Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network

Jiawei Chen, Han Zhang, Dong Nie, Li Wang, Gang Li, Weili Lin, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The human cerebellum has been recognized as a key brain structure for motor control and cognitive function regulation. Investigation of brain functional development in the early life has recently been focusing on both cerebral and cerebellar development. Accurate segmentation of the infant cerebellum into different tissues is among the most important steps for quantitative development studies. However, this is extremely challenging due to the weak tissue contrast, extremely folded structures, and severe partial volume effect. To date, there are very few works touching infant cerebellum segmentation. We tackle this challenge by proposing a densely connected convolutional network to learn robust feature representations of different cerebellar tissues towards automatic and accurate segmentation. Specifically, we develop a novel deep neural network architecture by directly connecting all the layers to ensure maximum information flow even among distant layers in the network. This is distinct from all previous studies. Importantly, the outputs from all previous layers are passed to all subsequent layers as contextual features that can guide the segmentation. Our method achieved superior performance than other state-of-the-art methods when applied to Baby Connectome Project (BCP) data consisting of both 6- and 12-month-old infant brain images.

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
EditorsMingxia Liu, Heung-Il Suk, Yinghuan Shi
PublisherSpringer Verlag
Pages233-240
Number of pages8
ISBN (Print)9783030009182
DOIs
Publication statusPublished - 2018 Jan 1
Externally publishedYes
Event9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 2018 Sep 162018 Sep 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11046 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period18/9/1618/9/16

Fingerprint

Cerebellum
Brain
Segmentation
Tissue
Network architecture
Motor Control
Maximum Flow
Information Flow
Network Architecture
Neural Networks
Distinct
Partial
Output

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Chen, J., Zhang, H., Nie, D., Wang, L., Li, G., Lin, W., & Shen, D. (2018). Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network. In M. Liu, H-I. Suk, & Y. Shi (Eds.), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings (pp. 233-240). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11046 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-00919-9_27

Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network. / Chen, Jiawei; Zhang, Han; Nie, Dong; Wang, Li; Li, Gang; Lin, Weili; Shen, Dinggang.

Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. ed. / Mingxia Liu; Heung-Il Suk; Yinghuan Shi. Springer Verlag, 2018. p. 233-240 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11046 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, J, Zhang, H, Nie, D, Wang, L, Li, G, Lin, W & Shen, D 2018, Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network. in M Liu, H-I Suk & Y Shi (eds), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11046 LNCS, Springer Verlag, pp. 233-240, 9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018, Granada, Spain, 18/9/16. https://doi.org/10.1007/978-3-030-00919-9_27
Chen J, Zhang H, Nie D, Wang L, Li G, Lin W et al. Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network. In Liu M, Suk H-I, Shi Y, editors, Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Springer Verlag. 2018. p. 233-240. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-00919-9_27
Chen, Jiawei ; Zhang, Han ; Nie, Dong ; Wang, Li ; Li, Gang ; Lin, Weili ; Shen, Dinggang. / Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network. Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. editor / Mingxia Liu ; Heung-Il Suk ; Yinghuan Shi. Springer Verlag, 2018. pp. 233-240 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{3168aa9330494522bf634c98dc8cf17f,
title = "Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network",
abstract = "The human cerebellum has been recognized as a key brain structure for motor control and cognitive function regulation. Investigation of brain functional development in the early life has recently been focusing on both cerebral and cerebellar development. Accurate segmentation of the infant cerebellum into different tissues is among the most important steps for quantitative development studies. However, this is extremely challenging due to the weak tissue contrast, extremely folded structures, and severe partial volume effect. To date, there are very few works touching infant cerebellum segmentation. We tackle this challenge by proposing a densely connected convolutional network to learn robust feature representations of different cerebellar tissues towards automatic and accurate segmentation. Specifically, we develop a novel deep neural network architecture by directly connecting all the layers to ensure maximum information flow even among distant layers in the network. This is distinct from all previous studies. Importantly, the outputs from all previous layers are passed to all subsequent layers as contextual features that can guide the segmentation. Our method achieved superior performance than other state-of-the-art methods when applied to Baby Connectome Project (BCP) data consisting of both 6- and 12-month-old infant brain images.",
author = "Jiawei Chen and Han Zhang and Dong Nie and Li Wang and Gang Li and Weili Lin and Dinggang Shen",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-00919-9_27",
language = "English",
isbn = "9783030009182",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "233--240",
editor = "Mingxia Liu and Heung-Il Suk and Yinghuan Shi",
booktitle = "Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings",

}

TY - GEN

T1 - Automatic accurate infant cerebellar tissue segmentation with densely connected convolutional network

AU - Chen, Jiawei

AU - Zhang, Han

AU - Nie, Dong

AU - Wang, Li

AU - Li, Gang

AU - Lin, Weili

AU - Shen, Dinggang

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The human cerebellum has been recognized as a key brain structure for motor control and cognitive function regulation. Investigation of brain functional development in the early life has recently been focusing on both cerebral and cerebellar development. Accurate segmentation of the infant cerebellum into different tissues is among the most important steps for quantitative development studies. However, this is extremely challenging due to the weak tissue contrast, extremely folded structures, and severe partial volume effect. To date, there are very few works touching infant cerebellum segmentation. We tackle this challenge by proposing a densely connected convolutional network to learn robust feature representations of different cerebellar tissues towards automatic and accurate segmentation. Specifically, we develop a novel deep neural network architecture by directly connecting all the layers to ensure maximum information flow even among distant layers in the network. This is distinct from all previous studies. Importantly, the outputs from all previous layers are passed to all subsequent layers as contextual features that can guide the segmentation. Our method achieved superior performance than other state-of-the-art methods when applied to Baby Connectome Project (BCP) data consisting of both 6- and 12-month-old infant brain images.

AB - The human cerebellum has been recognized as a key brain structure for motor control and cognitive function regulation. Investigation of brain functional development in the early life has recently been focusing on both cerebral and cerebellar development. Accurate segmentation of the infant cerebellum into different tissues is among the most important steps for quantitative development studies. However, this is extremely challenging due to the weak tissue contrast, extremely folded structures, and severe partial volume effect. To date, there are very few works touching infant cerebellum segmentation. We tackle this challenge by proposing a densely connected convolutional network to learn robust feature representations of different cerebellar tissues towards automatic and accurate segmentation. Specifically, we develop a novel deep neural network architecture by directly connecting all the layers to ensure maximum information flow even among distant layers in the network. This is distinct from all previous studies. Importantly, the outputs from all previous layers are passed to all subsequent layers as contextual features that can guide the segmentation. Our method achieved superior performance than other state-of-the-art methods when applied to Baby Connectome Project (BCP) data consisting of both 6- and 12-month-old infant brain images.

UR - http://www.scopus.com/inward/record.url?scp=85054477885&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054477885&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00919-9_27

DO - 10.1007/978-3-030-00919-9_27

M3 - Conference contribution

AN - SCOPUS:85054477885

SN - 9783030009182

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 233

EP - 240

BT - Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings

A2 - Liu, Mingxia

A2 - Suk, Heung-Il

A2 - Shi, Yinghuan

PB - Springer Verlag

ER -