Loading…
Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis
To evaluate whether radiomics features of late gadolinium enhancement (LGE) regions at cardiac MRI enable distinction between myocardial infarction (MI) and myocarditis and to compare radiomics with subjective visual analyses by readers with different experience levels. In this retrospective, instit...
Saved in:
| Published in: | Radiology. Cardiothoracic imaging 2019-12, Vol.1 (5), p.e180026-e180026 |
|---|---|
| Main Authors: | , , , , , , |
| Format: | Article |
| Language: | English |
| Subjects: | |
| Citations: | Items that this one cites Items that cite this one |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| cited_by | cdi_FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33 |
|---|---|
| cites | cdi_FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33 |
| container_end_page | e180026 |
| container_issue | 5 |
| container_start_page | e180026 |
| container_title | Radiology. Cardiothoracic imaging |
| container_volume | 1 |
| creator | Di Noto, Tommaso von Spiczak, Jochen Mannil, Manoj Gantert, Elena Soda, Paolo Manka, Robert Alkadhi, Hatem |
| description | To evaluate whether radiomics features of late gadolinium enhancement (LGE) regions at cardiac MRI enable distinction between myocardial infarction (MI) and myocarditis and to compare radiomics with subjective visual analyses by readers with different experience levels.
In this retrospective, institutional review board-approved study, consecutive MRI examinations of 111 patients with MI and 62 patients with myocarditis showing LGE were included. By using open-source software, classification performances attained from two-dimensional (2D) and three-dimensional (3D) texture analysis, shape, and first-order descriptors were compared, applying five different machine learning algorithms. A nested, stratified 10-fold cross-validation was performed. Classification performances were compared through Wilcoxon signed-rank tests. Supervised and unsupervised feature selection techniques were tested; the effect of resampling MR images was analyzed. Subjective image analysis was performed on 2D and 3D image sets by two independent, blinded readers with different experience levels.
When trained with recursive feature elimination (RFE), a support vector machine achieved the best results (accuracy: 88%) for 2D features, whereas linear discriminant analysis (LDA) showed the highest accuracy (85%) for 3D features ( |
| doi_str_mv | 10.1148/ryct.2019180026 |
| format | article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7977789</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2506510081</sourcerecordid><originalsourceid>FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33</originalsourceid><addsrcrecordid>eNpVkU1v1DAQhi0EolXpmRvykcu2_kichANStS1lpa2QysfVmthOd6rEXmynaH8I_xevWpZympHed54ZzUvIW87OOK_a87gz-Uww3vGWMaFekGOhZLtQXNYvn_VH5DSle1YsvKpq3r0mR1I2TVuL-pj8vgWLYUKT6BAivcSU0d_NmDal0JtdMBAtwkhXfoBoMgZPhximg5QxUch0DdnRa7BhRI_zRK_8Brxxk_N5L9_crj7QZZi2EDEVxC_MG_p17u9dQT44-gPTXJZceBh3CdMb8mqAMbnTp3pCvn-6-rb8vFh_uV4tL9YLIzuZF72yzjIrBQje1L1TVqneDkq0QlSiYVVfS84Flw7U0FVQVbLjTplBCtV3vZQn5OMjdzv3k7OmXBth1NuIE8SdDoD6f8XjRt-FB910TflgVwDvnwAx_JxdynrCZNw4gndhTlrUTNWcsZYX6_mj1cSQUnTDYQ1nep-n3uep_-VZJt49v-7g_5ue_APhop_f</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2506510081</pqid></control><display><type>article</type><title>Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis</title><source>Open Access: PubMed Central</source><creator>Di Noto, Tommaso ; von Spiczak, Jochen ; Mannil, Manoj ; Gantert, Elena ; Soda, Paolo ; Manka, Robert ; Alkadhi, Hatem</creator><creatorcontrib>Di Noto, Tommaso ; von Spiczak, Jochen ; Mannil, Manoj ; Gantert, Elena ; Soda, Paolo ; Manka, Robert ; Alkadhi, Hatem</creatorcontrib><description>To evaluate whether radiomics features of late gadolinium enhancement (LGE) regions at cardiac MRI enable distinction between myocardial infarction (MI) and myocarditis and to compare radiomics with subjective visual analyses by readers with different experience levels.
