Computerized lesion detection on breast ultrasound
We investigated the use of a radial gradient index (RGI) filtering technique to automatically detect lesions on breast ultrasound. After initial RGI filtering, a sensitivity of 87% at 0.76 false-positive detections per image was obtained on a database of 400 patients (757 images). Next, lesion candi...
Gespeichert in:
| Veröffentlicht in: | Medical physics (Lancaster) 2002-07, Vol.29 (7), p.1438-1446 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 1446 |
|---|---|
| container_issue | 7 |
| container_start_page | 1438 |
| container_title | Medical physics (Lancaster) |
| container_volume | 29 |
| creator | Drukker, Karen Giger, Maryellen L. Horsch, Karla Kupinski, Matthew A. Vyborny, Carl J. Mendelson, Ellen B. |
| description | We investigated the use of a radial gradient index (RGI) filtering technique to automatically detect lesions on breast ultrasound. After initial RGI filtering, a sensitivity of 87% at 0.76 false-positive detections per image was obtained on a database of 400 patients (757 images). Next, lesion candidates were segmented from the background by maximizing an average radial gradient (ARD) index for regions grown from the detected points. At an overlap of 0.4 with a radiologist lesion outline, 75% of the lesions were correctly detected. Subsequently, round robin analysis was used to assess the quality of the classification of lesion candidates into actual lesions and false-positives by a Bayesian neural network. The round robin analysis yielded an
A
z
value of 0.84, and an overall performance by case of 94% sensitivity at 0.48 false-positives per image. Use of computerized analysis of breast sonograms may ultimately facilitate the use of sonography in breast cancer screening programs. |
| doi_str_mv | 10.1118/1.1485995 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_scita</sourceid><recordid>TN_cdi_scitation_primary_10_1118_1_1485995</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>71956853</sourcerecordid><originalsourceid>FETCH-LOGICAL-c4185-9cf2fad1550e1bfdd445b8b6c06a5609d4abb0834df0d239c3b301794f41413a3</originalsourceid><addsrcrecordid>eNp90MtKxDAUBuAgijOOLnwBmZWg0DEnl7ZZyuANRnSh65ArVNppTVplfHo7tKggCoGTxXd-Dj9Cx4AXAJBfwAJYzoXgO2hKWEYTRrDYRVOMBUsIw3yCDmJ8wRinlON9NAHSL2SETRFZ1lXTtS4UH87OSxeLej23rnWm3f76p4NTsZ13ZRtUrLu1PUR7XpXRHY1zhp6vr56Wt8nq4eZueblKDIOcJ8J44pUFzrED7a1ljOtcpwaniqdYWKa0xjll1mNLqDBUUwyZYJ4BA6roDJ0OuU2oXzsXW1kV0biyVGtXd1FmIHiac9rDswGaUMcYnJdNKCoVNhKw3BYkQY4F9fZkDO105ey3HBvpQTKA96J0m7-T5P3jGHg--GiKVm1L-9p5q8MP31j_H_596iej8YlD</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>71956853</pqid></control><display><type>article</type><title>Computerized lesion detection on breast ultrasound</title><source>MEDLINE</source><source>Wiley Online Library Journals Frontfile Complete</source><creator>Drukker, Karen ; Giger, Maryellen L. ; Horsch, Karla ; Kupinski, Matthew A. ; Vyborny, Carl J. ; Mendelson, Ellen B.</creator><creatorcontrib>Drukker, Karen ; Giger, Maryellen L. ; Horsch, Karla ; Kupinski, Matthew A. ; Vyborny, Carl J. ; Mendelson, Ellen B.</creatorcontrib><description>We investigated the use of a radial gradient index (RGI) filtering technique to automatically detect lesions on breast ultrasound. After initial RGI filtering, a sensitivity of 87% at 0.76 false-positive detections per image was obtained on a database of 400 patients (757 images). Next, lesion candidates were segmented from the background by maximizing an average radial gradient (ARD) index for regions grown from the detected points. At an overlap of 0.4 with a radiologist lesion outline, 75% of the lesions were correctly detected. Subsequently, round robin analysis was used to assess the quality of the classification of lesion candidates into actual lesions and false-positives by a Bayesian neural network. The round robin analysis yielded an
A
z
