UK Biobank (UKB) conducts large-scale examinations of more than half a million volunteers, collecting health-related information on genetics, lifestyle, blood biochemistry, and more. Medical imaging of 100,000 subjects, with 70,000 follow-up sessions, enables measurements of organs, muscle, and body composition. With up to 170,000 mounting MR images, various methodologies are accordingly engaged in large-scale image analysis. This work presents an experimental inference engine that can automatically predict a comprehensive profile of subject metadata from UKB neck-to-knee body MRI. It was evaluated in cross-validation for baseline characteristics such as age, height, weight, and sex, but also measurements of body composition, organ volumes, and abstract properties like grip strength, pulse rate, and type 2 diabetic status. It predicted subsequently released test data covering twelve body composition metrics with a 3% median error. The proposed system can automatically analyze one thousand subjects within ten minutes, providing individual confidence intervals. The underlying methodology utilizes convolutional neural networks for image-based mean-variance regression on two-dimensional representations of the MRI data. This work aims to make the proposed system available for free to researchers, who can use it to obtain fast and fully-automated estimates of 72 different measurements immediately upon release of new UKB image data.
Large-scale medical studies such as the UK Biobank examine thousands of volunteer participants with medical imaging techniques. Combined with the vast amount of collected metadata, anatomical information from these images has the potential for medical analyses at unprecedented scale. However, their evaluation often requires manual input and long processing times, limiting the amount of reference values for biomarkers and other measurements available for research. Recent approaches with convolutional neural networks for regression can perform these evaluations automatically. On magnetic resonance imaging (MRI) data of more than 40,000 UK Biobank subjects, these systems can estimate human age, body composition and more. This style of analysis is almost entirely data-driven and no manual intervention or guidance with manually segmented ground truth images is required. The networks often closely emulate the reference method that provided their training data and can reach levels of agreement comparable to the expected variability between established medical gold standard techniques. The risk of silent failure can be individually quantified by predictive uncertainty obtained from a mean-variance criterion and ensembling. Saliency analysis furthermore enables an interpretation of the underlying relevant image features and showed that the networks learned to correctly target specific organs, limbs, and regions of interest.
Abstract Background and Aims Kidney parenchymal volume (KPV) presents a natural variation with respect to sex, age, and body size, and is also affected by diseases such as diabetes. The UK Biobank (UKBB) is a large-scale study including clinical and MRI data. The current project investigated the association between KPV and age in UKBB participants without diabetes and with diabetes type 1 (T1D), and type 2 (T2D). In addition, the effect of different treatments for T2D on KPV was investigated. Method KPV was estimated in 35,703 UKBB participants (52% women, age = 45-82 years) with a deep-learning-based segmentation of both kidneys (Dice = 0.956, error < 4%). The cohort was classified into Control, T1D and T2D subjects using an algorithm developed on UKBB clinical data (Eastwood et al 2016). Individuals with T2D were further divided into groups related to treatment: Lifestyle (no pharmaceuticals, i.e. light treatment, mean disease duration of 6.2 years), Metformin (metformin as the only pharmaceutical, i.e. intermediate treatment, mean disease duration of 8.3 years), and Other (more potent treatment, combination of pharmaceuticals, mean disease duration of 14.1 years). KPV was studied as a function of age in the different groups, divided according to sex. The statistical difference in mean KPV between groups was tested. For each group, the association between KPV and age was assessed by linear regression. Comparison of line slopes was conducted to investigate whether age-related patterns in KPV differed statistically between groups. Results The moving average curve of KPV vs age, in controls and subjects with T1D and T2D, is shown in Fig 1A (with a 15-year sliding window). The corresponding curve for different T2D treatment groups is depicted in Fig 1B. Fig 2A-D presents results for the comparison of KPV and regression line slope between the subject groups. According to Fig 1A and 2A, KPV is usually higher in subjects with T1D and T2D than in controls. As shown in Fig 1B and 2B, T2D subjects with longer disease duration and on pharmaceutical treatment (Metformin or Other) are generally more prone to large KPV than subjects with adapted lifestyle as treatment. The decreasing KPV pattern with age is faster in T2D subjects than controls but not significantly different between T1D subjects and controls (Fig 1A and 2C). Women in group Other also show a pattern of steeper age-related decline in KPV compared to remaining women with T2D treatment (Fig 1B and 2D). Conclusion Compared to controls, T1D subjects show enlarged KPV similar to that of T2D subjects, which is in line with previous literature. Subjects with T2D show a pattern of steeper age-related decline in KPV compared to controls. Female T2D subjects with longer disease duration, usually on a more potent treatment (beyond adaption of lifestyle and metformin as the only pharmaceutical), are more prone to enlarged KPV and exhibit a pattern of steeper age-related decline in KPV. This may be due to hyperfiltration caused by diabetes, resulting in increased kidney size. The normal loss of glomeruli with age could be accelerated by diabetes, leading to a greater loss in KPV per year in diabetics. Steeper KPV decline patterns in disease could also be caused by selection bias where old subjects with large KPV and related complications are less likely to participate in the study.
