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.

A wealth of information is contained in images obtained by whole-body magnetic resonance imaging (MRI).Studying the link between the imaged anatomy and properties known from outside sources has the potential to give new insights into the underlying factors that manifest themselves in individual human morphology. In this work we investigate the expression of age-related changes in the whole-body image. A large dataset of about 32,000 subjects scanned from neck to knee and aged 44-82 years from the UK Biobank study was used for a machine-based analysis. We trained a convolutional neural network based on the VGG16 architecture to predict the age of a given subject based on image data from these scans. In 10-fold crossvalidation on 23,000 of these images the network reached a mean absolute error (MAE) of 2.49 years (R 2 = 0.83) and showed consistent performance on a separate test set of another 8,000 images. On a second test set of 100 images the network outperformed the averaged estimates given by three experienced radiologists, which reached an MAE of 5.58 years (R 2 = 0.08), by more than three years on average. In an attempt to explain these findings, we employ saliency analysis that opens up the image-based criteria used by the automated method to human interpretation. We aggregate the saliency into a single anatomical visualization which clearly highlights structures in the aortic arch and knee as primary indicators of age.

The UK Biobank is collecting extensive data on health-related characteristics of over half a million volunteers. The biological samples of blood and urine can provide valuable insight on kidney function, with important links to cardiovascular and metabolic health. Further information on kidney anatomy could be obtained by medical imaging. In contrast to the brain, heart, liver, and pancreas, no dedicated Magnetic Resonance Imaging (MRI) is planned for the kidneys. An image-based assessment is nonetheless feasible in the neck-to-knee body MRI intended for abdominal body composition analysis, which also covers the kidneys. In this work, a pipeline for automated segmentation of parenchymal kidney volume in UK Biobank neck-to-knee body MRI is proposed. The underlying neural network reaches a relative error of 3.8%, with Dice score 0.956 in validation on 64 subjects, close to the 2.6% and Dice score 0.962 for repeated segmentation by one human operator. The released MRI of about 40,000 subjects can be processed within one day, yielding volume measurements of left and right kidney. Algorithmic quality ratings enabled the exclusion of outliers and potential failure cases. The resulting measurements can be studied and shared for large-scale investigation of associations and longitudinal changes in parenchymal kidney volume.