Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network
Brain tumors can appear anywhere in the brain and have vastly different sizes and morphology. Additionally, these tumors are often diffused and poorly contrasted. Consequently, the segmentation of brain tumor and intratumor subregions using magnetic resonance imaging (MRI) data with minimal human in...
Gespeichert in:
Veröffentlicht in: | Journal of healthcare engineering 2018-01, Vol.2018 (2018), p.1-14 |
---|---|
Hauptverfasser: | , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 14 |
---|---|
container_issue | 2018 |
container_start_page | 1 |
container_title | Journal of healthcare engineering |
container_volume | 2018 |
creator | Liu, Chang Jiang, Jingfeng Mao, Lei Cui, Shaoguo Xiong, Shuyu |
description | Brain tumors can appear anywhere in the brain and have vastly different sizes and morphology. Additionally, these tumors are often diffused and poorly contrasted. Consequently, the segmentation of brain tumor and intratumor subregions using magnetic resonance imaging (MRI) data with minimal human interventions remains a challenging task. In this paper, we present a novel fully automatic segmentation method from MRI data containing in vivo brain gliomas. This approach can not only localize the entire tumor region but can also accurately segment the intratumor structure. The proposed work was based on a cascaded deep learning convolutional neural network consisting of two subnetworks: (1) a tumor localization network (TLN) and (2) an intratumor classification network (ITCN). The TLN, a fully convolutional network (FCN) in conjunction with the transfer learning technology, was used to first process MRI data. The goal of the first subnetwork was to define the tumor region from an MRI slice. Then, the ITCN was used to label the defined tumor region into multiple subregions. Particularly, ITCN exploited a convolutional neural network (CNN) with deeper architecture and smaller kernel. The proposed approach was validated on multimodal brain tumor segmentation (BRATS 2015) datasets, which contain 220 high-grade glioma (HGG) and 54 low-grade glioma (LGG) cases. Dice similarity coefficient (DSC), positive predictive value (PPV), and sensitivity were used as evaluation metrics. Our experimental results indicated that our method could obtain the promising segmentation results and had a faster segmentation speed. More specifically, the proposed method obtained comparable and overall better DSC values (0.89, 0.77, and 0.80) on the combined (HGG + LGG) testing set, as compared to other methods reported in the literature. Additionally, the proposed approach was able to complete a segmentation task at a rate of 1.54 seconds per slice. |
doi_str_mv | 10.1155/2018/4940593 |
format | Article |
fullrecord | <record><control><sourceid>gale_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_5884212</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><galeid>A587654896</galeid><sourcerecordid>A587654896</sourcerecordid><originalsourceid>FETCH-LOGICAL-c471t-6e763a22d46c8518807654d788c6701ea9762b0a2d1ed77561e90ab25d930fa23</originalsourceid><addsrcrecordid>eNqNkc1v1DAQxSMEolXpjTOyxAUJlvoj_rogLUspKxWQgJ6t2WSSGhJ7iROq_vc4yrbAjbnMyPPT8xu9onjK6GvGpDzjlJmz0pZUWvGgOOa0pCsuqH14N3Mrj4rTlL7TXMKKkonHxRG3WkrN1HGB62mMPYy-Il-xh7AMbY9hzI8xkNiQtwP4QC46n8FEmiH25OOXLdn20GIiV8mHlgB5h7gnG0gV1FiTTzgN0OU23sThx5PiUQNdwtNDPymu3p9_23xYXX6-2G7Wl6uq1GxcKdRKAOd1qSojmTFUK1nW2phKacoQrFZ8R4HXDGutpWJoKey4rK2gDXBxUrxZdPfTrse6yldkF24_-B6GWxfBu383wV-7Nv5y0piSs1ngxUFgiD8nTKPrfaqw6yBgnJLjVBhNJVc2o88XtIUOnQ9NzIrVjLu1NLNxY1WmXi1UNcSUBmzuzTDq5gjdHKE7RJjxZ38fcA_fBZaBlwtw7UMNN_4_5XK0-W_4QzOjBbfiNwjfq3A</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2038705269</pqid></control><display><type>article</type><title>Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network</title><source>PubMed Central Open Access</source><source>EZB-FREE-00999 freely available EZB journals</source><source>Wiley Online Library (Open Access Collection)</source><source>PubMed Central</source><source>Alma/SFX Local Collection</source><creator>Liu, Chang ; Jiang, Jingfeng ; Mao, Lei ; Cui, Shaoguo ; Xiong, Shuyu</creator><contributor>Chang, Weide</contributor><creatorcontrib>Liu, Chang ; Jiang, Jingfeng ; Mao, Lei ; Cui, Shaoguo ; Xiong, Shuyu ; Chang, Weide</creatorcontrib><description>Brain tumors can appear anywhere in the brain and have vastly different sizes and morphology. Additionally, these tumors are often diffused and poorly contrasted. Consequently, the segmentation of brain tumor and intratumor subregions using magnetic resonance imaging (MRI) data with minimal human interventions remains a challenging task. In this paper, we present a novel fully automatic segmentation method from MRI data containing in vivo brain gliomas. This approach can not only localize the entire tumor region but can also accurately segment the intratumor structure. The proposed work was based on a cascaded deep learning convolutional neural network consisting of two subnetworks: (1) a tumor localization network (TLN) and (2) an intratumor classification network (ITCN). The TLN, a fully convolutional network (FCN) in conjunction with the transfer learning technology, was used to first process MRI data. The goal of the first subnetwork was to define the tumor region from an MRI slice. Then, the ITCN was used to label the defined tumor region into multiple subregions. Particularly, ITCN exploited a convolutional neural network (CNN) with deeper architecture and smaller kernel. The proposed approach was validated on multimodal brain tumor segmentation (BRATS 2015) datasets, which contain 220 high-grade glioma (HGG) and 54 low-grade glioma (LGG) cases. Dice similarity coefficient (DSC), positive predictive value (PPV), and sensitivity were used as evaluation metrics. Our experimental results indicated that our method could obtain the promising segmentation results and had a faster segmentation speed. More specifically, the proposed method obtained comparable and overall better DSC values (0.89, 0.77, and 0.80) on the combined (HGG + LGG) testing set, as compared to other methods reported in the literature. Additionally, the proposed approach was able to complete a segmentation task at a rate of 1.54 seconds per slice.</description><identifier>ISSN: 2040-2295</identifier><identifier>EISSN: 2040-2309</identifier><identifier>DOI: 10.1155/2018/4940593</identifier><identifier>PMID: 29755716</identifier><language>eng</language><publisher>Cairo, Egypt: Hindawi Publishing Corporation</publisher><subject>Brain tumors ; Gliomas ; Magnetic resonance imaging ; Medical imaging equipment ; Neural networks</subject><ispartof>Journal of healthcare engineering, 2018-01, Vol.2018 (2018), p.1-14</ispartof><rights>Copyright © 2018 Shaoguo Cui et al.</rights><rights>COPYRIGHT 2018 John Wiley & Sons, Inc.</rights><rights>Copyright © 2018 Shaoguo Cui et al. 2018</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c471t-6e763a22d46c8518807654d788c6701ea9762b0a2d1ed77561e90ab25d930fa23</citedby><cites>FETCH-LOGICAL-c471t-6e763a22d46c8518807654d788c6701ea9762b0a2d1ed77561e90ab25d930fa23</cites><orcidid>0000-0001-8812-6246 ; 0000-0003-3064-0490</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC5884212/pdf/$$EPDF$$P50$$Gpubmedcentral$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC5884212/$$EHTML$$P50$$Gpubmedcentral$$Hfree_for_read</linktohtml><link.rule.ids>230,314,727,780,784,885,27924,27925,53791,53793</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/29755716$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><contributor>Chang, Weide</contributor><creatorcontrib>Liu, Chang</creatorcontrib><creatorcontrib>Jiang, Jingfeng</creatorcontrib><creatorcontrib>Mao, Lei</creatorcontrib><creatorcontrib>Cui, Shaoguo</creatorcontrib><creatorcontrib>Xiong, Shuyu</creatorcontrib><title>Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network</title><title>Journal of healthcare engineering</title><addtitle>J Healthc Eng</addtitle><description>Brain tumors can appear anywhere in the brain and have vastly different sizes and morphology. Additionally, these tumors are often diffused and poorly contrasted. Consequently, the segmentation of brain tumor and intratumor subregions