Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network

Abnormal iron accumulation in the brain subcortical nuclei has been reported to be correlated to various neurodegenerative diseases, which can be measured through the magnetic susceptibility from the quantitative susceptibility mapping (QSM). To quantitively measure the magnetic susceptibility, the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chai, Chao, Qiao, Pengchong, Zhao, Bin, Wang, Huiying, Liu, Guohua, Wu, Hong, Haacke, E Mark, Shen, Wen, Cao, Chen, Ye, Xinchen, Liu, Zhiyang, Xia, Shuang
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Chai, Chao
Qiao, Pengchong
Zhao, Bin
Wang, Huiying
Liu, Guohua
Wu, Hong
Haacke, E Mark
Shen, Wen
Cao, Chen
Ye, Xinchen
Liu, Zhiyang
Xia, Shuang
description Abnormal iron accumulation in the brain subcortical nuclei has been reported to be correlated to various neurodegenerative diseases, which can be measured through the magnetic susceptibility from the quantitative susceptibility mapping (QSM). To quantitively measure the magnetic susceptibility, the nuclei should be accurately segmented, which is a tedious task for clinicians. In this paper, we proposed a double-branch residual-structured U-Net (DB-ResUNet) based on 3D convolutional neural network (CNN) to automatically segment such brain gray matter nuclei. To better tradeoff between segmentation accuracy and the memory efficiency, the proposed DB-ResUNet fed image patches with high resolution and the patches with low resolution but larger field of view into the local and global branches, respectively. Experimental results revealed that by jointly using QSM and T$_\text{1}$ weighted imaging (T$_\text{1}$WI) as inputs, the proposed method was able to achieve better segmentation accuracy over its single-branch counterpart, as well as the conventional atlas-based method and the classical 3D-UNet structure. The susceptibility values and the volumes were also measured, which indicated that the measurements from the proposed DB-ResUNet are able to present high correlation with values from the manually annotated regions of interest.
doi_str_mv 10.48550/arxiv.2008.00901
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2008_00901</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2008_00901</sourcerecordid><originalsourceid>FETCH-LOGICAL-a671-c2dca50289a7eea684bceefec84a86915761d789eedd0abe013dbf618d9144303</originalsourceid><addsrcrecordid>eNotkL1OwzAUhb0woMIDMOEXSLDz64wlQEEqRahljm7sm8oiiSPHDnTkzWkCyznD-XSGj5AbzsJEpCm7A_utpzBiTISMFYxfkp-1d6YDh4ru8dhh78Bp01PT0HsLuqcbCyf6Cs6hpTsvW9T0PL976J2e2Qnp3o8SB6dr3Wo3w8Og-yP9GOd8QBxoafrJtH5-hpbu0Nul3Jexn1fkooF2xOv_XpHD0-OhfA62b5uXcr0NIMt5ICMlIWWRKCBHhEwktURsUIoERFbwNM-4ykWBqBSDGhmPVd1kXKiCJ0nM4hW5_btdHFSD1R3YUzW7qBYX8S9vrl0S</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network</title><source>arXiv.org</source><creator>Chai, Chao ; Qiao, Pengchong ; Zhao, Bin ; Wang, Huiying ; Liu, Guohua ; Wu, Hong ; Haacke, E Mark ; Shen, Wen ; Cao, Chen ; Ye, Xinchen ; Liu, Zhiyang ; Xia, Shuang</creator><creatorcontrib>Chai, Chao ; Qiao, Pengchong ; Zhao, Bin ; Wang, Huiying ; Liu, Guohua ; Wu, Hong ; Haacke, E Mark ; Shen, Wen ; Cao, Chen ; Ye, Xinchen ; Liu, Zhiyang ; Xia, Shuang</creatorcontrib><description>Abnormal iron accumulation in the brain subcortical nuclei has been reported to be correlated to various neurodegenerative diseases, which can be measured through the magnetic susceptibility from the quantitative susceptibility mapping (QSM). To quantitively measure the magnetic susceptibility, the nuclei should be accurately segmented, which is a tedious task for clinicians. In this paper, we proposed a double-branch residual-structured U-Net (DB-ResUNet) based on 3D convolutional neural network (CNN) to automatically segment such brain gray matter nuclei. To better tradeoff between segmentation accuracy and the memory efficiency, the proposed DB-ResUNet fed image patches with high resolution and the patches with low resolution but larger field of view into the local and global branches, respectively. Experimental results revealed that by jointly using QSM and T$_\text{1}$ weighted imaging (T$_\text{1}$WI) as inputs, the proposed method was able to achieve better segmentation accuracy over its single-branch counterpart, as well as the conventional atlas-based method and the classical 3D-UNet structure. The susceptibility values and the volumes were also measured, which indicated that the measurements from the proposed DB-ResUNet are able to present high correlation with values from the manually annotated regions of interest.</description><identifier>DOI: 10.48550/arxiv.2008.00901</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2020-08</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,778,883</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2008.00901$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2008.00901$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Chai, Chao</creatorcontrib><creatorcontrib>Qiao, Pengchong</creatorcontrib><creatorcontrib>Zhao, Bin</creatorcontrib><creatorcontrib>Wang, Huiying</creatorcontrib><creatorcontrib>Liu, Guohua</creatorcontrib><creatorcontrib>Wu, Hong</creatorcontrib><creatorcontrib>Haacke, E Mark</creatorcontrib><creatorcontrib>Shen, Wen</creatorcontrib><creatorcontrib>Cao, Chen</creatorcontrib><creatorcontrib>Ye, Xinchen</creatorcontrib><creatorcontrib>Liu, Zhiyang</creatorcontrib><creatorcontrib>Xia, Shuang</creatorcontrib><title>Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network</title><description>Abnormal