2024/10/15 by Suma Anand, Kaiwen Xu, Anand, Suma +5 · 1 voice
Computer Science · Medicine · #Advanced MRI Techniques and Applications #Cardiovascular Disease and Adiposity #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Radiomics and Machine Learning in Medical Imaging #cs.CV #cs.LG
paper · pdf · doi:10.48550/arxiv.2410.11186
openalex publication_date 2024/10/15 · arxiv published 2024/10/15 · arxiv updated 2024/10/15 · openalex created_date 2024/10/20 · openalex updated_date 2026/07/28
Magnetic Resonance Imaging (MRI) is the gold standard for measuring fat and iron content non-invasively in the body via measures known as Proton Density Fat Fraction (PDFF) and R2^*, respectively. However, conventional PDFF and R2^* quantification methods operate on MR images voxel-wise and require at least three measurements to estimate three quantities: water, fat, and R2^*. Alternatively, the two-point Dixon MRI protocol is widely used and fast because it acquires only two measurements; however, these cannot be used to estimate three quantities voxel-wise. Leveraging the fact that neighboring voxels have similar values, we propose using a generative machine learning approach to learn PDFF and R2^* from Dixon MRI. We use paired Dixon-IDEAL data from UK Biobank in the liver and a Pix2Pix conditional GAN to demonstrate the first large-scale R2^* imputation from two-point Dixon MRIs. Using our proposed approach, we synthesize PDFF and R2^* maps that show significantly greater correlation with ground-truth than conventional voxel-wise baselines.