Labeling colorectal NBI zoom-videoendoscope image sequences with MRF and SVM
In this paper, we propose a sequence labeling method by using SVM posterior probabilities with a Markov Random Field (MRF) model for colorectal Narrow Band Imaging (NBI) zoom-videoendoscope. Classifying each frame of a video sequence by SVM classifiers independently leads to an output sequence which...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 4834 |
|---|---|
| container_issue | |
| container_start_page | 4831 |
| container_title | |
| container_volume | 2013 |
| creator | Hirakawa, Tsubasa Tamaki, Toru Raytchev, Bisser Kaneda, Kazufumi Koide, Tetsushi Yoshida, Shigeta Kominami, Yoko Matsuo, Taiji Miyaki, Rie Tanaka, Shinji |
| description | In this paper, we propose a sequence labeling method by using SVM posterior probabilities with a Markov Random Field (MRF) model for colorectal Narrow Band Imaging (NBI) zoom-videoendoscope. Classifying each frame of a video sequence by SVM classifiers independently leads to an output sequence which is unstable and hard to understand by endoscopists. To make it more stable and readable, we use an MRF model to label the sequence of posterior probabilities. In addition, we introduce class asymmetry for the NBI images in order to keep and enhance frames where there is a possibility that cancers might have been detected. Experimental results with NBI video sequences demonstrate that the proposed MRF model with class asymmetry performs much better than a model without asymmetry. |
| doi_str_mv | 10.1109/EMBC.2013.6610629 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>pubmed_6IE</sourceid><recordid>TN_cdi_ieee_primary_6610629</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6610629</ieee_id><sourcerecordid>24110816</sourcerecordid><originalsourceid>FETCH-LOGICAL-i301t-313e9a5d14465f397cdc62ff17c696ce5ce565906da517c1b81599e7a44569833</originalsourceid><addsrcrecordid>eNo9UO1Kw0AQPEWxtfYBRJB7gdTb3N1e7qctrRZSBb_wX7leNjWS5mrSKvr0BlqFhR1mhmFnGTsHMQAQ9mo8G44GsQA5QASBsT1gp6C0MSIGtIesC1onkULQRy0WVkWYmNcO6zfNuxACDKKM5QnrxKrNSwC7LE3dgsqiWnIfylCT37iS3w2n_CeEVfRZZBSoykLjw5p4sXJL4g19bKny1PCvYvPGZw8T7qqMP77Mzthx7sqG-vvdY8-T8dPoNkrvb6aj6zQqpIBNJEGSdToDpVDn0hqfeYzzHIxHi550O6itwMzploNFAtpaMk4pjTaRsscud7nr7WJF2Xxdt5fV3_O_Wq3hYmcoiOhf3j9N_gKl5VnQ</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Labeling colorectal NBI zoom-videoendoscope image sequences with MRF and SVM</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Hirakawa, Tsubasa ; Tamaki, Toru ; Raytchev, Bisser ; Kaneda, Kazufumi ; Koide, Tetsushi ; Yoshida, Shigeta ; Kominami, Yoko ; Matsuo, Taiji ; Miyaki, Rie ; Tanaka, Shinji</creator><creatorcontrib>Hirakawa, Tsubasa ; Tamaki, Toru ; Raytchev, Bisser ; Kaneda, Kazufumi ; Koide, Tetsushi ; Yoshida, Shigeta ; Kominami, Yoko ; Matsuo, Taiji ; Miyaki, Rie ; Tanaka, Shinji</creatorcontrib><description>In this paper, we propose a sequence labeling method by using SVM posterior probabilities with a Markov Random Field (MRF) model for colorectal Narrow Band Imaging (NBI) zoom-videoendoscope. Classifying each frame of a video sequence by SVM classifiers independently leads to an output sequence which is unstable and hard to understand by endoscopists. To make it more stable and readable, we use an MRF model to label the sequence of posterior probabilities. In addition, we introduce class asymmetry for the NBI images in order to keep and enhance frames where there is a possibility that cancers might have been detected. Experimental results with NBI video sequences demonstrate that the proposed MRF model with class asymmetry performs much better than a model without asymmetry.