iron accumulation in the brain subcortical nuclei has been reported to be correlated to various neurodegenerative diseases, which can be measured through the magnetic susceptibility from the quantitative susceptibility mapping (QSM). To quantitively measure the magnetic susceptibility, the nuclei should be accurately segmented, which is a tedious task for clinicians. In this paper, we proposed a double-branch residual-structured U-Net (DB-ResUNet) based on 3D convolutional neural network (CNN) to automatically segment such brain gray matter nuclei. To better tradeoff between segmentation accuracy and the memory efficiency, the proposed DB-ResUNet fed image patches with high resolution and the patches with low resolution but larger field of view into the local and global branches, respectively. Experimental results revealed that by jointly using QSM and T$_\text{1}$ weighted imaging (T$_\text{1}$WI) as inputs, the proposed method was able to achieve better segmentation accuracy over its single-branch counterpart, as well as the conventional atlas-based method and the classical 3D-UNet structure. The susceptibility values and the volumes were also measured, which indicated that the measurements from the proposed DB-ResUNet are able to present high correlation with values from the manually annotated regions of interest.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotkL1OwzAUhb0woMIDMOEXSLDz64wlQEEqRahljm7sm8oiiSPHDnTkzWkCyznD-XSGj5AbzsJEpCm7A_utpzBiTISMFYxfkp-1d6YDh4ru8dhh78Bp01PT0HsLuqcbCyf6Cs6hpTsvW9T0PL976J2e2Qnp3o8SB6dr3Wo3w8Og-yP9GOd8QBxoafrJtH5-hpbu0Nul3Jexn1fkooF2xOv_XpHD0-OhfA62b5uXcr0NIMt5ICMlIWWRKCBHhEwktURsUIoERFbwNM-4ykWBqBSDGhmPVd1kXKiCJ0nM4hW5_btdHFSD1R3YUzW7qBYX8S9vrl0S</recordid><startdate>20200803</startdate><enddate>20200803</enddate><creator>Chai, Chao</creator><creator>Qiao, Pengchong</creator><creator>Zhao, Bin</creator><creator>Wang, Huiying</creator><creator>Liu, Guohua</creator><creator>Wu, Hong</creator><creator>Haacke, E Mark</creator><creator>Shen, Wen</creator><creator>Cao, Chen</creator><creator>Ye, Xinchen</creator><creator>Liu, Zhiyang</creator><creator>Xia, Shuang</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20200803</creationdate><title>Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network</title><author>Chai, Chao ; Qiao, Pengchong ; Zhao, Bin ; Wang, Huiying ; Liu, Guohua ; Wu, Hong ; Haacke, E Mark ; Shen, Wen ; Cao, Chen ; Ye, Xinchen ; Liu, Zhiyang ; Xia, Shuang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a671-c2dca50289a7eea684bceefec84a86915761d789eedd0abe013dbf618d9144303</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Chai, Chao</creatorcontrib><creatorcontrib>Qiao, Pengchong</creatorcontrib><creatorcontrib>Zhao, Bin</creatorcontrib><creatorcontrib>Wang, Huiying</creatorcontrib><creatorcontrib>Liu, Guohua</creatorcontrib><creatorcontrib>Wu, Hong</creatorcontrib><creatorcontrib>Haacke, E Mark</creatorcontrib><creatorcontrib>Shen, Wen</creatorcontrib><creatorcontrib>Cao, Chen</creatorcontrib><creatorcontrib>Ye, Xinchen</creatorcontrib><creatorcontrib>Liu, Zhiyang</creatorcontrib><creatorcontrib>Xia, Shuang</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Chai, Chao</au><au>Qiao, Pengchong</au><au>Zhao, Bin</au><au>Wang, Huiying</au><au>Liu, Guohua</au><au>Wu, Hong</au><au>Haacke, E Mark</au><au>Shen, Wen</au><au>Cao, Chen</au><au>Ye, Xinchen</au><au>Liu, Zhiyang</au><au>Xia, Shuang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network</atitle><date>2020-08-03</date><risdate>2020</risdate><abstract>Abnormal iron accumulation in the brain subcortical nuclei has been reported to be correlated to various neurodegenerative diseases, which can be measured through the magnetic susceptibility from the quantitative susceptibility mapping (QSM). To quantitively measure the magnetic susceptibility, the nuclei should be accurately segmented, which is a tedious task for clinicians. In this paper, we proposed a double-branch residual-structured U-Net (DB-ResUNet) based on 3D convolutional neural network (CNN) to automatically segment such brain gray matter nuclei. To better tradeoff between segmentation accuracy and the memory efficiency, the proposed DB-ResUNet fed image patches with high resolution and the patches with low resolution but larger field of view into the local and global branches, respectively. Experimental results revealed that by jointly using QSM and T$_\text{1}$ weighted imaging (T$_\text{1}$WI) as inputs, the proposed method was able to achieve better segmentation accuracy over its single-branch counterpart, as well as the conventional atlas-based method and the classical 3D-UNet structure. The susceptibility values and the volumes were also measured, which indicated that the measurements from the proposed DB-ResUNet are able to present high correlation with values from the manually annotated regions of interest.</abstract><doi>10.48550/arxiv.2008.00901</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2008.00901
ispartof
issn
language eng
recordid cdi_arxiv_primary_2008_00901
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-16T09%3A39%3A14IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Automated%20Segmentation%20of%20Brain%20Gray%20Matter%20Nuclei%20on%20Quantitative%20Susceptibility%20Mapping%20Using%20Deep%20Convolutional%20Neural%20Network&rft.au=Chai,%20Chao&rft.date=2020-08-03&rft_id=info:doi/10.48550/arxiv.2008.00901&rft_dat=%3Carxiv_GOX%3E2008_00901%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true