Estimation of Attenuation Coefficients for Simultaneous PET/MRI Using Both MRI and PET Data Combining Bayesian Deep Learning pseudo-CT and Maximum Likelihood Estimation of Activity and Attenuation

10 Aug 2020 Leynes Andrew P Ahn Sangtae P. Wangerin Kristen A. Kaushik Sandeep S. Wiesinger Florian Hope Thomas A. Larson Peder E. Z.

A major remaining challenge for magnetic resonance-based attenuation correction methods (MRAC) is their susceptibility to sources of MRI artifacts (e.g. implants, motion) as well as uncertainties due to the limitations of MRI contrast (e.g. accurate bone delineation and density, and separation of air/bone). We propose using a Bayesian deep convolutional neural network that, in addition to generating an initial pseudo-CT from MR data, also produces uncertainty estimates of the pseudo-CT to quantify the limitations of the MR data... (read more)

PDF Abstract
No code implementations yet. Submit your code now