</description><identifier>ISSN: 1094-687X</identifier><identifier>ISSN: 1557-170X</identifier><identifier>EISSN: 1558-4615</identifier><identifier>EISBN: 1457702169</identifier><identifier>EISBN: 9781457702167</identifier><identifier>DOI: 10.1109/EMBC.2013.6610629</identifier><identifier>PMID: 24110816</identifier><language>eng</language><publisher>United States: IEEE</publisher><subject>Cancer ; Capsule Endoscopy ; Colorectal Neoplasms - diagnosis ; Endoscopes ; Humans ; Image Processing, Computer-Assisted ; Image segmentation ; Labeling ; Markov Chains ; Narrow Band Imaging - methods ; Support Vector Machine ; Support vector machines ; Tumors ; Video sequences</subject><ispartof>2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2013, Vol.2013, p.4831-4834</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6610629$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,2052,27902,54895</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6610629$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/24110816$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Hirakawa, Tsubasa</creatorcontrib><creatorcontrib>Tamaki, Toru</creatorcontrib><creatorcontrib>Raytchev, Bisser</creatorcontrib><creatorcontrib>Kaneda, Kazufumi</creatorcontrib><creatorcontrib>Koide, Tetsushi</creatorcontrib><creatorcontrib>Yoshida, Shigeta</creatorcontrib><creatorcontrib>Kominami, Yoko</creatorcontrib><creatorcontrib>Matsuo, Taiji</creatorcontrib><creatorcontrib>Miyaki, Rie</creatorcontrib><creatorcontrib>Tanaka, Shinji</creatorcontrib><title>Labeling colorectal NBI zoom-videoendoscope image sequences with MRF and SVM</title><title>2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)</title><addtitle>EMBC</addtitle><addtitle>Conf Proc IEEE Eng Med Biol Soc</addtitle><description>In this paper, we propose a sequence labeling method by using SVM posterior probabilities with a Markov Random Field (MRF) model for colorectal Narrow Band Imaging (NBI) zoom-videoendoscope. Classifying each frame of a video sequence by SVM classifiers independently leads to an output sequence which is unstable and hard to understand by endoscopists. To make it more stable and readable, we use an MRF model to label the sequence of posterior probabilities. In addition, we introduce class asymmetry for the NBI images in order to keep and enhance frames where there is a possibility that cancers might have been detected. Experimental results with NBI video sequences demonstrate that the proposed MRF model with class asymmetry performs much better than a model without asymmetry.</description><subject>Cancer</subject><subject>Capsule Endoscopy</subject><subject>Colorectal Neoplasms - diagnosis</subject><subject>Endoscopes</subject><subject>Humans</subject><subject>Image Processing, Computer-Assisted</subject><subject>Image segmentation</subject><subject>Labeling</subject><subject>Markov Chains</subject><subject>Narrow Band Imaging - methods</subject><subject>Support Vector Machine</subject><subject>Support vector machines</subject><subject>Tumors</subject><subject>Video sequences</subject><issn>1094-687X</issn><issn>1557-170X</issn><issn>1558-4615</issn><isbn>1457702169</isbn><isbn>9781457702167</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2013</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><sourceid>EIF</sourceid><recordid>eNo9UO1Kw0AQPEWxtfYBRJB7gdTb3N1e7qctrRZSBb_wX7leNjWS5mrSKvr0BlqFhR1mhmFnGTsHMQAQ9mo8G44GsQA5QASBsT1gp6C0MSIGtIesC1onkULQRy0WVkWYmNcO6zfNuxACDKKM5QnrxKrNSwC7LE3dgsqiWnIfylCT37iS3w2n_CeEVfRZZBSoykLjw5p4sXJL4g19bKny1PCvYvPGZw8T7qqMP77Mzthx7sqG-vvdY8-T8dPoNkrvb6aj6zQqpIBNJEGSdToDpVDn0hqfeYzzHIxHi550O6itwMzploNFAtpaMk4pjTaRsscud7nr7WJF2Xxdt5fV3_O_Wq3hYmcoiOhf3j9N_gKl5VnQ</recordid><startdate>20130101</startdate><enddate>20130101</enddate><creator>Hirakawa, Tsubasa</creator><creator>Tamaki, Toru</creator><creator>Raytchev, Bisser</creator><creator>Kaneda, Kazufumi</creator><creator>Koide, Tetsushi</creator><creator>Yoshida, Shigeta</creator><creator>Kominami, Yoko</creator><creator>Matsuo, Taiji</creator><creator>Miyaki, Rie</creator><creator>Tanaka, Shinji</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope></search><sort><creationdate>20130101</creationdate><title>Labeling colorectal NBI zoom-videoendoscope image sequences with MRF and SVM</title><author>Hirakawa, Tsubasa ; Tamaki, Toru ; Raytchev, Bisser ; Kaneda, Kazufumi ; Koide, Tetsushi ; Yoshida, Shigeta ; Kominami, Yoko ; Matsuo, Taiji ; Miyaki, Rie ; Tanaka, Shinji</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i301t-313e9a5d14465f397cdc62ff17c696ce5ce565906da517c1b81599e7a44569833</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Cancer</topic><topic>Capsule