Improving DCE-MRI through Unfolded Low-Rank + Sparse Optimisation

A method for perfusion imaging with DCE-MRI is developed based on a combination of two popular paradigms: the low-rank + sparse model for optimisation-based reconstruction, and the deep unfolding. A learnable algorithm derived from a proximal method is designed with emphasis on simplicity and interp...

Full description

Saved in:
Bibliographic Details
Main Authors: Mokry, Ondrej, Vitous, Jiri, Rajmic, Pavel, Jirik, Radovan
Format: Conference Proceeding
Language:English
Subjects:
Adaptation models
Analytical models
Biological system modeling
Computational modeling
DCE-MRI
deep unfolding
L+S model
perfusion analysis
proximal splitting algorithms
Rats
Thresholding (Imaging)
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A method for perfusion imaging with DCE-MRI is developed based on a combination of two popular paradigms: the low-rank + sparse model for optimisation-based reconstruction, and the deep unfolding. A learnable algorithm derived from a proximal method is designed with emphasis on simplicity and interpretability. The resulting deep network is trained and evaluated using a simulated measurement of a rat with a brain tumor, showing large performance gain over the classical low-rank + sparse baseline. Moreover, a quantitative perfusion analysis is performed based on the reconstructed sequence, proving that even training based on a simple pixel-wise error can lead to a significant improvement of the quality of the perfusion maps.
ISSN:1945-8452
DOI:10.1109/ISBI56570.2024.10635295