Convolutional neural network-based regression for quantification of brain characteristics using MRI

João Fernandes, Victor Alves, Nadieh Khalili, Manon J.N.L. Benders, Ivana Išgum, Josien Pluim, Pim Moeskops

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Downloads (Pure)

Abstract

Preterm birth is connected to impairments and altered brain growth. Compared to their term born peers, preterm infants have a higher risk of behavioral and cognitive problems since most part of their brain development is in extra-uterine conditions. This paper presents different deep learning approaches with the objective of quantifying the volumes of 8 brain tissues and 5 other image-based descriptors that quantify the state of brain development. Two datasets were used: one with 86 MR brain images of patients around 30 weeks PMA and the other with 153 patients around 40 weeks PMA. Two approaches were evaluated: (1) using the full image as 3D input and (2) using multiple image slices as 3D input, both achieving promising results. A second study, using a dataset of MR brain images of rats, was also performed to assess the performance of this method with other brains. A 2D approach was used to estimate the volumes of 3 rat brain tissues.

Original languageEnglish
Title of host publicationNew Knowledge in Information Systems and Technologies
EditorsLuís Paulo Reis, Sandra Costanzo, Hojjat Adeli, Álvaro Rocha
Place of PublicationBerlin
PublisherSpringer
Pages577-586
Number of pages10
Volume2
ISBN (Print)9783030161835
DOIs
Publication statusPublished - 30 Mar 2019
EventWorld Conference on Information Systems and Technologies, WorldCIST 2019 - Galicia, Spain
Duration: 16 Apr 201919 Apr 2019

Publication series

NameAdvances in Intelligent Systems and Computing
Volume931
ISSN (Print)2194-5357
ISSN (Electronic)2194-5365

Conference

ConferenceWorld Conference on Information Systems and Technologies, WorldCIST 2019
CountrySpain
CityGalicia
Period16/04/1919/04/19

Fingerprint

Magnetic resonance imaging
Brain
Neural networks
Rats
Tissue

Keywords

  • Brain quantification
  • Convolutional neural networks
  • Deep learning
  • Magnetic resonance imaging
  • Preterm infants
  • Rat brain
  • Regression

Cite this

Fernandes, J., Alves, V., Khalili, N., Benders, M. J. N. L., Išgum, I., Pluim, J., & Moeskops, P. (2019). Convolutional neural network-based regression for quantification of brain characteristics using MRI. In L. P. Reis, S. Costanzo, H. Adeli, & Á. Rocha (Eds.), New Knowledge in Information Systems and Technologies (Vol. 2, pp. 577-586). (Advances in Intelligent Systems and Computing; Vol. 931). Berlin: Springer. https://doi.org/10.1007/978-3-030-16184-2_55
Fernandes, João ; Alves, Victor ; Khalili, Nadieh ; Benders, Manon J.N.L. ; Išgum, Ivana ; Pluim, Josien ; Moeskops, Pim. / Convolutional neural network-based regression for quantification of brain characteristics using MRI. New Knowledge in Information Systems and Technologies. editor / Luís Paulo Reis ; Sandra Costanzo ; Hojjat Adeli ; Álvaro Rocha. Vol. 2 Berlin : Springer, 2019. pp. 577-586 (Advances in Intelligent Systems and Computing).
@inproceedings{a20b36629ec84c738d8be015c33e2cc2,
title = "Convolutional neural network-based regression for quantification of brain characteristics using MRI",
abstract = "Preterm birth is connected to impairments and altered brain growth. Compared to their term born peers, preterm infants have a higher risk of behavioral and cognitive problems since most part of their brain development is in extra-uterine conditions. This paper presents different deep learning approaches with the objective of quantifying the volumes of 8 brain tissues and 5 other image-based descriptors that quantify the state of brain development. Two datasets were used: one with 86 MR brain images of patients around 30 weeks PMA and the other with 153 patients around 40 weeks PMA. Two approaches were evaluated: (1) using the full image as 3D input and (2) using multiple image slices as 3D input, both achieving promising results. A second study, using a dataset of MR brain images of rats, was also performed to assess the performance of this method with other brains. A 2D approach was used to estimate the volumes of 3 rat brain tissues.",
keywords = "Brain quantification, Convolutional neural networks, Deep learning, Magnetic resonance imaging, Preterm infants, Rat brain, Regression",
author = "Jo{\~a}o Fernandes and Victor Alves and Nadieh Khalili and Benders, {Manon J.N.L.} and Ivana Išgum and Josien Pluim and Pim Moeskops",
year = "2019",
month = "3",
day = "30",
doi = "10.1007/978-3-030-16184-2_55",
language = "English",
isbn = "9783030161835",
volume = "2",
series = "Advances in Intelligent Systems and Computing",
publisher = "Springer",
pages = "577--586",
editor = "Reis, {Lu{\'i}s Paulo} and Sandra Costanzo and Hojjat Adeli and {\'A}lvaro Rocha",
booktitle = "New Knowledge in Information Systems and Technologies",
address = "Germany",