Endoscopy</topic><topic>Colorectal Neoplasms - diagnosis</topic><topic>Endoscopes</topic><topic>Humans</topic><topic>Image Processing, Computer-Assisted</topic><topic>Image segmentation</topic><topic>Labeling</topic><topic>Markov Chains</topic><topic>Narrow Band Imaging - methods</topic><topic>Support Vector Machine</topic><topic>Support vector machines</topic><topic>Tumors</topic><topic>Video sequences</topic><toplevel>online_resources</toplevel><creatorcontrib>Hirakawa, Tsubasa</creatorcontrib><creatorcontrib>Tamaki, Toru</creatorcontrib><creatorcontrib>Raytchev, Bisser</creatorcontrib><creatorcontrib>Kaneda, Kazufumi</creatorcontrib><creatorcontrib>Koide, Tetsushi</creatorcontrib><creatorcontrib>Yoshida, Shigeta</creatorcontrib><creatorcontrib>Kominami, Yoko</creatorcontrib><creatorcontrib>Matsuo, Taiji</creatorcontrib><creatorcontrib>Miyaki, Rie</creatorcontrib><creatorcontrib>Tanaka, Shinji</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Hirakawa, Tsubasa</au><au>Tamaki, Toru</au><au>Raytchev, Bisser</au><au>Kaneda, Kazufumi</au><au>Koide, Tetsushi</au><au>Yoshida, Shigeta</au><au>Kominami, Yoko</au><au>Matsuo, Taiji</au><au>Miyaki, Rie</au><au>Tanaka, Shinji</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Labeling colorectal NBI zoom-videoendoscope image sequences with MRF and SVM</atitle><btitle>2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)</btitle><stitle>EMBC</stitle><addtitle>Conf Proc IEEE Eng Med Biol Soc</addtitle><date>2013-01-01</date><risdate>2013</risdate><volume>2013</volume><spage>4831</spage><epage>4834</epage><pages>4831-4834</pages><issn>1094-687X</issn><issn>1557-170X</issn><eissn>1558-4615</eissn><eisbn>1457702169</eisbn><eisbn>9781457702167</eisbn><abstract>In this paper, we propose a sequence labeling method by using SVM posterior probabilities with a Markov Random Field (MRF) model for colorectal Narrow Band Imaging (NBI) zoom-videoendoscope. Classifying each frame of a video sequence by SVM classifiers independently leads to an output sequence which is unstable and hard to understand by endoscopists. To make it more stable and readable, we use an MRF model to label the sequence of posterior probabilities. In addition, we introduce class asymmetry for the NBI images in order to keep and enhance frames where there is a possibility that cancers might have been detected. Experimental results with NBI video sequences demonstrate that the proposed MRF model with class asymmetry performs much better than a model without asymmetry.</abstract><cop>United States</cop><pub>IEEE</pub><pmid>24110816</pmid><doi>10.1109/EMBC.2013.6610629</doi><tpages>4</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1094-687X |
| ispartof | 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2013, Vol.2013, p.4831-4834 |
| issn | 1094-687X 1557-170X 1558-4615 |
| language | eng |
| recordid | cdi_ieee_primary_6610629 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Cancer Capsule Endoscopy Colorectal Neoplasms - diagnosis Endoscopes Humans Image Processing, Computer-Assisted Image segmentation Labeling Markov Chains Narrow Band Imaging - methods Support Vector Machine Support vector machines Tumors Video sequences |
| title | Labeling colorectal NBI zoom-videoendoscope image sequences with MRF and SVM |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T01%3A09%3A59IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-pubmed_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Labeling%20colorectal%20NBI%20zoom-videoendoscope%20image%20sequences%20with%20MRF%20and%20SVM&rft.btitle=2013%2035th%20Annual%20International%20Conference%20of%20the%20IEEE%20Engineering%20in%20Medicine%20and%20Biology%20Society%20(EMBC)&rft.au=Hirakawa,%20Tsubasa&rft.date=2013-01-01&rft.volume=2013&rft.spage=4831&rft.epage=4834&rft.pages=4831-4834&rft.issn=1094-687X&rft.eissn=1558-4615&rft_id=info:doi/10.1109/EMBC.2013.6610629&rft_dat=%3Cpubmed_6IE%3E24110816%3C/pubmed_6IE%3E%3Curl%3E%3C/url%3E&rft.eisbn=1457702169&rft.eisbn_list=9781457702167&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/24110816&rft_ieee_id=6610629&rfr_iscdi=true |