Loading…

Simple delineations cannot substitute full 3d tumor delineations for MR-based radiomics prediction of locoregional control in oropharyngeal cancer

Manual delineation of head and neck tumor contours for radiomics analyses is tedious and time consuming. This study investigates if fast or readily available tumor contours can substitute full tumor contours by an experienced observer for an MR-based radiomics model to predict locoregional control (...

Full description

Saved in:
Bibliographic Details
Published in:European journal of radiology 2022-03, Vol.148, p.110167-110167, Article 110167
Main Authors: Bos, Paula, van den Brekel, Michiel W.M., Taghavi, Marjaneh, Gouw, Zeno A.R., Al-Mamgani, Abrahim, Waktola, Selam, Aerts, Hugo J.W.L., Beets-Tan, Regina G.H., Castelijns, Jonas A., Jasperse, Bas
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-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873
cites cdi_FETCH-LOGICAL-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873
container_end_page 110167
container_issue
container_start_page 110167
container_title European journal of radiology
container_volume 148
creator Bos, Paula
van den Brekel, Michiel W.M.
Taghavi, Marjaneh
Gouw, Zeno A.R.
Al-Mamgani, Abrahim
Waktola, Selam
Aerts, Hugo J.W.L.
Beets-Tan, Regina G.H.
Castelijns, Jonas A.
Jasperse, Bas
description Manual delineation of head and neck tumor contours for radiomics analyses is tedious and time consuming. This study investigates if fast or readily available tumor contours can substitute full tumor contours by an experienced observer for an MR-based radiomics model to predict locoregional control (LRC) in oropharyngeal squamous cell carcinoma (OPSCC) tumors. Radiomic features were extracted from postcontrast T1-weighted MRIs of 177 OPSCC primary tumors using six different manual delineation strategies. LRC prediction models based on recursive feature elimination combined with logistic regression were built. Models were trained and tested on data from each separate delineation. Additionally, the model derived from segmentations from the experienced reader was tested by each of the alternative delineations. Complementary, this was repeated with removal of size and shape features. Model performance was evaluated using area under the curve (AUC). Prediction performance of the experienced radiologist tumor delineation (AUC: 0.74) was superior compared to all other delineations when trained and tested (AUCs: 0.41–0.56) or trained on experienced delineations and tested (AUC: 0.56–0.67) on alternative segmentations. Removal of size and shape features considerably decreases prediction performance (AUC: 0.54). Applying the model based on expert delineations to spherical or single slice delineations makes prediction worthless since these models predict one class. Fast or readily available contours cannot substitute full expert tumor delineations in radiomics models predictive of LRC in OPSCC.
doi_str_mv 10.1016/j.ejrad.2022.110167
format article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2623884108</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0720048X22000171</els_id><sourcerecordid>2623884108</sourcerecordid><originalsourceid>FETCH-LOGICAL-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873</originalsourceid><addsrcrecordid>eNp9kc1u1TAQhS0EopfSJ6iEvGSTi-3EsbNggSr-pCIkaCV2lmOPW185cWo7SLwGT4zTWyqxYWXNzDeemXMQOqdkTwnt3xz2cEja7hlhbE-3lHiCdlQK1gjBxFO0I4KRhnTyxwl6kfOBEMK7gT1HJy0nsq_RDv3-7qclALYQ_Ay6-DhnbPQ8x4LzOubiy1oAuzUE3Fpc1immf2FXE1--NaPOYHHdx8fJm4yXBNabDcHR4RBNTHBTIx2wiXNJMWBfSykutzr9mm9gK-jZQHqJnjkdMpw9vKfo-sP7q4tPzeXXj58v3l02puVDaQzowXbcdkbY3g0j76CTox3FSFtwjgs-GgrOSOr6wbkeenDcSuv0wFstRXuKXh__XVK8WyEXNflsIAQ9Q1yzYj1rpewokRVtj6hJMecETi3JT3VvRYnalFcHdW-G2sxQRzNq16uHAes4gX3s-at-Bd4eAahn_vSQVDYeqgbWJzBF2ej_O-APTzKhLg</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2623884108</pqid></control><display><type>article</type><title>Simple