Fetal MRI by Robust Deep Generative Prior Reconstruction and Diffeomorphic Registration

Lucilio Cordero-Grande; Juan Enrique Ortuño-Fisac; Alejandra Aguado del Hoyo; Alena Uus; Maria Deprez; Andres Santos; Joseph V. Hajnal; María J. Ledesma-Carbayo

doi:10.1109/TMI.2022.3217725

Fetal MRI by Robust Deep Generative Prior Reconstruction and Diffeomorphic Registration

Lucilio Cordero-Grande^*, Juan Enrique Ortuño-Fisac, Alejandra Aguado del Hoyo, Alena Uus, Maria Deprez, Andres Santos, Joseph V. Hajnal, María J. Ledesma-Carbayo

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

7 Citations (Scopus)

Abstract

Magnetic resonance imaging of whole fetal body and placenta is limited by different sources of motion affecting the womb. Usual scanning techniques employ single-shot multi-slice sequences where anatomical information in different slices may be subject to different deformations, contrast variations or artifacts. Volumetric reconstruction formulations have been proposed to correct for these factors, but they must accommodate a non-homogeneous and non-isotropic sampling, so regularization becomes necessary. Thus, in this paper we propose a deep generative prior for robust volumetric reconstructions integrated with a diffeomorphic volume to slice registration method. Experiments are performed to validate our contributions and compare with ifdefined tmiformat R2.5a state of the art method methods in the literature in a cohort of 72 fetal datasets in the range of 20-36 weeks gestational age. Results suggest improved image resolution Quantitative as well as radiological assessment suggest improved image quality and more accurate prediction of gestational age at scan is obtained when comparing to a state of the art reconstruction method methods. In addition, gestational age prediction results from our volumetric reconstructions compare favourably are competitive with existing brain-based approaches, with boosted accuracy when integrating information of organs other than the brain. Namely, a mean absolute error of 0.618 weeks (R²=0.958) is achieved when combining fetal brain and trunk information.

Original language	English
Pages (from-to)	810-822
Number of pages	13
Journal	IEEE Transactions on Medical Imaging
Volume	42
Issue number	3
Early online date	26 Oct 2022
DOIs	https://doi.org/10.1109/TMI.2022.3217725
Publication status	Published - Mar 2023

Keywords

diffeomorphic image registration
Fetal magnetic resonance imaging
generative image priors
gestational age prediction
slice to volume reconstruction

Access to Document

10.1109/TMI.2022.3217725Licence: CC BY

Cite this

@article{3f71f758393f4c288fa0a2f4418019a0,

title = "Fetal MRI by Robust Deep Generative Prior Reconstruction and Diffeomorphic Registration",

abstract = "Magnetic resonance imaging of whole fetal body and placenta is limited by different sources of motion affecting the womb. Usual scanning techniques employ single-shot multi-slice sequences where anatomical information in different slices may be subject to different deformations, contrast variations or artifacts. Volumetric reconstruction formulations have been proposed to correct for these factors, but they must accommodate a non-homogeneous and non-isotropic sampling, so regularization becomes necessary. Thus, in this paper we propose a deep generative prior for robust volumetric reconstructions integrated with a diffeomorphic volume to slice registration method. Experiments are performed to validate our contributions and compare with ifdefined tmiformat R2.5a state of the art method methods in the literature in a cohort of 72 fetal datasets in the range of 20-36 weeks gestational age. Results suggest improved image resolution Quantitative as well as radiological assessment suggest improved image quality and more accurate prediction of gestational age at scan is obtained when comparing to a state of the art reconstruction method methods. In addition, gestational age prediction results from our volumetric reconstructions compare favourably are competitive with existing brain-based approaches, with boosted accuracy when integrating information of organs other than the brain. Namely, a mean absolute error of 0.618 weeks (R2=0.958) is achieved when combining fetal brain and trunk information.",

keywords = "diffeomorphic image registration, Fetal magnetic resonance imaging, generative image priors, gestational age prediction, slice to volume reconstruction",

