Uncertainty-aware high-fidelity anatomical mri synthesis using deep convolutional network with monte carlo dropout

HIGHLIGHTS

• What: The authors propose a convolution neural network to obtain highresolution T1-weighted MRI from lower-resolution T2-weighted input that can be acquired within a shorter scan time. Specifically, the authors propose an uncertainty-aware convolutional neural_network (CNN) which synthesises high-quality anatomical MRI with uncertainty estimation. The authors propose a deep convolutional neural_network architecture for a multi-contrast MRI model. The authors design a Monte Carlo dropout-based approach to enable the proposed super-resolution (SR) model to be uncertainty-aware.
• Who: Synthesis and collaborators from the Harrow International School Shenzhen Qianhai, Shenzhen . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.