delineations cannot substitute full 3d tumor delineations for MR-based radiomics prediction of locoregional control in oropharyngeal cancer</title><source>ScienceDirect Journals</source><creator>Bos, Paula ; van den Brekel, Michiel W.M. ; Taghavi, Marjaneh ; Gouw, Zeno A.R. ; Al-Mamgani, Abrahim ; Waktola, Selam ; Aerts, Hugo J.W.L. ; Beets-Tan, Regina G.H. ; Castelijns, Jonas A. ; Jasperse, Bas</creator><creatorcontrib>Bos, Paula ; van den Brekel, Michiel W.M. ; Taghavi, Marjaneh ; Gouw, Zeno A.R. ; Al-Mamgani, Abrahim ; Waktola, Selam ; Aerts, Hugo J.W.L. ; Beets-Tan, Regina G.H. ; Castelijns, Jonas A. ; Jasperse, Bas</creatorcontrib><description>Manual delineation of head and neck tumor contours for radiomics analyses is tedious and time consuming. This study investigates if fast or readily available tumor contours can substitute full tumor contours by an experienced observer for an MR-based radiomics model to predict locoregional control (LRC) in oropharyngeal squamous cell carcinoma (OPSCC) tumors. Radiomic features were extracted from postcontrast T1-weighted MRIs of 177 OPSCC primary tumors using six different manual delineation strategies. LRC prediction models based on recursive feature elimination combined with logistic regression were built. Models were trained and tested on data from each separate delineation. Additionally, the model derived from segmentations from the experienced reader was tested by each of the alternative delineations. Complementary, this was repeated with removal of size and shape features. Model performance was evaluated using area under the curve (AUC). Prediction performance of the experienced radiologist tumor delineation (AUC: 0.74) was superior compared to all other delineations when trained and tested (AUCs: 0.41–0.56) or trained on experienced delineations and tested (AUC: 0.56–0.67) on alternative segmentations. Removal of size and shape features considerably decreases prediction performance (AUC: 0.54). Applying the model based on expert delineations to spherical or single slice delineations makes prediction worthless since these models predict one class. Fast or readily available contours cannot substitute full expert tumor delineations in radiomics models predictive of LRC in OPSCC.</description><identifier>ISSN: 0720-048X</identifier><identifier>EISSN: 1872-7727</identifier><identifier>DOI: 10.1016/j.ejrad.2022.110167</identifier><identifier>PMID: 35086005</identifier><language>eng</language><publisher>Ireland: Elsevier B.V</publisher><subject>Head and neck neoplasms ; Head and Neck Neoplasms - diagnostic imaging ; Humans ; Machine learning ; Magnetic Resonance Imaging ; Oropharyngeal neoplasms ; Oropharyngeal Neoplasms - diagnostic imaging ; Outcome prediction ; Radiomics ; Retrospective Studies ; Squamous Cell Carcinoma of Head and Neck</subject><ispartof>European journal of radiology, 2022-03, Vol.148, p.110167-110167, Article 110167</ispartof><rights>2022 Elsevier B.V.</rights><rights>Copyright © 2022 Elsevier B.V. All rights reserved.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873</citedby><cites>FETCH-LOGICAL-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/35086005$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Bos, Paula</creatorcontrib><creatorcontrib>van den Brekel, Michiel W.M.</creatorcontrib><creatorcontrib>Taghavi, Marjaneh</creatorcontrib><creatorcontrib>Gouw, Zeno A.R.</creatorcontrib><creatorcontrib>Al-Mamgani, Abrahim</creatorcontrib><creatorcontrib>Waktola, Selam</creatorcontrib><creatorcontrib>Aerts, Hugo J.W.L.</creatorcontrib><creatorcontrib>Beets-Tan, Regina G.H.</creatorcontrib><creatorcontrib>Castelijns, Jonas A.