author = "Lucilio Cordero-Grande and Ortu{\~n}o-Fisac, {Juan Enrique} and {del Hoyo}, {Alejandra Aguado} and Alena Uus and Maria Deprez and Andres Santos and Hajnal, {Joseph V.} and Ledesma-Carbayo, {Mar{\'i}a J.}",

year = "2023",

month = mar,

doi = "10.1109/TMI.2022.3217725",

language = "English",

volume = "42",

pages = "810--822",

journal = "IEEE Transactions on Medical Imaging",

issn = "0278-0062",

publisher = "I E E E",

number = "3",

}

TY - JOUR

T1 - Fetal MRI by Robust Deep Generative Prior Reconstruction and Diffeomorphic Registration

AU - Cordero-Grande, Lucilio

AU - Ortuño-Fisac, Juan Enrique

AU - del Hoyo, Alejandra Aguado

AU - Uus, Alena

AU - Deprez, Maria

AU - Santos, Andres

AU - Hajnal, Joseph V.

AU - Ledesma-Carbayo, María J.

PY - 2023/3

Y1 - 2023/3

N2 - Magnetic resonance imaging of whole fetal body and placenta is limited by different sources of motion affecting the womb. Usual scanning techniques employ single-shot multi-slice sequences where anatomical information in different slices may be subject to different deformations, contrast variations or artifacts. Volumetric reconstruction formulations have been proposed to correct for these factors, but they must accommodate a non-homogeneous and non-isotropic sampling, so regularization becomes necessary. Thus, in this paper we propose a deep generative prior for robust volumetric reconstructions integrated with a diffeomorphic volume to slice registration method. Experiments are performed to validate our contributions and compare with ifdefined tmiformat R2.5a state of the art method methods in the literature in a cohort of 72 fetal datasets in the range of 20-36 weeks gestational age. Results suggest improved image resolution Quantitative as well as radiological assessment suggest improved image quality and more accurate prediction of gestational age at scan is obtained when comparing to a state of the art reconstruction method methods. In addition, gestational age prediction results from our volumetric reconstructions compare favourably are competitive with existing brain-based approaches, with boosted accuracy when integrating information of organs other than the brain. Namely, a mean absolute error of 0.618 weeks (R2=0.958) is achieved when combining fetal brain and trunk information.

AB - Magnetic resonance imaging of whole fetal body and placenta is limited by different sources of motion affecting the womb. Usual scanning techniques employ single-shot multi-slice sequences where anatomical information in different slices may be subject to different deformations, contrast variations or artifacts. Volumetric reconstruction formulations have been proposed to correct for these factors, but they must accommodate a non-homogeneous and non-isotropic sampling, so regularization becomes necessary. Thus, in this paper we propose a deep generative prior for robust volumetric reconstructions integrated with a diffeomorphic volume to slice registration method. Experiments are performed to validate our contributions and compare with ifdefined tmiformat R2.5a state of the art method methods in the literature in a cohort of 72 fetal datasets in the range of 20-36 weeks gestational age. Results suggest improved image resolution Quantitative as well as radiological assessment suggest improved image quality and more accurate prediction of gestational age at scan is obtained when comparing to a state of the art reconstruction method methods. In addition, gestational age prediction results from our volumetric reconstructions compare favourably are competitive with existing brain-based approaches, with boosted accuracy when integrating information of organs other than the brain. Namely, a mean absolute error of 0.618 weeks (R2=0.958) is achieved when combining fetal brain and trunk information.

KW - diffeomorphic image registration

KW - Fetal magnetic resonance imaging

KW - generative image priors

KW - gestational age prediction

KW - slice to volume reconstruction

UR - http://www.scopus.com/inward/record.url?scp=85141476278&partnerID=8YFLogxK

U2 - 10.1109/TMI.2022.3217725

DO - 10.1109/TMI.2022.3217725

M3 - Article

C2 - 36288233

AN - SCOPUS:85141476278

SN - 0278-0062

VL - 42

SP - 810

EP - 822

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

IS - 3

ER -

Fetal MRI by Robust Deep Generative Prior Reconstruction and Diffeomorphic Registration

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this