated segmentation techniques applicable to a 3-slice CT imaging protocol, consisting of single slices at the level of the liver, abdomen, and thigh, allowing detailed analysis of numerous tissues and organs.

METHODS

The study used more than 4000 CT subjects acquired from the large-scale SCAPIS and IGT cohort to train and evaluate four convolutional neural network based architectures: ResUNET, UNET++, Ghost-UNET, and the proposed Ghost-UNET++. The segmentation techniques were developed and evaluated for automated segmentation of the liver, spleen, skeletal muscle, bone marrow, cortical bone, and various adipose tissue depots, including visceral (VAT), intraperitoneal (IPAT), retroperitoneal (RPAT), subcutaneous (SAT), deep (DSAT), and superficial SAT (SSAT), as well as intermuscular adipose tissue (IMAT). The models were trained and validated for each target using tenfold cross-validation and test sets.

RESULTS

The Dice scores on cross validation in SCAPIS were: ResUNET 0.964 (0.909-0.996), UNET++ 0.981 (0.927-0.996), Ghost-UNET 0.961 (0.904-0.991), and Ghost-UNET++ 0.968 (0.910-0.994). All four models showed relatively strong results, however UNET++ had the best performance overall. Ghost-UNET++ performed competitively compared to UNET++ and showed a more computationally efficient approach.

CONCLUSION

Fully automated segmentation techniques can be successfully applied to a 3-slice CT imaging protocol to analyze multiple tissues and organs related to BC. The overall best performance was achieved by UNET++, against which Ghost-UNET++ showed competitive results based on a more computationally efficient approach. The use of fully automated segmentation methods can reduce analysis time and provide objective results in large-scale studies of BC.

Using Free-Breathing MRI to Quantify Pancreatic Fat and Investigate Spatial Heterogeneity in Children

BACKGROUND

MRI acquisition for pediatric pancreatic fat quantification is limited by breath-holds (BH). Full segmentation (FS) or small region of interest (ROI) analysis methods may not account for pancreatic fat spatial heterogeneity, which may limit accuracy.

PURPOSE

To improve MRI acquisition and analysis for quantifying pancreatic proton-density fat fraction (pPDFF) in children by investigating free-breathing (FB)-MRI, characterizing pPDFF spatial heterogeneity, and relating pPDFF to clinical markers.

STUDY TYPE

Prospective.

POPULATION

A total of 34 children, including healthy (N = 16, 8 female) and overweight (N = 18, 5 female) subjects.

FIELD STRENGTH AND SEQUENCES

3 T; multiecho gradient-echo three-dimensional (3D) stack-of-stars FB-MRI, multiecho gradient-echo 3D Cartesian BH-MRI.

ASSESSMENT

A radiologist measured FS- and ROI-based pPDFF on FB-MRI and BH-MRI PDFF maps, with anatomical images as references. Regional pPDFF in the pancreatic head, body, and tail were measured on FB-MRI. FS-pPDFF, ROI-pPDFF, and regional pPDFF were compared, and related to clinical markers, including hemoglobin A1c.

STATISTICAL TESTS

T-test, Bland-Altman analysis, Lin's concordance correlation coefficient (CCC), one-way analysis of variance, and Spearman's rank correlation coefficient were used. P < 0.05 was considered significant.

RESULTS

FS-pPDFF and ROI-pPDFF from FB-MRI and BH-MRI had mean difference = 0.4%; CCC was 0.95 for FS-pPDFF and 0.62 for ROI-pPDFF. FS-pPDFF was higher than ROI-pPDFF (10.4% ± 6.4% vs. 4.2% ± 2.8%). Tail-pPDFF (11.6% ± 8.1%) was higher than body-pPDFF (8.9% ± 6.3%) and head-pPDFF (8.7% ± 5.2%). Head-pPDFF and body-pPDFF positively correlated with hemoglobin A1c.

DATA CONCLUSION

FB-MRI pPDFF is comparable to BH-MRI. Spatial heterogeneity affects pPDFF quantification. Regional measurements of pPDFF in the head and body were correlated with hemoglobin A1c, a marker of insulin sensitivity.

