Memory-Efficient Training for Fully Unrolled Deep Learned PET Image Reconstruction with Iteration-Dependent Targets

We propose a new version of the forward-backward splitting expectation-maximisation network (FBSEM-Net) along with a new memory-efficient training method enabling the training of fully unrolled implementations of 3D FBSEM-Net. FBSEM-Net unfolds the maximum a posteriori expectation-maximisation algor...

Descripción completa

Detalles Bibliográficos
Autores principales: Corda-D’Incan, Guillaume, Schnabel, Julia A., Reader, Andrew J.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: 2022
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7612803/
https://www.ncbi.nlm.nih.gov/pubmed/35664091
http://dx.doi.org/10.1109/TRPMS.2021.3101947

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7612803/
https://www.ncbi.nlm.nih.gov/pubmed/35664091
http://dx.doi.org/10.1109/TRPMS.2021.3101947

Ejemplares similares