In this retrospective, institutional review board-approved study, consecutive MRI examinations of 111 patients with MI and 62 patients with myocarditis showing LGE were included. By using open-source software, classification performances attained from two-dimensional (2D) and three-dimensional (3D) texture analysis, shape, and first-order descriptors were compared, applying five different machine learning algorithms. A nested, stratified 10-fold cross-validation was performed. Classification performances were compared through Wilcoxon signed-rank tests. Supervised and unsupervised feature selection techniques were tested; the effect of resampling MR images was analyzed. Subjective image analysis was performed on 2D and 3D image sets by two independent, blinded readers with different experience levels.
When trained with recursive feature elimination (RFE), a support vector machine achieved the best results (accuracy: 88%) for 2D features, whereas linear discriminant analysis (LDA) showed the highest accuracy (85%) for 3D features (
<.05). When trained with principal component analysis (PCA), LDA attained the highest accuracy with both 2D (86%) and 3D (89%;
=.4) features. Results found for classifiers trained with spline resampling were less accurate than those achieved with one-dimensional (1D) nearest-neighbor interpolation (
<.05), whereas results for classifiers trained with 1D nearest-neighbor interpolation and without resampling were similar (
=.1). As compared with the radiomics approach, subjective visual analysis performance was lower for the less experienced and higher for the experienced reader for both 2D and 3D data.
Radiomics features of LGE permit the distinction between MI and myocarditis with high accuracy by using either 2D features and RFE or 3D features and PCA.© RSNA, 2019</description><identifier>ISSN: 2638-6135</identifier><identifier>EISSN: 2638-6135</identifier><identifier>DOI: 10.1148/ryct.2019180026</identifier><identifier>PMID: 33778525</identifier><language>eng</language><publisher>United States: Radiological Society of North America</publisher><subject>Original Research</subject><ispartof>Radiology. Cardiothoracic imaging, 2019-12, Vol.1 (5), p.e180026-e180026</ispartof><rights>2019 by the Radiological Society of North America, Inc.</rights><rights>2019 by the Radiological Society of North America, Inc. 2019</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33</citedby><cites>FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33</cites><orcidid>0000-0002-5161-055X ; 0000-0002-2581-2166 ; 0000-0002-1978-1535</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/PMC7977789/pdf/$$EPDF$$P50$$Gpubmedcentral$$H</linktopdf><linktohtml>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC7977789/$$EHTML$$P50$$Gpubmedcentral$$H</linktohtml><link.rule.ids>230,314,727,780,784,885,27924,27925,53791,53793</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33778525$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Di Noto, Tommaso</creatorcontrib><creatorcontrib>von Spiczak, Jochen</creatorcontrib><creatorcontrib>Mannil, Manoj</creatorcontrib><creatorcontrib>Gantert, Elena</creatorcontrib><creatorcontrib>Soda, Paolo</creatorcontrib><creatorcontrib>Manka, Robert</creatorcontrib><creatorcontrib>Alkadhi, Hatem</creatorcontrib><title>Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis</title><title>Radiology. Cardiothoracic imaging</title><addtitle>Radiol Cardiothorac Imaging</addtitle><description>To evaluate whether radiomics features of late gadolinium enhancement (LGE) regions at cardiac MRI enable distinction between myocardial infarction (MI) and myocarditis and to compare radiomics with subjective visual analyses by readers with different experience levels.
