Publication
Title
Sparse-view medical tomosynthesis via mixed scale dense convolutional framelet networks
Author
Abstract
X-ray tomosynthesis is a low-dose and relatively inexpensive 3D imaging technique that relies on limited-angle and sparse-view tomography. Unfortunately, tomosynthesis often leads to reconstructed images that are corrupted by ripple artifacts. The current state-of-the-art for artifact suppression in tomographic data involves the use of Convolutional Neural Networks for mapping corrupted reconstructions into artifact-free images. Recently, Deep Convolutional Framelet Networks (DCFNs) were proposed in which max-pooling layers in the U-net were replaced by fixed Wavelet decompositions. In this work, we show that replacing the regular convolutional blocks in the DCFNs by Mixed Scaled Dense (MSD) blocks for exploiting multi-scale features allows us to better represent and hence suppress tomosynthesis artifacts at different scales. Experiments using simulated data show that our Mixed Scale Dense Convolutional Framelet Network (MSDCFN) outperforms the state-of-the-art methods in the vast majority of the tomosynthesis scans evaluated.
Language
English
Source (journal)
Proceedings. - Piscataway, NJ, 2002, currens
Source (book)
20th IEEE International Symposium on Biomedical Imaging (ISBI), APR 18-21, 2023, Cartagena, Colombia
Publication
New york : Ieee , 2023
ISBN
978-1-6654-7358-3
DOI
10.1109/ISBI53787.2023.10230645
Volume/pages
(2023) , p. 1-5
ISI
001062050500322
Full text (Publisher's DOI)
Full text (publisher's version - intranet only)
UAntwerpen
Faculty/Department
Research group
Project info
Next generation X-ray metrology for meeting industry standards (MetroFlex).
Publication type
Subject
Affiliation
Publications with a UAntwerp address
External links
Web of Science
Record
Identifier
Creation 04.12.2023
Last edited 08.12.2023
To cite this reference