Skip to content

Restore low-dose DBT projections using VCT software

License

Notifications You must be signed in to change notification settings

LAVI-USP/SPIE2022

Repository files navigation

Assessment of training strategies for convolutional neural network to restore low-dose digital breast tomosynthesis projections

======

This repository contains the training and testing codes for the paper "Assessment of training strategies for convolutional neural network to restore low-dose digital breast tomosynthesis projections", submitted to the SPIE Medical Imaging 2022 conference. We used the OpenVCT from the University of Pennsylvania, available here. Also, we used a model-based (MB) restoration as a benchmark, also available here, which uses the commonly known BM3D.

Network architecture:

Some results:

Soon

Reference:

Rodrigo de Barros Vimieiro, Lucas Rodrigues Borges, Bruno Barufaldi, Andrew D. A. Maidment, Ge Wang, and Marcelo Andrade da Costa Vieira "Assessment of training strategies for convolutional neural network to restore low-dose digital breast tomosynthesis projections", Proc. SPIE 12031, Medical Imaging 2022: Physics of Medical Imaging, (4 April 2022); https://doi.org/10.1117/12.2609945

@inproceedings{vimieiro2022assessment,
author = {Rodrigo de Barros Vimieiro and Lucas Rodrigues Borges and Bruno Barufaldi and Andrew D. A. Maidment and Ge Wang and Marcelo Andrade da Costa Vieira},
title = {{Assessment of training strategies for convolutional neural network to restore low-dose digital breast tomosynthesis projections}},
volume = {12031},
booktitle = {Medical Imaging 2022: Physics of Medical Imaging},
organization = {International Society for Optics and Photonics},
publisher = {SPIE},
year = {2022},
doi = {10.1117/12.2609945},
URL = {https://doi.org/10.1117/12.2609945}
}

Laboratory of Computer Vision (Lavi)
Department of Electrical and Computer Engineering
São Carlos School of Engineering, University of São Paulo
São Carlos - Brazil

AI-based X-ray Imaging System (AXIS)
Department of Biomedical Engineering
Rensselaer Polytechnic Institute
Troy - USA