Convolutional neural networks based robust Denoising of Low-Dose Computed Tomography Perfusion maps

Matlab Codes* : (requires MatConvNet )

Data: Training data is generated using the tool digital-brain-perfusion-phantom

#Matlab code for generating the network architecture.

#Matlab code for Testing the trained network for Phantom data.

#Matlab codes for Online SPD (Comparision codes and preprocessing codes)

#Matlab codes for BM3D

Models for both temporal and map denoising can be accessed here: Link

Venkata S. Kadimesetty, Sreedevi Gutta, Sriram Ganapathy, and Phaneendra K. Yalavarthy Convolutional Neural Network based RobustDenoising of Low-Dose Computed Tomography Perfusion Maps,Accepted in IEEE TRPMS PDF

This Matlab code is used as part of the work presented in:

"Convolutional neural networks based robust Denoising of Low-Dose Computed Tomography Perfusion maps" Submitted to IEEE TRPMS

Updated on March 14, 2018.