using magnetic resonance imaging (MRI) data with minimal human interventions remains a challenging task. In this paper, we present a novel fully automatic segmentation method from MRI data containing in vivo brain gliomas. This approach can not only localize the entire tumor region but can also accurately segment the intratumor structure. The proposed work was based on a cascaded deep learning convolutional neural network consisting of two subnetworks: (1) a tumor localization network (TLN) and (2) an intratumor classification network (ITCN). The TLN, a fully convolutional network (FCN) in conjunction with the transfer learning technology, was used to first process MRI data. The goal of the first subnetwork was to define the tumor region from an MRI slice. Then, the ITCN was used to label the defined tumor region into multiple subregions. Particularly, ITCN exploited a convolutional neural network (CNN) with deeper architecture and smaller kernel. The proposed approach was validated on multimodal brain tumor segmentation (BRATS 2015) datasets, which contain 220 high-grade glioma (HGG) and 54 low-grade glioma (LGG) cases. Dice similarity coefficient (DSC), positive predictive value (PPV), and sensitivity were used as evaluation metrics. Our experimental results indicated that our method could obtain the promising segmentation results and had a faster segmentation speed. More specifically, the proposed method obtained comparable and overall better DSC values (0.89, 0.77, and 0.80) on the combined (HGG + LGG) testing set, as compared to other methods reported in the literature. Additionally, the proposed approach was able to complete a segmentation task at a rate of 1.54 seconds per slice.</description><subject>Brain tumors</subject><subject>Gliomas</subject><subject>Magnetic resonance imaging</subject><subject>Medical imaging equipment</subject><subject>Neural networks</subject><issn>2040-2295</issn><issn>2040-2309</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>RHX</sourceid><recordid>eNqNkc1v1DAQxSMEolXpjTOyxAUJlvoj_rogLUspKxWQgJ6t2WSSGhJ7iROq_vc4yrbAjbnMyPPT8xu9onjK6GvGpDzjlJmz0pZUWvGgOOa0pCsuqH14N3Mrj4rTlL7TXMKKkonHxRG3WkrN1HGB62mMPYy-Il-xh7AMbY9hzI8xkNiQtwP4QC46n8FEmiH25OOXLdn20GIiV8mHlgB5h7gnG0gV1FiTTzgN0OU23sThx5PiUQNdwtNDPymu3p9_23xYXX6-2G7Wl6uq1GxcKdRKAOd1qSojmTFUK1nW2phKacoQrFZ8R4HXDGutpWJoKey4rK2gDXBxUrxZdPfTrse6yldkF24_-B6GWxfBu383wV-7Nv5y0piSs1ngxUFgiD8nTKPrfaqw6yBgnJLjVBhNJVc2o88XtIUOnQ9NzIrVjLu1NLNxY1WmXi1UNcSUBmzuzTDq5gjdHKE7RJjxZ38fcA_fBZaBlwtw7UMNN_4_5XK0-W_4QzOjBbfiNwjfq3A</recordid><startdate>20180101</startdate><enddate>20180101</enddate><creator>Liu, Chang</creator><creator>Jiang, Jingfeng</creator><creator>Mao, Lei</creator><creator>Cui, Shaoguo</creator><creator>Xiong, Shuyu</creator><general>Hindawi Publishing Corporation</general><general>Hindawi</general><general>John Wiley & Sons, Inc</general><scope>ADJCN</scope><scope>AHFXO</scope><scope>RHU</scope><scope>RHW</scope><scope>RHX</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0001-8812-6246</orcidid><orcidid>https://orcid.org/0000-0003-3064-0490</orcidid></search><sort><creationdate>20180101</creationdate><title>Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network</title><author>Liu, Chang ; Jiang, Jingfeng ; Mao, Lei ; Cui, Shaoguo ; Xiong, Shuyu</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c471t-6e763a22d46c8518807654d788c6701ea9762b0a2d1ed77561e90ab25d930fa23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Brain tumors</topic><topic>Gliomas</topic><topic>Magnetic resonance imaging</topic><topic>Medical imaging equipment</topic><topic>Neural networks</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Liu, Chang</creatorcontrib><creatorcontrib>Jiang, Jingfeng</creatorcontrib><creatorcontrib>Mao, Lei</creatorcontrib><creatorcontrib>Cui, Shaoguo</creatorcontrib><creatorcontrib>Xiong, Shuyu</creatorcontrib><collection>الدوريات العلمية والإحصائية - e-Marefa Academic and Statistical Periodicals</collection><collection>معرفة - المحتوى العربي الأكاديمي المتكامل - e-Marefa Academic Complete</collection><collection>Hindawi Publishing Complete</collection><collection>Hindawi