}

Fernandes, J, Alves, V, Khalili, N, Benders, MJNL, Išgum, I, Pluim, J & Moeskops, P 2019, Convolutional neural network-based regression for quantification of brain characteristics using MRI. in LP Reis, S Costanzo, H Adeli & Á Rocha (eds), New Knowledge in Information Systems and Technologies. vol. 2, Advances in Intelligent Systems and Computing, vol. 931, Springer, Berlin, pp. 577-586, World Conference on Information Systems and Technologies, WorldCIST 2019, Galicia, Spain, 16/04/19. https://doi.org/10.1007/978-3-030-16184-2_55

Convolutional neural network-based regression for quantification of brain characteristics using MRI. / Fernandes, João; Alves, Victor; Khalili, Nadieh; Benders, Manon J.N.L.; Išgum, Ivana; Pluim, Josien; Moeskops, Pim.

New Knowledge in Information Systems and Technologies. ed. / Luís Paulo Reis; Sandra Costanzo; Hojjat Adeli; Álvaro Rocha. Vol. 2 Berlin : Springer, 2019. p. 577-586 (Advances in Intelligent Systems and Computing; Vol. 931).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Convolutional neural network-based regression for quantification of brain characteristics using MRI

AU - Fernandes, João

AU - Alves, Victor

AU - Khalili, Nadieh

AU - Benders, Manon J.N.L.

AU - Išgum, Ivana

AU - Pluim, Josien

AU - Moeskops, Pim

PY - 2019/3/30

Y1 - 2019/3/30

N2 - Preterm birth is connected to impairments and altered brain growth. Compared to their term born peers, preterm infants have a higher risk of behavioral and cognitive problems since most part of their brain development is in extra-uterine conditions. This paper presents different deep learning approaches with the objective of quantifying the volumes of 8 brain tissues and 5 other image-based descriptors that quantify the state of brain development. Two datasets were used: one with 86 MR brain images of patients around 30 weeks PMA and the other with 153 patients around 40 weeks PMA. Two approaches were evaluated: (1) using the full image as 3D input and (2) using multiple image slices as 3D input, both achieving promising results. A second study, using a dataset of MR brain images of rats, was also performed to assess the performance of this method with other brains. A 2D approach was used to estimate the volumes of 3 rat brain tissues.

AB - Preterm birth is connected to impairments and altered brain growth. Compared to their term born peers, preterm infants have a higher risk of behavioral and cognitive problems since most part of their brain development is in extra-uterine conditions. This paper presents different deep learning approaches with the objective of quantifying the volumes of 8 brain tissues and 5 other image-based descriptors that quantify the state of brain development. Two datasets were used: one with 86 MR brain images of patients around 30 weeks PMA and the other with 153 patients around 40 weeks PMA. Two approaches were evaluated: (1) using the full image as 3D input and (2) using multiple image slices as 3D input, both achieving promising results. A second study, using a dataset of MR brain images of rats, was also performed to assess the performance of this method with other brains. A 2D approach was used to estimate the volumes of 3 rat brain tissues.

KW - Brain quantification

KW - Convolutional neural networks

KW - Deep learning

KW - Magnetic resonance imaging

KW - Preterm infants

KW - Rat brain

KW - Regression

UR - http://www.scopus.com/inward/record.url?scp=85065091198&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-16184-2_55

DO - 10.1007/978-3-030-16184-2_55

M3 - Conference contribution

AN - SCOPUS:85065091198

SN - 9783030161835

VL - 2

T3 - Advances in Intelligent Systems and Computing

SP - 577

EP - 586

BT - New Knowledge in Information Systems and Technologies

A2 - Reis, Luís Paulo

A2 - Costanzo, Sandra

A2 - Adeli, Hojjat

A2 - Rocha, Álvaro

PB - Springer

CY - Berlin

ER -

Fernandes J, Alves V, Khalili N, Benders MJNL, Išgum I, Pluim J et al. Convolutional neural network-based regression for quantification of brain characteristics using MRI. In Reis LP, Costanzo S, Adeli H, Rocha Á, editors, New Knowledge in Information Systems and Technologies. Vol. 2. Berlin: Springer. 2019. p. 577-586. (Advances in Intelligent Systems and Computing). https://doi.org/10.1007/978-3-030-16184-2_55