Deep Fully Convolutional Network for MR Fingerprinting

Dongdong Chen; Mohammad Golbabaee; Pedro A.  Gómez; Marion I. Menzel; Mike Davies

Deep Fully Convolutional Network for MR Fingerprinting

Dongdong Chen, Mohammad Golbabaee, Pedro A. Gómez, Marion I. Menzel, Mike Davies

University of Edinburgh

Research output: Contribution to conference › Paper › peer-review

Abstract

This work proposes an end-to-end deep fully convolutional neural network for MRF reconstruction (MRF-FCNN), which firstly employs linear dimensionality reduction and then uses a neural network to project the data into the tissue parameters. The MRF dictionary is only used for training the network and not during image reconstruction. We show that MRF-FCNN is capable of achieving accuracy comparable to the ground-truth maps thanks to capturing spatio-temporal data structures without a need for the non-scalable dictionary matching step used in the baseline reconstructions.

Original language	English
Publication status	Published - 10 Jul 2019

Access to Document

https://openreview.net/pdf?id=SJxUdvJTtNLicence: CC BY

Cite this

@conference{f6749ac85f7b47ac9cbf3deec1a1755e,

title = "Deep Fully Convolutional Network for MR Fingerprinting",

abstract = "This work proposes an end-to-end deep fully convolutional neural network for MRF reconstruction (MRF-FCNN), which firstly employs linear dimensionality reduction and then uses a neural network to project the data into the tissue parameters. The MRF dictionary is only used for training the network and not during image reconstruction. We show that MRF-FCNN is capable of achieving accuracy comparable to the ground-truth maps thanks to capturing spatio-temporal data structures without a need for the non-scalable dictionary matching step used in the baseline reconstructions.",

author = "Dongdong Chen and Mohammad Golbabaee and G{\'o}mez, {Pedro A.} and Menzel, {Marion I.} and Mike Davies",

year = "2019",

month = jul,

day = "10",

language = "English",

}

TY - CONF

T1 - Deep Fully Convolutional Network for MR Fingerprinting

AU - Chen, Dongdong

AU - Golbabaee, Mohammad

AU - Gómez, Pedro A.

AU - Menzel, Marion I.

AU - Davies, Mike

PY - 2019/7/10

Y1 - 2019/7/10

N2 - This work proposes an end-to-end deep fully convolutional neural network for MRF reconstruction (MRF-FCNN), which firstly employs linear dimensionality reduction and then uses a neural network to project the data into the tissue parameters. The MRF dictionary is only used for training the network and not during image reconstruction. We show that MRF-FCNN is capable of achieving accuracy comparable to the ground-truth maps thanks to capturing spatio-temporal data structures without a need for the non-scalable dictionary matching step used in the baseline reconstructions.

AB - This work proposes an end-to-end deep fully convolutional neural network for MRF reconstruction (MRF-FCNN), which firstly employs linear dimensionality reduction and then uses a neural network to project the data into the tissue parameters. The MRF dictionary is only used for training the network and not during image reconstruction. We show that MRF-FCNN is capable of achieving accuracy comparable to the ground-truth maps thanks to capturing spatio-temporal data structures without a need for the non-scalable dictionary matching step used in the baseline reconstructions.

M3 - Paper

ER -