EVIDENCE LEVEL

2 TECHNICAL EFFICACY: Stage 2.

Changes in abdominal fat depots after bariatric surgery are associated with improved metabolic profile

BACKGROUND AND AIMS

Obesity associated with a change in the quantity and quality of fat depots. Using computed tomography (CT), we analyzed abdominal fat depots in patients with obesity after bariatric surgery according to their metabolic health status.

METHODS AND RESULTS

We recruited 79 individuals with metabolically unhealthy obesity before bariatric surgery and compared them with age-sex matched healthy controls. The volume and fat attenuation index (FAI) of fat depots were measured using CT scans that were conducted prior to and a year after bariatric surgery. 'Metabolically healthy' was defined as having no hypertension, normal fasting glucose and a waist-to-hip ratio of <1.05 for men and <0.95 for women. Individuals who achieved a metabolic health status conversion (MHC) (n = 29, 37%)-from unhealthy to healthy-were younger (p < 0.001) as compared to individuals without MHC. Pre-surgery BMI and reduction of BMI did not differ between the two groups (p = 0.099, p = 0.5730). Bariatric surgery reduced the volume and increased the FAI of fat depots. Baseline lower abdominal periaortic adipose tissue (AT) volume (p = 0.014) and great percent reduction in renal sinus AT volume after surgery (p = 0.019) were associated with MHC after surgery. Increased intraperitoneal AT FAI (p = 0.031) was also associated with MHC.

CONCLUSION

MHC was not associated with improvement in general obesity, based on indicators such as reduction of BMI after surgery. Weight reduction induced specific abdominal fat depot changes measured by CT are positively associated with MHC.

Keeping It Local in Metabolic Disease: Adipose Tissue Paracrine Signaling and Insulin Resistance

Alterations in adipose tissue composition and function are associated with obesity and contribute to the development of type 2 diabetes. While the significance of this relationship has been cemented, our understanding of the multifaceted role of adipose tissue in metabolic heath and disease continues to evolve and expand. Heterogenous populations of cells that make up adipose tissue throughout the body generate diverse secretomes containing a mosaic of bioactive compounds with vast structural and signaling capabilities. While there are many reports highlighting the important role of adipose tissue endocrine signaling in insulin resistance and type 2 diabetes, the direct, local, paracrine effect of adipose tissue has received less attention. Recent studies have begun to underscore the importance of considering anatomically discrete adipose depots for their specific impact on local microenvironments and metabolic function in neighboring tissues as well as regulation of whole-body physiology. This article highlights the important role of adipose tissue paracrine signaling on metabolic function and insulin sensitivity in nearby tissues and organs, specifically focusing on visceral, pancreatic, subcutaneous, intermuscular, and perivascular adipose tissue depots.

Mouse model of the adipose organ: the heterogeneous anatomical characteristics

Adipose tissue plays a pivotal role in energy storage, hormone secretion, and temperature control. Mammalian adipose tissue is largely divided into white adipose tissue and brown adipose tissue, although recent studies have discovered the existence of beige adipocytes. Adipose tissues are widespread over the whole body and each location shows distinctive metabolic features. Mice are used as a representative experimental model system in metabolic studies due to their numerous advantages. Importantly, the adipose tissues of experimental animals and humans are not perfectly matched, and each adipose tissue exhibits both similar and specific characteristics. Nevertheless, the diversity and characteristics of mouse adipose tissue have not yet been comprehensively summarized. This review summarizes diverse information about the different types of adipose tissue being studied in mouse models. The types and characteristics of adipocytes were described, and each adipose tissue was classified by type, and features such as its distribution, origin, differences from humans, and metabolic characteristics were described. In particular, the distribution of widely studied adipose tissues was illustrated so that researchers can comprehensively grasp its location. Also, the adipose tissues misused or confusingly used among researchers were described. This review will provide researchers with comprehensive information and cautions needed to study adipose tissues in mouse models.