Using UK Biobank data to establish population-specific atlases from whole body MRI

Reliable reference data in medical imaging is largely unavailable. Developing tools that allow for the comparison of individual patient data to reference data has a high potential to improve diagnostic imaging. Population atlases are a commonly used tool in medical imaging to facilitate this. Constructing such atlases becomes particularly challenging when working with highly heterogeneous datasets, such as whole-body images, which contain significant anatomical variations. In this work, we propose a pipeline for generating a standardised whole-body atlas for a highly heterogeneous population by partitioning the population into anatomically meaningful subgroups. Using magnetic resonance images from the UK Biobank dataset, we create six whole-body atlases representing a healthy population average. We furthermore unbias them, and this way obtain a realistic representation of the population. In addition to the anatomical atlases, we generate probabilistic atlases that capture the distributions of abdominal fat (visceral and subcutaneous) and five abdominal organs across the population (liver, spleen, pancreas, left and right kidneys). Our pipeline effectively generates high-quality, realistic whole-body atlases with clinical applicability. The probabilistic atlases show differences in fat distribution between subjects with medical conditions such as diabetes and cardiovascular diseases and healthy subjects in the atlas space. With this work, we make the constructed anatomical and label atlases publically available, with the expectation that they will support medical research involving whole-body MR images. Medical imaging requires examples of healthy images to be available for comparison with individual patient data. This comparison is important to detect changes that are indicative of disease and enable diagnosis. Population atlases consist of healthy images that can be used for comparisons. The images should match population characteristics as much as possible. However, building these atlases can be difficult, especially if the images used to compile the atlas show differences. In this study, we provide a method to create a standardised whole-body atlas using whole-body (neck to knee) magnetic resonance images. We produce a set of atlases that represent a healthy population. These atlases have been made publicly available and should assist medical researchers and improve healthcare outcomes for patients. Starck and Sideri-Lampretsa et al. propose a pipeline for generating whole-body atlases from a heterogeneous population by dividing it into anatomically meaningful subgroups. They demonstrate the use of these atlases for studying differences between healthy individuals and those with conditions such as diabetes or cardiovascular disease.