value of 0.84, and an overall performance by case of 94% sensitivity at 0.48 false-positives per image. Use of computerized analysis of breast sonograms may ultimately facilitate the use of sonography in breast cancer screening programs.</description><identifier>ISSN: 0094-2405</identifier><identifier>EISSN: 2473-4209</identifier><identifier>DOI: 10.1118/1.1485995</identifier><identifier>PMID: 12148724</identifier><identifier>CODEN: MPHYA6</identifier><language>eng</language><publisher>United States: American Association of Physicists in Medicine</publisher><subject>Artificial neural networks ; Bayes methods ; Bayes Theorem ; biological tissues ; biomedical ultrasonics ; Breast Neoplasms - diagnosis ; Breast Neoplasms - diagnostic imaging ; breast sonography ; cancer ; computer‐aided diagnosis ; Databases as Topic ; Diseases ; False Positive Reactions ; Female ; Humans ; Image analysis ; image classification ; Image Processing, Computer-Assisted ; image segmentation ; Mammography ; Mass Screening ; Medical diagnosis with acoustics ; medical image processing ; Medical imaging ; Models, Statistical ; neural nets ; Neural networks, fuzzy logic, artificial intelligence ; performance of lesion detection ; Physicists ; Probability theory ; Probability theory, stochastic processes, and statistics ; Radiologists ; ROC Curve ; Sensitivity and Specificity ; Software ; Ultrasonography ; Ultrasonography - methods</subject><ispartof>Medical physics (Lancaster), 2002-07, Vol.29 (7), p.1438-1446</ispartof><rights>American Association of Physicists in Medicine</rights><rights>2002 American Association of Physicists in Medicine</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c4185-9cf2fad1550e1bfdd445b8b6c06a5609d4abb0834df0d239c3b301794f41413a3</citedby><cites>FETCH-LOGICAL-c4185-9cf2fad1550e1bfdd445b8b6c06a5609d4abb0834df0d239c3b301794f41413a3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1118%2F1.1485995$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1118%2F1.1485995$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>314,776,780,1411,27903,27904,45553,45554</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/12148724$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Drukker, Karen</creatorcontrib><creatorcontrib>Giger, Maryellen L.</creatorcontrib><creatorcontrib>Horsch, Karla</creatorcontrib><creatorcontrib>Kupinski, Matthew A.</creatorcontrib><creatorcontrib>Vyborny, Carl J.</creatorcontrib><creatorcontrib>Mendelson, Ellen B.</creatorcontrib><title>Computerized lesion detection on breast ultrasound</title><title>Medical physics (Lancaster)</title><addtitle>Med Phys</addtitle><description>We investigated the use of a radial gradient index (RGI) filtering technique to automatically detect lesions on breast ultrasound. After initial RGI filtering, a sensitivity of 87% at 0.76 false-positive detections per image was obtained on a database of 400 patients (757 images). Next, lesion candidates were segmented from the background by maximizing an average radial gradient (ARD) index for regions grown from the detected points. At an overlap of 0.4 with a radiologist lesion outline, 75% of the lesions were correctly detected. Subsequently, round robin analysis was used to assess the quality of the classification of lesion candidates into actual lesions and false-positives by a Bayesian neural network. The round robin analysis yielded an
A
z