Along with rich health-related metadata, an ongoing imaging study has acquired MRI of over 40,000 male and female UK Biobank participants aged 44-82 since 2014. Phenotypes derived from these images, such as measurements of body composition, can reveal new links between genetics, cardiovascular disease, and metabolic conditions. In this retrospective study, six measurements of body composition were automatically estimated by ResNet50 neural networks for image-based regression from neck-to-knee body MRI. Despite the potential for high speed and accuracy, these networks produce no output segmentations that could indicate the reliability of individual measurements. The presented experiments therefore examine mean-variance regression and ensembling for predictive uncertainty estimation, which can quantify individual measurement errors and thereby help to identify potential outliers, anomalies, and other failure cases automatically. In 10-fold cross-validation on data of about 8,500 subjects, mean-variance regression and ensembling showed complementary benefits, reducing the mean absolute error across all predictions by 12%. Both improved the calibration of uncertainties and their ability to identify high prediction errors. With intra-class correlation coefficients (ICC) above 0.97, all targets except the liver fat content yielded relative measurement errors below 5%. Testing on another 1,000 subjects showed consistent performance, and the method was finally deployed for inference to 30,000 subjects with missing reference values. The results indicate that deep regression ensembles could ultimately provide automated, uncertainty-aware measurements of body composition for more than 120,000 UK Biobank neck-to-knee body MRI that are to be acquired within the coming years.
UK Biobank (UKB) is conducting a large-scale study of more than half a million volunteers, collecting health-related information on genetics, lifestyle, blood biochemistry, and more. Medical imaging furthermore targets 100,000 subjects, with 70,000 follow-up sessions, enabling measurements of organs, muscle, and body composition. With up to 170,000 mounting MR images, various methodologies are accordingly engaged in large-scale image analysis. This work presents an experimental inference engine that can automatically predict a comprehensive profile of subject metadata from UKB neck-to-knee body MRI. In cross-validation, it accurately inferred baseline characteristics such as age, height, weight, and sex, but also emulated measurements of body composition by DXA, organ volumes, and abstract properties like grip strength, pulse rate, and type 2 diabetic status (AUC: 0.866). The proposed system can automatically analyze thousands of subjects within hours and provide individual confidence intervals. The underlying methodology is based on convolutional neural networks for image-based mean-variance regression on two-dimensional representations of the MRI data. This work aims to make the proposed system available for free to researchers, who can use it to obtain fast and fully-automated estimates of 72 different measurements immediately upon release of new UK Biobank image data.
Non-alcoholic fatty liver disease (NAFLD) can lead to irreversible liver damage manifesting in systemic effects (e.g., elevated portal vein pressure and splenomegaly) with increased risk of deadly outcomes. However, the association of spleen volume with NAFLD and related type 2-diabetes (T2D) is not fully understood. The UK Biobank contains comprehensive health-data of 500,000 participants, including clinical data and MR-images of >40,000 individuals. The present study estimated the spleen volume of 37,066 participants through automated deep learning-based image segmentation of neck-to-knee MR-images. The aim was to investigate the associations of spleen volume with NAFLD, T2D and liver fibrosis, while adjusting for natural confounders. The recent redefinition and new nomenclature of NAFLD to metabolic dysfunction-associated steatotic liver disease (MASLD), promoted by major organisations of studies on liver disease, was not employed as introduced after the conduct or this study. The results showed that spleen volume decreased with age, correlated positively with body size and was smaller in females compared to males. Larger spleens were observed in subjects with NAFLD and T2D compared to controls. Spleen volume was also positively and independently associated with liver fat fraction, liver volume and the fibrosis-4 score, with notable volumetric increases already at low liver fat fractions and volumes, but not independently associated with T2D. These results suggest a link between spleen volume and NAFLD already at an early stage of the disease, potentially due to initial rise in portal vein pressure.
