Evaluation of automatic neonatal brain segmentation algorithms: The NeoBrainS12 challenge

Ivana Išgum; Manon J N L Benders; Brian Avants; M Jorge Cardoso; Serena J Counsell; Elda Fischi Gomez; Laura Gui; Petra S Hűppi; Karina J Kersbergen; Antonios Makropoulos; Andrew Melbourne; Pim Moeskops; Christian P Mol; Maria Kuklisova-Murgasova; Daniel Rueckert; Julia A Schnabel; Vedran Srhoj-Egekher; Jue Wu; Siying Wang; Linda S de Vries; Max A Viergever

doi:10.1016/j.media.2014.11.001

Evaluation of automatic neonatal brain segmentation algorithms: The NeoBrainS12 challenge

Ivana Išgum^*, Manon J N L Benders, Brian Avants, M Jorge Cardoso, Serena J Counsell, Elda Fischi Gomez, Laura Gui, Petra S Hűppi, Karina J Kersbergen, Antonios Makropoulos, Andrew Melbourne, Pim Moeskops, Christian P Mol, Maria Kuklisova-Murgasova, Daniel Rueckert, Julia A Schnabel, Vedran Srhoj-Egekher, Jue Wu, Siying Wang, Linda S de VriesMax A Viergever

^*Corresponding author for this work

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

80 Citations (Scopus)

Abstract

A number of algorithms for brain segmentation in preterm born infants have been published, but a reliable comparison of their performance is lacking. The NeoBrainS12 study (http://neobrains12.isi.uu.nl), providing three different image sets of preterm born infants, was set up to provide such a comparison. These sets are (i) axial scans acquired at 40 weeks corrected age, (ii) coronal scans acquired at 30 weeks corrected age and (iii) coronal scans acquired at 40 weeks corrected age. Each of these three sets consists of three T1- and T2-weighted MR images of the brain acquired with a 3T MRI scanner. The task was to segment cortical grey matter, non-myelinated and myelinated white matter, brainstem, basal ganglia and thalami, cerebellum, and cerebrospinal fluid in the ventricles and in the extracerebral space separately. Any team could upload the results and all segmentations were evaluated in the same way. This paper presents the results of eight participating teams. The results demonstrate that the participating methods were able to segment all tissue classes well, except myelinated white matter.

Original language	English
Pages (from-to)	135-151
Number of pages	17
Journal	Medical Image Analysis
Volume	20
Issue number	1
Early online date	15 Nov 2014
DOIs	https://doi.org/10.1016/j.media.2014.11.001
Publication status	Published - 1 Feb 2015

Keywords

Brain segmentation
MRI
Neonatal brain
Segmentation comparison
Segmentation evaluation

Access to Document

10.1016/j.media.2014.11.001

Cite this

Išgum, I., Benders, M. J. N. L., Avants, B., Cardoso, M. J., Counsell, S. J., Gomez, E. F., Gui, L., Hűppi, P. S., Kersbergen, K. J., Makropoulos, A., Melbourne, A., Moeskops, P., Mol, C. P., Kuklisova-Murgasova, M., Rueckert, D., Schnabel, J. A., Srhoj-Egekher, V., Wu, J., Wang, S., ... Viergever, M. A. (2015). Evaluation of automatic neonatal brain segmentation algorithms: The NeoBrainS12 challenge. Medical Image Analysis, 20(1), 135-151. https://doi.org/10.1016/j.media.2014.11.001

@article{8afc6c449866415d83aed0f67b7103e2,

title = "Evaluation of automatic neonatal brain segmentation algorithms: The NeoBrainS12 challenge",

abstract = "A number of algorithms for brain segmentation in preterm born infants have been published, but a reliable comparison of their performance is lacking. The NeoBrainS12 study (http://neobrains12.isi.uu.nl), providing three different image sets of preterm born infants, was set up to provide such a comparison. These sets are (i) axial scans acquired at 40 weeks corrected age, (ii) coronal scans acquired at 30 weeks corrected age and (iii) coronal scans acquired at 40 weeks corrected age. Each of these three sets consists of three T1- and T2-weighted MR images of the brain acquired with a 3T MRI scanner. The task was to segment cortical grey matter, non-myelinated and myelinated white matter, brainstem, basal ganglia and thalami, cerebellum, and cerebrospinal fluid in the ventricles and in the extracerebral space separately. Any team could upload the results and all segmentations were evaluated in the same way. This paper presents the results of eight participating teams. The results demonstrate that the participating methods were able to segment all tissue classes well, except myelinated white matter.",

keywords = "Brain segmentation, MRI, Neonatal brain, Segmentation comparison, Segmentation evaluation",