In this retrospective, institutional review board-approved study, consecutive MRI examinations of 111 patients with MI and 62 patients with myocarditis showing LGE were included. By using open-source software, classification performances attained from two-dimensional (2D) and three-dimensional (3D) texture analysis, shape, and first-order descriptors were compared, applying five different machine learning algorithms. A nested, stratified 10-fold cross-validation was performed. Classification performances were compared through Wilcoxon signed-rank tests. Supervised and unsupervised feature selection techniques were tested; the effect of resampling MR images was analyzed. Subjective image analysis was performed on 2D and 3D image sets by two independent, blinded readers with different experience levels.
When trained with recursive feature elimination (RFE), a support vector machine achieved the best results (accuracy: 88%) for 2D features, whereas linear discriminant analysis (LDA) showed the highest accuracy (85%) for 3D features (
<.05). When trained with principal component analysis (PCA), LDA attained the highest accuracy with both 2D (86%) and 3D (89%;
=.4) features. Results found for classifiers trained with spline resampling were less accurate than those achieved with one-dimensional (1D) nearest-neighbor interpolation (
<.05), whereas results for classifiers trained with 1D nearest-neighbor interpolation and without resampling were similar (
=.1). As compared with the radiomics approach, subjective visual analysis performance was lower for the less experienced and higher for the experienced reader for both 2D and 3D data.
Radiomics features of LGE permit the distinction between MI and myocarditis with high accuracy by using either 2D features and RFE or 3D features and PCA.© RSNA, 2019</description><subject>Original Research</subject><issn>2638-6135</issn><issn>2638-6135</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNpVkU1v1DAQhi0EolXpmRvykcu2_kichANStS1lpa2QysfVmthOd6rEXmynaH8I_xevWpZympHed54ZzUvIW87OOK_a87gz-Uww3vGWMaFekGOhZLtQXNYvn_VH5DSle1YsvKpq3r0mR1I2TVuL-pj8vgWLYUKT6BAivcSU0d_NmDal0JtdMBAtwkhXfoBoMgZPhximg5QxUch0DdnRa7BhRI_zRK_8Brxxk_N5L9_crj7QZZi2EDEVxC_MG_p17u9dQT44-gPTXJZceBh3CdMb8mqAMbnTp3pCvn-6-rb8vFh_uV4tL9YLIzuZF72yzjIrBQje1L1TVqneDkq0QlSiYVVfS84Flw7U0FVQVbLjTplBCtV3vZQn5OMjdzv3k7OmXBth1NuIE8SdDoD6f8XjRt-FB910TflgVwDvnwAx_JxdynrCZNw4gndhTlrUTNWcsZYX6_mj1cSQUnTDYQ1nep-n3uep_-VZJt49v-7g_5ue_APhop_f</recordid><startdate>20191201</startdate><enddate>20191201</enddate><creator>Di Noto, Tommaso</creator><creator>von Spiczak, Jochen</creator><creator>Mannil, Manoj</creator><creator>Gantert, Elena</creator><creator>Soda, Paolo</creator><creator>Manka, Robert</creator><creator>Alkadhi, Hatem</creator><general>Radiological Society of North America</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-5161-055X</orcidid><orcidid>https://orcid.org/0000-0002-2581-2166</orcidid><orcidid>https://orcid.org/0000-0002-1978-1535</orcidid></search><sort><creationdate>20191201</creationdate><title>Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis</title><author>Di Noto, Tommaso ; von Spiczak, Jochen ; Mannil, Manoj ; Gantert, Elena ; Soda, Paolo ; Manka, Robert ; Alkadhi, Hatem</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Original Research</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Di Noto, Tommaso</creatorcontrib><creatorcontrib>von Spiczak, Jochen</creatorcontrib><creatorcontrib>Mannil, Manoj</creatorcontrib><creatorcontrib>Gantert, Elena</creatorcontrib><creatorcontrib>Soda, Paolo</creatorcontrib><creatorcontrib>Manka, Robert</creatorcontrib><creatorcontrib>Alkadhi, Hatem</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Radiology. Cardiothoracic imaging</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Di Noto, Tommaso</au><au>von Spiczak, Jochen</au><au>Mannil, Manoj</au><au>Gantert, Elena</au><au>Soda, Paolo</au><au>Manka, Robert</au><au>Alkadhi, Hatem</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis</atitle><jtitle>Radiology. Cardiothoracic imaging</jtitle><addtitle>Radiol Cardiothorac Imaging</addtitle><date>2019-12-01</date><risdate>2019</risdate><volume>1</volume><issue>5</issue><spage>e180026</spage><epage>e180026</epage><pages>e180026-e180026</pages><issn>2638-6135</issn><eissn>2638-6135</eissn><abstract>To evaluate whether radiomics features of late gadolinium enhancement (LGE) regions at cardiac MRI enable distinction between myocardial infarction (MI) and myocarditis and to compare radiomics with subjective visual analyses by readers with different experience levels.
