The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data

David Iclănzan; Rodica Ioana Lung; Zsolt-Levente Kucsván; Béla Surányi; Levente Kovács; László Szilágyi

doi:10.1109/AFRICON51333.2021.9570952

The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data

David Iclănzan, Rodica Ioana Lung, Zsolt-Levente Kucsván, Béla Surányi, Levente Kovács, László Szilágyi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

2 Citations (Scopus)

Abstract

Atlas assisted image segmentation has been quite popular in medical imaging during the last two decades. The atlas is able to provide prior information on the imaged organ's shape, appearance, and local texture or intensity distribution. In case of segmenting images via pixelwise classification, the final segmentation result is obtained through a fusion of the classification outcome with the local atlas information. In other words, the atlas guides the classifier towards the shape of local structures normally situated at the given location. This paper proposes to demonstrate the advantages a multi-atlas can bring in a segmentation process of the main tissues in infant brain based on multi-spectral MRI records. Three supervised machine learning methods are deployed to segment brain tissues, with and without the use of the atlas. Differences are evaluated using statistical accuracy indicators. Atlases improved the overall segmentation accuracy by 2.5-3.5%, depending on the deployed classifier method.

Original language	English
Title of host publication	Proceedings of 2021 IEEE AFRICON
Place of Publication	Piscataway, NJ
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	978-1-6654-1984-0
ISBN (Print)	978-1-6654-4748-5, 978-1-6654-1983-3 (USB)
DOIs	https://doi.org/10.1109/AFRICON51333.2021.9570952
Publication status	Published - 13 Sept 2021
Externally published	Yes
Event	IEEE AFRICON, AFRICON 2021 - Virtual, Arusha, Tanzania, United Republic of Duration: 13 Sept 2021 → 15 Sept 2021

Publication series

Name	IEEE AFRICON Conference
Publisher	IEEE
Volume	2021
ISSN (Print)	2153-0025
ISSN (Electronic)	2153-0033

Conference

Conference	IEEE AFRICON, AFRICON 2021
Country/Territory	Tanzania, United Republic of
City	Virtual, Arusha
Period	13/09/21 → 15/09/21

Keywords

Atlas-based segmentation
Brain tissue segmentation
Infant brain
Magnetic resonance imaging
n/a OA procedure

Access to Document

10.1109/AFRICON51333.2021.9570952

Cite this

@inproceedings{e6faed1e142c4882a6419311fa71f9d8,

title = "The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data",

abstract = "Atlas assisted image segmentation has been quite popular in medical imaging during the last two decades. The atlas is able to provide prior information on the imaged organ's shape, appearance, and local texture or intensity distribution. In case of segmenting images via pixelwise classification, the final segmentation result is obtained through a fusion of the classification outcome with the local atlas information. In other words, the atlas guides the classifier towards the shape of local structures normally situated at the given location. This paper proposes to demonstrate the advantages a multi-atlas can bring in a segmentation process of the main tissues in infant brain based on multi-spectral MRI records. Three supervised machine learning methods are deployed to segment brain tissues, with and without the use of the atlas. Differences are evaluated using statistical accuracy indicators. Atlases improved the overall segmentation accuracy by 2.5-3.5%, depending on the deployed classifier method.",

keywords = "Atlas-based segmentation, Brain tissue segmentation, Infant brain, Magnetic resonance imaging, n/a OA procedure",

author = "David Icl{\u a}nzan and Lung, {Rodica Ioana} and Zsolt-Levente Kucsv{\'a}n and B{\'e}la Sur{\'a}nyi and Levente Kov{\'a}cs and L{\'a}szl{\'o} Szil{\'a}gyi",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; IEEE AFRICON, AFRICON 2021 ; Conference date: 13-09-2021 Through 15-09-2021",

year = "2021",

month = sep,

day = "13",

doi = "10.1109/AFRICON51333.2021.9570952",

language = "English",

isbn = "978-1-6654-4748-5",

series = "IEEE AFRICON Conference",

publisher = "IEEE",

booktitle = "Proceedings of 2021 IEEE AFRICON",

address = "United States",

}

Iclănzan, D, Lung, RI, Kucsván, Z-L, Surányi, B, Kovács, L & Szilágyi, L 2021, The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data. in Proceedings of 2021 IEEE AFRICON. IEEE AFRICON Conference, vol. 2021, IEEE, Piscataway, NJ, IEEE AFRICON, AFRICON 2021, Virtual, Arusha, Tanzania, United Republic of, 13/09/21. https://doi.org/10.1109/AFRICON51333.2021.9570952

The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data. / Iclănzan, David; Lung, Rodica Ioana; Kucsván, Zsolt-Levente et al.
Proceedings of 2021 IEEE AFRICON. Piscataway, NJ: IEEE, 2021. (IEEE AFRICON Conference; Vol. 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data

AU - Iclănzan, David

AU - Lung, Rodica Ioana

AU - Kucsván, Zsolt-Levente

AU - Surányi, Béla

AU - Kovács, Levente

AU - Szilágyi, László

PY - 2021/9/13

Y1 - 2021/9/13

N2 - Atlas assisted image segmentation has been quite popular in medical imaging during the last two decades. The atlas is able to provide prior information on the imaged organ's shape, appearance, and local texture or intensity distribution. In case of segmenting images via pixelwise classification, the final segmentation result is obtained through a fusion of the classification outcome with the local atlas information. In other words, the atlas guides the classifier towards the shape of local structures normally situated at the given location. This paper proposes to demonstrate the advantages a multi-atlas can bring in a segmentation process of the main tissues in infant brain based on multi-spectral MRI records. Three supervised machine learning methods are deployed to segment brain tissues, with and without the use of the atlas. Differences are evaluated using statistical accuracy indicators. Atlases improved the overall segmentation accuracy by 2.5-3.5%, depending on the deployed classifier method.

AB - Atlas assisted image segmentation has been quite popular in medical imaging during the last two decades. The atlas is able to provide prior information on the imaged organ's shape, appearance, and local texture or intensity distribution. In case of segmenting images via pixelwise classification, the final segmentation result is obtained through a fusion of the classification outcome with the local atlas information. In other words, the atlas guides the classifier towards the shape of local structures normally situated at the given location. This paper proposes to demonstrate the advantages a multi-atlas can bring in a segmentation process of the main tissues in infant brain based on multi-spectral MRI records. Three supervised machine learning methods are deployed to segment brain tissues, with and without the use of the atlas. Differences are evaluated using statistical accuracy indicators. Atlases improved the overall segmentation accuracy by 2.5-3.5%, depending on the deployed classifier method.

KW - Atlas-based segmentation

KW - Brain tissue segmentation

KW - Infant brain

KW - Magnetic resonance imaging

KW - n/a OA procedure

UR - http://www.scopus.com/inward/record.url?scp=85118457160&partnerID=8YFLogxK

U2 - 10.1109/AFRICON51333.2021.9570952

DO - 10.1109/AFRICON51333.2021.9570952

M3 - Conference contribution

AN - SCOPUS:85118457160

SN - 978-1-6654-4748-5

SN - 978-1-6654-1983-3 (USB)

T3 - IEEE AFRICON Conference

BT - Proceedings of 2021 IEEE AFRICON

PB - IEEE

CY - Piscataway, NJ

T2 - IEEE AFRICON, AFRICON 2021

Y2 - 13 September 2021 through 15 September 2021

ER -

The role of atlases and multi-atlases in brain tissue segmentation based on multispectral magnetic resonance image data

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this