OBJECTIVES: Bone marrow adipose tissue (BMAT) represents >10% of total fat mass in healthy humans and further increases in diverse clinical conditions, but the impact of BMAT on human health and disease remains poorly understood. Magnetic resonance imaging (MRI) allows non-invasive measurement of the bone marrow fat fraction (BMFF), and human MRI studies have begun identifying associations between BMFF and skeletal or metabolic diseases. However, such studies have so far been limited to smaller cohorts: analysis of BMFF on a larger, population scale therefore has huge potential to reveal fundamental new knowledge of BMAT's formation and pathophysiological functions. The UK Biobank (UKBB) is undertaking whole-body MRI of 100,000 participants, providing the ideal opportunity for such advances. MATERIALS AND METHODS: Herein, we developed a deep learning pipeline for high-throughput BMFF analysis of these UKBB MRI data. Automatic bone marrow segmentation was achieved by designing new lightweight attention-based 3D U-Net convolutional neural networks that allowed more-accurate segmentation of small structures from large volumetric data. Using manual segmentations from 61-64 subjects, the models were trained against four bone marrow regions of interest: the spine, femoral head, total hip and femoral diaphysis. Models were validated using a further 10-12 datasets for each region and then used to segment datasets from a further 729 UKBB participants. BMFF was then determined and assessed for expected and new pathophysiological characteristics. RESULTS: Dice scores confirmed the accuracy of the models, which matched or exceeded that for conventional U-Net models. The BMFF measurements from the 729-subject cohort confirmed previously reported relationships between BMFF and age, sex and bone mineral density, while also identifying new site- and sex-specific BMFF characteristics. CONCLUSIONS: We have established a new deep learning method for accurate segmentation of small structures from large volumetric data. This method works well for accurate, large-scale BMFF analysis from UKBB MRI data and has the potential to reveal novel clinical insights. The application of our method across the full UKBB imaging cohort will therefore allow identification of the genetic and pathophysiological factors associated with altered BMAT. Together, our findings establish the utility of deep learning for population-level BMFF analysis and promise to help elucidate the full impact of BMAT on human health and disease.

G.P. is currently an employee of Pfizer. All other authors declare no competing interests.

This work was funded by a grant from the Medical Research Council (MR/S010505/1 to W.P.C.). W.P.C. was further supported by a Chancellor's Fellowship from the University of Edinburgh. C.W. was further supported by the British Heart Foundation (RG/16/10/32375). C.G. and T.M. were supported by the Edinburgh Clinical Research Facility and NHS Lothian R&D.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

The ethics committee of the University of Edinburgh waived ethical approval for this work. This is because the phenotypic and imaging data used in this study were obtained from UK Biobank and analysed under an approved project application (ID 48697); the UK Biobank has it's own ethical framework and does not require further approval from institutional ethic committees. All work reported herein was done in accordance with UK Biobank ethical requirements.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.

Yes

All data for fat fraction (FF) and segmentation volumes will be uploaded to the UK Biobank. Code for the deep learning models will be made available via GitHub. Code for regression analyses will be made available via DataShare (https://datashare.ed.ac.uk). Until these data are publicly available, the authors will agree to all reasonable requests for code and data sharing, in accordance with UK Biobank guidelines.