value of 0.84, and an overall performance by case of 94% sensitivity at 0.48 false-positives per image. Use of computerized analysis of breast sonograms may ultimately facilitate the use of sonography in breast cancer screening programs.</description><subject>Artificial neural networks</subject><subject>Bayes methods</subject><subject>Bayes Theorem</subject><subject>biological tissues</subject><subject>biomedical ultrasonics</subject><subject>Breast Neoplasms - diagnosis</subject><subject>Breast Neoplasms - diagnostic imaging</subject><subject>breast sonography</subject><subject>cancer</subject><subject>computer‐aided diagnosis</subject><subject>Databases as Topic</subject><subject>Diseases</subject><subject>False Positive Reactions</subject><subject>Female</subject><subject>Humans</subject><subject>Image analysis</subject><subject>image classification</subject><subject>Image Processing, Computer-Assisted</subject><subject>image segmentation</subject><subject>Mammography</subject><subject>Mass Screening</subject><subject>Medical diagnosis with acoustics</subject><subject>medical image processing</subject><subject>Medical imaging</subject><subject>Models, Statistical</subject><subject>neural nets</subject><subject>Neural networks, fuzzy logic, artificial intelligence</subject><subject>performance of lesion detection</subject><subject>Physicists</subject><subject>Probability theory</subject><subject>Probability theory, stochastic processes, and statistics</subject><subject>Radiologists</subject><subject>ROC Curve</subject><subject>Sensitivity and Specificity</subject><subject>Software</subject><subject>Ultrasonography</subject><subject>Ultrasonography - methods</subject><issn>0094-2405</issn><issn>2473-4209</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2002</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp90MtKxDAUBuAgijOOLnwBmZWg0DEnl7ZZyuANRnSh65ArVNppTVplfHo7tKggCoGTxXd-Dj9Cx4AXAJBfwAJYzoXgO2hKWEYTRrDYRVOMBUsIw3yCDmJ8wRinlON9NAHSL2SETRFZ1lXTtS4UH87OSxeLej23rnWm3f76p4NTsZ13ZRtUrLu1PUR7XpXRHY1zhp6vr56Wt8nq4eZueblKDIOcJ8J44pUFzrED7a1ljOtcpwaniqdYWKa0xjll1mNLqDBUUwyZYJ4BA6roDJ0OuU2oXzsXW1kV0biyVGtXd1FmIHiac9rDswGaUMcYnJdNKCoVNhKw3BYkQY4F9fZkDO105ey3HBvpQTKA96J0m7-T5P3jGHg--GiKVm1L-9p5q8MP31j_H_596iej8YlD</recordid><startdate>200207</startdate><enddate>200207</enddate><creator>Drukker, Karen</creator><creator>Giger, Maryellen L.</creator><creator>Horsch, Karla</creator><creator>Kupinski, Matthew A.</creator><creator>Vyborny, Carl J.</creator><creator>Mendelson, Ellen B.</creator><general>American Association of Physicists in Medicine</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope></search><sort><creationdate>200207</creationdate><title>Computerized lesion detection on breast ultrasound</title><author>Drukker, Karen ; Giger, Maryellen L. ; Horsch, Karla ; Kupinski, Matthew A. ; Vyborny, Carl J. ; Mendelson, Ellen B.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c4185-9cf2fad1550e1bfdd445b8b6c06a5609d4abb0834df0d239c3b301794f41413a3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2002</creationdate><topic>Artificial neural networks</topic><topic>Bayes methods</topic><topic>Bayes Theorem</topic><topic>biological tissues</topic><topic>biomedical ultrasonics</topic><topic>Breast Neoplasms - diagnosis</topic><topic>Breast Neoplasms - diagnostic imaging</topic><topic>breast sonography</topic><topic>cancer</topic><topic>computer‐aided diagnosis</topic><topic>Databases as Topic</topic><topic>Diseases</topic><topic>False Positive Reactions</topic><topic>Female</topic><topic>Humans</topic><topic>Image analysis</topic><topic>image classification</topic><topic>Image Processing, Computer-Assisted</topic><topic>image segmentation</topic><topic>Mammography</topic><topic>Mass Screening</topic><topic>Medical diagnosis with acoustics</topic><topic>medical image processing</topic><topic>Medical imaging</topic><topic>Models, Statistical</topic><topic>neural nets</topic><topic>Neural networks, fuzzy logic, artificial intelligence</topic><topic>performance of lesion detection</topic><topic>Physicists</topic><topic>Probability theory</topic><topic>Probability theory, stochastic processes, and statistics</topic><topic>Radiologists</topic><topic>ROC Curve</topic><topic>Sensitivity and Specificity</topic><topic>Software</topic><topic>Ultrasonography</topic><topic>Ultrasonography - methods</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Drukker, Karen</creatorcontrib><creatorcontrib>Giger, Maryellen L.