Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework

Abstract: 

Model-Based Iterative Reconstruction (MBIR) has shown promising results in clinical studies as they allow significant
dose reduction during CT scans while maintaining the diagnostic image quality. MBIR improves the image quality over analytical reconstruction by modeling both the sensor (e.g., forward model) and the image being reconstructed (e.g., prior model). While the forward model is typically based on the physics of the sensor, accurate prior modeling remains a challenging problem. Markov Random Field (MRF) has been widely used as prior models in MBIR due to simple structure, but they cannot completely capture the subtle characteristics of complex images. To tackle this challenge, we generate a prior model by learning the desirable image property from a large dataset. Toward this, we use Plug-and-Play (PnP) framework which decouples the forward model and the prior model in the optimization procedure, replacing the prior model optimization by a image denoising operator. Then, we adopt the state-of-the-art deep residual learning for the image denoising operator which represents the prior model
in MBIR. Experimental results on real CT scans demonstrate that our PnP MBIR with deep residual learning prior significantly
reduces the noise and artifacts compared to analytical reconstruction and standard MBIR with MRF prior.

up
0 users have voted:

Paper Details

Authors:
Dong Hye Ye, Somesh Srivastava, Jean-Baptiste Thibault, Ken Sauer, Charles Bouman
Submitted On:
19 April 2018 - 7:12pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Dong Hye Ye
Paper Code:
1370
Document Year:
2018
Cite

Document Files

ICASSP_DongHyeYe.pdf

(58 downloads)

Keywords

Additional Categories

Subscribe

[1] Dong Hye Ye, Somesh Srivastava, Jean-Baptiste Thibault, Ken Sauer, Charles Bouman, "Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3038. Accessed: Aug. 19, 2018.
@article{3038-18,
url = {http://sigport.org/3038},
author = {Dong Hye Ye; Somesh Srivastava; Jean-Baptiste Thibault; Ken Sauer; Charles Bouman },
publisher = {IEEE SigPort},
title = {Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework},
year = {2018} }
TY - EJOUR
T1 - Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework
AU - Dong Hye Ye; Somesh Srivastava; Jean-Baptiste Thibault; Ken Sauer; Charles Bouman
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3038
ER -
Dong Hye Ye, Somesh Srivastava, Jean-Baptiste Thibault, Ken Sauer, Charles Bouman. (2018). Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework. IEEE SigPort. http://sigport.org/3038
Dong Hye Ye, Somesh Srivastava, Jean-Baptiste Thibault, Ken Sauer, Charles Bouman, 2018. Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework. Available at: http://sigport.org/3038.
Dong Hye Ye, Somesh Srivastava, Jean-Baptiste Thibault, Ken Sauer, Charles Bouman. (2018). "Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework." Web.
1. Dong Hye Ye, Somesh Srivastava, Jean-Baptiste Thibault, Ken Sauer, Charles Bouman. Deep Residual Learning for Model-Based Iterative CT Reconstruction using Plug-and-Play Framework [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3038