In this retrospective, institutional review board-approved study, consecutive MRI examinations of 111 patients with MI and 62 patients with myocarditis showing LGE were included. By using open-source software, classification performances attained from two-dimensional (2D) and three-dimensional (3D) texture analysis, shape, and first-order descriptors were compared, applying five different machine learning algorithms. A nested, stratified 10-fold cross-validation was performed. Classification performances were compared through Wilcoxon signed-rank tests. Supervised and unsupervised feature selection techniques were tested; the effect of resampling MR images was analyzed. Subjective image analysis was performed on 2D and 3D image sets by two independent, blinded readers with different experience levels.
When trained with recursive feature elimination (RFE), a support vector machine achieved the best results (accuracy: 88%) for 2D features, whereas linear discriminant analysis (LDA) showed the highest accuracy (85%) for 3D features (
<.05). When trained with principal component analysis (PCA), LDA attained the highest accuracy with both 2D (86%) and 3D (89%;
=.4) features. Results found for classifiers trained with spline resampling were less accurate than those achieved with one-dimensional (1D) nearest-neighbor interpolation (
<.05), whereas results for classifiers trained with 1D nearest-neighbor interpolation and without resampling were similar (
=.1). As compared with the radiomics approach, subjective visual analysis performance was lower for the less experienced and higher for the experienced reader for both 2D and 3D data.
Radiomics features of LGE permit the distinction between MI and myocarditis with high accuracy by using either 2D features and RFE or 3D features and PCA.© RSNA, 2019</abstract><cop>United States</cop><pub>Radiological Society of North America</pub><pmid>33778525</pmid><doi>10.1148/ryct.2019180026</doi><orcidid>https://orcid.org/0000-0002-5161-055X</orcidid><orcidid>https://orcid.org/0000-0002-2581-2166</orcidid><orcidid>https://orcid.org/0000-0002-1978-1535</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 2638-6135 |
| ispartof | Radiology. Cardiothoracic imaging, 2019-12, Vol.1 (5), p.e180026-e180026 |
| issn | 2638-6135 2638-6135 |
| language | eng |
| recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7977789 |
| source | Open Access: PubMed Central |
| subjects | Original Research |
| title | Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T16%3A18%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Radiomics%20for%20Distinguishing%20Myocardial%20Infarction%20from%20Myocarditis%20at%20Late%20Gadolinium%20Enhancement%20at%20MRI:%20Comparison%20with%20Subjective%20Visual%20Analysis&rft.jtitle=Radiology.%20Cardiothoracic%20imaging&rft.au=Di%20Noto,%20Tommaso&rft.date=2019-12-01&rft.volume=1&rft.issue=5&rft.spage=e180026&rft.epage=e180026&rft.pages=e180026-e180026&rft.issn=2638-6135&rft.eissn=2638-6135&rft_id=info:doi/10.1148/ryct.2019180026&rft_dat=%3Cproquest_pubme%3E2506510081%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c393t-b6ded0d32a2175be6d66bdf6282242704b5311213ea6f94a44391e6cf326b9b33%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2506510081&rft_id=info:pmid/33778525&rfr_iscdi=true |