</creatorcontrib><creatorcontrib>Horsch, Karla</creatorcontrib><creatorcontrib>Kupinski, Matthew A.</creatorcontrib><creatorcontrib>Vyborny, Carl J.</creatorcontrib><creatorcontrib>Mendelson, Ellen B.</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Medical physics (Lancaster)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Drukker, Karen</au><au>Giger, Maryellen L.</au><au>Horsch, Karla</au><au>Kupinski, Matthew A.</au><au>Vyborny, Carl J.</au><au>Mendelson, Ellen B.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Computerized lesion detection on breast ultrasound</atitle><jtitle>Medical physics (Lancaster)</jtitle><addtitle>Med Phys</addtitle><date>2002-07</date><risdate>2002</risdate><volume>29</volume><issue>7</issue><spage>1438</spage><epage>1446</epage><pages>1438-1446</pages><issn>0094-2405</issn><eissn>2473-4209</eissn><coden>MPHYA6</coden><abstract>We investigated the use of a radial gradient index (RGI) filtering technique to automatically detect lesions on breast ultrasound. After initial RGI filtering, a sensitivity of 87% at 0.76 false-positive detections per image was obtained on a database of 400 patients (757 images). Next, lesion candidates were segmented from the background by maximizing an average radial gradient (ARD) index for regions grown from the detected points. At an overlap of 0.4 with a radiologist lesion outline, 75% of the lesions were correctly detected. Subsequently, round robin analysis was used to assess the quality of the classification of lesion candidates into actual lesions and false-positives by a Bayesian neural network. The round robin analysis yielded an
A
z
value of 0.84, and an overall performance by case of 94% sensitivity at 0.48 false-positives per image. Use of computerized analysis of breast sonograms may ultimately facilitate the use of sonography in breast cancer screening programs.</abstract><cop>United States</cop><pub>American Association of Physicists in Medicine</pub><pmid>12148724</pmid><doi>10.1118/1.1485995</doi><tpages>9</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0094-2405 |
| ispartof | Medical physics (Lancaster), 2002-07, Vol.29 (7), p.1438-1446 |
| issn | 0094-2405 2473-4209 |
| language | eng |
| recordid | cdi_scitation_primary_10_1118_1_1485995 |
| source | MEDLINE; Wiley Online Library Journals Frontfile Complete |
| subjects | Artificial neural networks Bayes methods Bayes Theorem biological tissues biomedical ultrasonics Breast Neoplasms - diagnosis Breast Neoplasms - diagnostic imaging breast sonography cancer computer‐aided diagnosis Databases as Topic Diseases False Positive Reactions Female Humans Image analysis image classification Image Processing, Computer-Assisted image segmentation Mammography Mass Screening Medical diagnosis with acoustics medical image processing Medical imaging Models, Statistical neural nets Neural networks, fuzzy logic, artificial intelligence performance of lesion detection Physicists Probability theory Probability theory, stochastic processes, and statistics Radiologists ROC Curve Sensitivity and Specificity Software Ultrasonography Ultrasonography - methods |
| title | Computerized lesion detection on breast ultrasound |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-22T06%3A32%3A44IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_scita&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Computerized%20lesion%20detection%20on%20breast%20ultrasound&rft.jtitle=Medical%20physics%20(Lancaster)&rft.au=Drukker,%20Karen&rft.date=2002-07&rft.volume=29&rft.issue=7&rft.spage=1438&rft.epage=1446&rft.pages=1438-1446&rft.issn=0094-2405&rft.eissn=2473-4209&rft.coden=MPHYA6&rft_id=info:doi/10.1118/1.1485995&rft_dat=%3Cproquest_scita%3E71956853%3C/proquest_scita%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=71956853&rft_id=info:pmid/12148724&rfr_iscdi=true |