We develop pipelines for reconstructing susceptibility source separation (χ-separation) maps, which requires a T2 map, from UK Biobank protocol and routine clinical protocol data that have no T2 map but have various T2-weighted contrasts (e.g., FLAIR and T2-weighted images). Using these and additional contrast-weighted images, we propose a deep neural network framework that generates an R2 (=1/T2) map, with which χ-separation is conducted. The proposed pipelines successfully generated positive and negative susceptibility maps that are highly similar to gold standard results. The results suggest that χ-separation is applicable to various clinical routine protocols and open-source data.
This abstract and the presentation materials are available to members only; a login is required.