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.

Anthropometric estimators of abdominal fat volume in adults with overweight and obesity.

BACKGROUND/OBJECTIVES: To evaluate anthropometric measures for the prediction of whole-abdominal adipose tissue volumes VXAT (subcutaneous VSAT, visceral VVAT and total VTAT) in patients with obesity. SUBJECTS/METHODS: A total of 181 patients (108 women) with overweight or obesity were analyzed retrospectively. MRI data (1.5 T) were available from independent clinical trials at a single institution (Integrated Research and Treatment Center of Obesity, University of Leipzig). A custom-made software was used for automated tissue segmentation. Anthropometric parameters (AP) were circumferences of the waist (WC) and hip (HC), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR) and the (hypothetical) hip-to-height ratio (HHtR). Agreement was evaluated by standard deviations sd% of percent differences between estimated volumes (using results of linear AP-VXAT regression) and measured ones as well as Pearson's correlation coefficient r. RESULTS: For SAT volume estimation, the smallest sd% for all patients was seen for HC (25.1%) closely followed by HHtR (25.2%). Sex-specific results for females (17.5% for BMI and 17.2% for HC) and males (20.7% for WC) agreed better. VAT volumes could not be estimated reliably by any of the anthropometric measures considered here. TAT volumes in a mixed population could be best estimated by BMI closely followed by WC (roughly 17.5%). A sex-specific consideration reduced the deviations to around 16% for females (BMI and WC) and below 14% for males (WC). CONCLUSIONS: We suggest the use of sex-specific parameters-BMI or HC for females and WC for males-for the estimation of abdominal SAT and TAT volumes in patients with overweight or obesity.

Abdominal subcutaneous fat quantification in obese patients from limited field-of-view MRI data.

Different types of adipose tissue can be accurately localized and quantified by tomographic imaging techniques (MRI or CT). One common shortcoming for the abdominal subcutaneous adipose tissue (ASAT) of obese subjects is the technically restricted imaging field of view (FOV). This work derives equations for the conversion between six surrogate measures and fully segmented ASAT volume and discusses the predictive power of these image-based quantities. Clinical (gender, age, anthropometry) and MRI data (1.5 T, two-point Dixon sequence) of 193 overweight and obese patients (116 female, 77 male) from a single research center for obesity were analyzed retrospectively. Six surrogate measures of fully segmented ASAT volume (VASAT) were considered: two simple ASAT lengths, two partial areas (Ap-FH, Ap-ASIS) and two partial volumes (Vp-FH, Vp-ASIS) limited by either the femoral heads (FH) or the anterior superior iliac spine (ASIS). Least-squares regression between each measure and VASAT provided slope and intercept for the computation of estimated ASAT volumes (V~ASAT). Goodness of fit was evaluated by coefficient of determination (R2) and standard deviation of percent differences (sd%) between V~ASAT and VASAT. Best agreement was observed for partial volume Vp-FH (sd% = 14.4% and R2 = 0.78), followed by Vp-ASIS (sd% = 18.1% and R2 = 0.69) and AWFASIS (sd% = 23.9% and R2 = 0.54), with minor gender differences only. Other estimates from simple lengths and partial areas were moderate only (sd% > 23.0% and R2 < 0.50). Gender differences in R2 generally ranged between 0.02 (dven) and 0.29 (Ap-FH). The common FOV restriction for MRI volumetry of ASAT in obese subjects can best be overcome by estimating VASAT from Vp-FH using the equation derived here. The very simple AWFASIS can be used with reservation.

Estimation of abdominal subcutaneous fat volume of obese adults from single-slice MRI data - Regression coefficients and agreement.

