Large-scale, deep learning image analysis to generate insights for cardiorenal and metabolic disease

Cardiorenal and metabolic diseases are the most common cause of death worldwide. There is a huge need to understand more in this area to be able to guide both clinical management and clinical research. Researchers from Uppsala University and Antaros Medical have developed deep-learning image analysis approaches to enable investigation of the wealth of data available in the UK Biobank, potentially including 100,000 participants performing Magnetic Resonance Imaging (MRI).

The different deep-learning image analysis approaches have been applied on neck-to-knee body MRI generating data for more than 40,000 currently available UK Biobank participants for biometrics, liver fat and kidney parenchymal volume (see our news from ASN 2020 and AASLD 2020). Linking this data with other measurements in the UK Biobank has the potential to generate insights on large-scale associations and longitudinal changes bringing value to the understanding of disease.

Find out more in the recent publications:

Kidney segmentation in neck-to-knee body MRI of 40,000 UK Biobank participants
Authors: Taro Langner, Andreas Östling, Lukas Maldonis, Albin Karlsson, Daniel Olmo, Dag Lindgren, Andreas Wallin, Lowe Lundin, Robin Strand, Håkan Ahlström & Joel Kullberg

Large-Scale Inference of Liver Fat with Neural Networks on UK Biobank Body MRI

Authors: Taro Langner, Robin Strand, Håkan Ahlström & Joel Kullberg

Large-scale biometry with interpretable neural network regression on UK Biobank body MRI
Authors: Taro Langner, Robin Strand, Håkan Ahlström & Joel Kullberg

Identifying Morphological Indicators of Aging With Neural Networks on Large-Scale Whole-Body MRI

Authors: Taro Langner, Johan Wikström, Tomas Bjerner, Håkan Ahlström, and Joel Kullberg

Pressify Share on Linkedin Share on Twitter