Publishing Subscription Journals</collection><collection>Hindawi Publishing Open Access Journals</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of healthcare engineering</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Liu, Chang</au><au>Jiang, Jingfeng</au><au>Mao, Lei</au><au>Cui, Shaoguo</au><au>Xiong, Shuyu</au><au>Chang, Weide</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network</atitle><jtitle>Journal of healthcare engineering</jtitle><addtitle>J Healthc Eng</addtitle><date>2018-01-01</date><risdate>2018</risdate><volume>2018</volume><issue>2018</issue><spage>1</spage><epage>14</epage><pages>1-14</pages><issn>2040-2295</issn><eissn>2040-2309</eissn><abstract>Brain tumors can appear anywhere in the brain and have vastly different sizes and morphology. Additionally, these tumors are often diffused and poorly contrasted. Consequently, the segmentation of brain tumor and intratumor subregions using magnetic resonance imaging (MRI) data with minimal human interventions remains a challenging task. In this paper, we present a novel fully automatic segmentation method from MRI data containing in vivo brain gliomas. This approach can not only localize the entire tumor region but can also accurately segment the intratumor structure. The proposed work was based on a cascaded deep learning convolutional neural network consisting of two subnetworks: (1) a tumor localization network (TLN) and (2) an intratumor classification network (ITCN). The TLN, a fully convolutional network (FCN) in conjunction with the transfer learning technology, was used to first process MRI data. The goal of the first subnetwork was to define the tumor region from an MRI slice. Then, the ITCN was used to label the defined tumor region into multiple subregions. Particularly, ITCN exploited a convolutional neural network (CNN) with deeper architecture and smaller kernel. The proposed approach was validated on multimodal brain tumor segmentation (BRATS 2015) datasets, which contain 220 high-grade glioma (HGG) and 54 low-grade glioma (LGG) cases. Dice similarity coefficient (DSC), positive predictive value (PPV), and sensitivity were used as evaluation metrics. Our experimental results indicated that our method could obtain the promising segmentation results and had a faster segmentation speed. More specifically, the proposed method obtained comparable and overall better DSC values (0.89, 0.77, and 0.80) on the combined (HGG + LGG) testing set, as compared to other methods reported in the literature. Additionally, the proposed approach was able to complete a segmentation task at a rate of 1.54 seconds per slice.</abstract><cop>Cairo, Egypt</cop><pub>Hindawi Publishing Corporation</pub><pmid>29755716</pmid><doi>10.1155/2018/4940593</doi><tpages>14</tpages><orcidid>https://orcid.org/0000-0001-8812-6246</orcidid><orcidid>https://orcid.org/0000-0003-3064-0490</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2040-2295 |
ispartof | Journal of healthcare engineering, 2018-01, Vol.2018 (2018), p.1-14 |
issn | 2040-2295 2040-2309 |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_5884212 |
source | PubMed Central Open Access; EZB-FREE-00999 freely available EZB journals; Wiley Online Library (Open Access Collection); PubMed Central; Alma/SFX Local Collection |
subjects | Brain tumors Gliomas Magnetic resonance imaging Medical imaging equipment Neural networks |
title | Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T08%3A28%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-gale_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Automatic%20Semantic%20Segmentation%20of%20Brain%20Gliomas%20from%20MRI%20Images%20Using%20a%20Deep%20Cascaded%20Neural%20Network&rft.jtitle=Journal%20of%20healthcare%20engineering&rft.au=Liu,%20Chang&rft.date=2018-01-01&rft.volume=2018&rft.issue=2018&rft.spage=1&rft.epage=14&rft.pages=1-14&rft.issn=2040-2295&rft.eissn=2040-2309&rft_id=info:doi/10.1155/2018/4940593&rft_dat=%3Cgale_pubme%3EA587654896%3C/gale_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2038705269&rft_id=info:pmid/29755716&rft_galeid=A587654896&rfr_iscdi=true |