author = "Ivana I{\v s}gum and Benders, {Manon J N L} and Brian Avants and Cardoso, {M Jorge} and Counsell, {Serena J} and Gomez, {Elda Fischi} and Laura Gui and H{\H u}ppi, {Petra S} and Kersbergen, {Karina J} and Antonios Makropoulos and Andrew Melbourne and Pim Moeskops and Mol, {Christian P} and Maria Kuklisova-Murgasova and Daniel Rueckert and Schnabel, {Julia A} and Vedran Srhoj-Egekher and Jue Wu and Siying Wang and {de Vries}, {Linda S} and Viergever, {Max A}",

year = "2015",

month = feb,

day = "1",

doi = "10.1016/j.media.2014.11.001",

language = "English",

volume = "20",

pages = "135--151",

journal = "Medical Image Analysis",

issn = "1361-8415",

publisher = "Elsevier",

number = "1",

}

Išgum, I, Benders, MJNL, Avants, B, Cardoso, MJ , Counsell, SJ, Gomez, EF, Gui, L, Hűppi, PS, Kersbergen, KJ, Makropoulos, A, Melbourne, A, Moeskops, P, Mol, CP, Kuklisova-Murgasova, M, Rueckert, D, Schnabel, JA, Srhoj-Egekher, V, Wu, J, Wang, S, de Vries, LS & Viergever, MA 2015, 'Evaluation of automatic neonatal brain segmentation algorithms: The NeoBrainS12 challenge', Medical Image Analysis, vol. 20, no. 1, pp. 135-151. https://doi.org/10.1016/j.media.2014.11.001

TY - JOUR

T1 - Evaluation of automatic neonatal brain segmentation algorithms

T2 - The NeoBrainS12 challenge

AU - Išgum, Ivana

AU - Benders, Manon J N L

AU - Avants, Brian

AU - Cardoso, M Jorge

AU - Counsell, Serena J

AU - Gomez, Elda Fischi

AU - Gui, Laura

AU - Hűppi, Petra S

AU - Kersbergen, Karina J

AU - Makropoulos, Antonios

AU - Melbourne, Andrew

AU - Moeskops, Pim

AU - Mol, Christian P

AU - Kuklisova-Murgasova, Maria

AU - Rueckert, Daniel

AU - Schnabel, Julia A

AU - Srhoj-Egekher, Vedran

AU - Wu, Jue

AU - Wang, Siying

AU - de Vries, Linda S

AU - Viergever, Max A

PY - 2015/2/1

Y1 - 2015/2/1

N2 - A number of algorithms for brain segmentation in preterm born infants have been published, but a reliable comparison of their performance is lacking. The NeoBrainS12 study (http://neobrains12.isi.uu.nl), providing three different image sets of preterm born infants, was set up to provide such a comparison. These sets are (i) axial scans acquired at 40 weeks corrected age, (ii) coronal scans acquired at 30 weeks corrected age and (iii) coronal scans acquired at 40 weeks corrected age. Each of these three sets consists of three T1- and T2-weighted MR images of the brain acquired with a 3T MRI scanner. The task was to segment cortical grey matter, non-myelinated and myelinated white matter, brainstem, basal ganglia and thalami, cerebellum, and cerebrospinal fluid in the ventricles and in the extracerebral space separately. Any team could upload the results and all segmentations were evaluated in the same way. This paper presents the results of eight participating teams. The results demonstrate that the participating methods were able to segment all tissue classes well, except myelinated white matter.

AB - A number of algorithms for brain segmentation in preterm born infants have been published, but a reliable comparison of their performance is lacking. The NeoBrainS12 study (http://neobrains12.isi.uu.nl), providing three different image sets of preterm born infants, was set up to provide such a comparison. These sets are (i) axial scans acquired at 40 weeks corrected age, (ii) coronal scans acquired at 30 weeks corrected age and (iii) coronal scans acquired at 40 weeks corrected age. Each of these three sets consists of three T1- and T2-weighted MR images of the brain acquired with a 3T MRI scanner. The task was to segment cortical grey matter, non-myelinated and myelinated white matter, brainstem, basal ganglia and thalami, cerebellum, and cerebrospinal fluid in the ventricles and in the extracerebral space separately. Any team could upload the results and all segmentations were evaluated in the same way. This paper presents the results of eight participating teams. The results demonstrate that the participating methods were able to segment all tissue classes well, except myelinated white matter.

KW - Brain segmentation

KW - MRI

KW - Neonatal brain

KW - Segmentation comparison

KW - Segmentation evaluation

UR - http://www.scopus.com/inward/record.url?scp=84920895738&partnerID=8YFLogxK

U2 - 10.1016/j.media.2014.11.001

DO - 10.1016/j.media.2014.11.001

M3 - Article

C2 - 25487610

AN - SCOPUS:84920895738

SN - 1361-8415

VL - 20

SP - 135

EP - 151

JO - Medical Image Analysis

JF - Medical Image Analysis

IS - 1

ER -

Evaluation of automatic neonatal brain segmentation algorithms: The NeoBrainS12 challenge

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this