PURPOSE: Cross-sectional imaging is increasingly used to quantify adipose tissue compartments in subjects with overweight or obesity. The lack of ionizing radiation makes magnetic resonance imaging (MRI) highly preferable to computed tomography (CT) although it is generally less standardized and time-consuming. Fat areas of single or stacks of neighboring slices have previously been considered as surrogates to avoid laborious processing of whole abdominal data-but studies are inconsistent in design and results. The present work therefore analyzed a relatively large number of overweight or obese adults and involved a total of eight landmarks and two surrogates (slice and stack). The goals were to identify the most reliable estimators of abdominal subcutaneous adipose tissue (ASAT) volume for both genders and to relate the findings to the pertinent literature. MATERIAL AND METHODS: Anthropometric and fat-sensitive 1.5 T MRI data of 193 patients (116 female, 77 male) from different IRB-approved studies at a single clinical research institution (IFB Adiposity Diseases, University Medicine Leipzig, Germany) were analyzed retrospectively. Mean (± SD) age and BMI were 51.5 (± 12.4) years and 33.7 (± 3.9) kg/m2 for females versus 57.6 (± 12.4) years and 32.1 (± 3.7) kg/m2 for males. Areas of selected axial slices (10 mm thick, 0.5 mm gap) and of stacks of five slices at common landmarks - intervertebral disc spaces L1/L2 to L5/S1, anterior superior iliac spine (ASIS), femoral head (FH) and umbilicus (UM) - were considered as estimators for ASAT volume (reference). Agreement between simple areas and reference volumes was asssessed by linear regression (coefficient of determination R2) as well as standard deviations of percent differences sd% between estimated and measured volumes. RESULTS: ASAT volumes ranged from 6.61 to 21.94 L for females (mean: 13.37 L) and from 5.42 to 17.90 L (mean: 9.89 L) for males. The smallest sd% (8.4 %-10.1 %) and largest R2 values (0.86-0.92) for single slices were observed for three candidate slice positions that were also associated with the highest ASAT volume fraction: L4/L5, L5/S1 and UM. The stack estimates for these landmarks were overall somewhat better (7.3 %-9.7 %, 0.88-0.94, respectively). The differences in sd% between genders ranged between 0.2 % and 1.1 %. CONCLUSION: ASAT volume in overweight or obese patients can be readily estimated with good accuracy from a single MRI slice centered at intervertebral disc space L5/S1 for both genders. Disc space L4/L5 or the umbilicus are nearly equivalent landmarks, in particular for male subjects. The extension to stack measures may yield too little improvement to justify the extra effort. Landmarks like ASIS, FH or the remaining lumbar disc spaces are considered as unreliable.

Half-body MRI volumetry of abdominal adipose tissue in patients with obesity.

BACKGROUND: The purpose of this study was to determine to what extent the whole volumes of abdominal subcutaneous (ASAT) and visceral adipose tissue (VAT) of patients with obesity can be predicted by using data of one body half only. Such a workaround has already been reported for dual-energy x-ray absorption (DEXA) scans and becomes feasible whenever the field of view of an imaging technique is not large enough. METHODS: Full-body abdominal MRI data of 26 patients from an obesity treatment center (13 females and 13 males, BMI range 30.8-41.2 kg/m2, 32.6-61.5 years old) were used as reference (REF). MRI was performed with IRB approval on a clinical 1.5 T MRI (Achieva dStream, Philips Healthcare, Best, Netherlands). Segmentation of adipose tissue was performed with a custom-made Matlab software tool. Statistical measures of agreement were the coefficient of determination R2 of a linear fit. RESULTS: Mean ASATREF was 12,976 (7812-24,161) cm3 and mean VATREF was 4068 (1137-7518) cm3. Mean half-body volumes relative to the whole-body values were 50.8% (48.2-53.7%) for ASATL and 49.2% (46.3-51.8%) for ASATR. Corresponding volume fractions were 56.4% (51.4-65.9%) for VATL and 43.6% (34.1-48.6%) for VATR. Correlations of ASATREF with ASATL as well as with ASATR were both excellent (R2 > 0.99, p < 0.01). Corresponding correlations of VATREF were marginally lower (R2 = 0.98 for VATL, p < 0.01, and R2 = 0.97 for VATR, p < 0.01). CONCLUSIONS: In conclusion, abdominal fat volumes can be reliably assessed by half-body MRI data, in particular the subcutaneous fat compartment.