</creatorcontrib><creatorcontrib>Jasperse, Bas</creatorcontrib><title>Simple delineations cannot substitute full 3d tumor delineations for MR-based radiomics prediction of locoregional control in oropharyngeal cancer</title><title>European journal of radiology</title><addtitle>Eur J Radiol</addtitle><description>Manual delineation of head and neck tumor contours for radiomics analyses is tedious and time consuming. This study investigates if fast or readily available tumor contours can substitute full tumor contours by an experienced observer for an MR-based radiomics model to predict locoregional control (LRC) in oropharyngeal squamous cell carcinoma (OPSCC) tumors. Radiomic features were extracted from postcontrast T1-weighted MRIs of 177 OPSCC primary tumors using six different manual delineation strategies. LRC prediction models based on recursive feature elimination combined with logistic regression were built. Models were trained and tested on data from each separate delineation. Additionally, the model derived from segmentations from the experienced reader was tested by each of the alternative delineations. Complementary, this was repeated with removal of size and shape features. Model performance was evaluated using area under the curve (AUC). Prediction performance of the experienced radiologist tumor delineation (AUC: 0.74) was superior compared to all other delineations when trained and tested (AUCs: 0.41–0.56) or trained on experienced delineations and tested (AUC: 0.56–0.67) on alternative segmentations. Removal of size and shape features considerably decreases prediction performance (AUC: 0.54). Applying the model based on expert delineations to spherical or single slice delineations makes prediction worthless since these models predict one class. Fast or readily available contours cannot substitute full expert tumor delineations in radiomics models predictive of LRC in OPSCC.</description><subject>Head and neck neoplasms</subject><subject>Head and Neck Neoplasms - diagnostic imaging</subject><subject>Humans</subject><subject>Machine learning</subject><subject>Magnetic Resonance Imaging</subject><subject>Oropharyngeal neoplasms</subject><subject>Oropharyngeal Neoplasms - diagnostic imaging</subject><subject>Outcome prediction</subject><subject>Radiomics</subject><subject>Retrospective Studies</subject><subject>Squamous Cell Carcinoma of Head and Neck</subject><issn>0720-048X</issn><issn>1872-7727</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><recordid>eNp9kc1u1TAQhS0EopfSJ6iEvGSTi-3EsbNggSr-pCIkaCV2lmOPW185cWo7SLwGT4zTWyqxYWXNzDeemXMQOqdkTwnt3xz2cEja7hlhbE-3lHiCdlQK1gjBxFO0I4KRhnTyxwl6kfOBEMK7gT1HJy0nsq_RDv3-7qclALYQ_Ay6-DhnbPQ8x4LzOubiy1oAuzUE3Fpc1immf2FXE1--NaPOYHHdx8fJm4yXBNabDcHR4RBNTHBTIx2wiXNJMWBfSykutzr9mm9gK-jZQHqJnjkdMpw9vKfo-sP7q4tPzeXXj58v3l02puVDaQzowXbcdkbY3g0j76CTox3FSFtwjgs-GgrOSOr6wbkeenDcSuv0wFstRXuKXh__XVK8WyEXNflsIAQ9Q1yzYj1rpewokRVtj6hJMecETi3JT3VvRYnalFcHdW-G2sxQRzNq16uHAes4gX3s-at-Bd4eAahn_vSQVDYeqgbWJzBF2ej_O-APTzKhLg</recordid><startdate>202203</startdate><enddate>202203</enddate><creator>Bos, Paula</creator><creator>van den Brekel, Michiel W.M.</creator><creator>Taghavi, Marjaneh</creator><creator>Gouw, Zeno A.R.</creator><creator>Al-Mamgani, Abrahim</creator><creator>Waktola, Selam</creator><creator>Aerts, Hugo J.W.L.</creator><creator>Beets-Tan, Regina G.H.</creator><creator>Castelijns, Jonas A.</creator><creator>Jasperse, Bas</creator><general>Elsevier B.V</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>202203</creationdate><title>Simple delineations cannot substitute full 3d tumor delineations for MR-based radiomics prediction of locoregional control in oropharyngeal cancer</title><author>Bos, Paula ; van den Brekel, Michiel W.M. ; Taghavi, Marjaneh ; Gouw, Zeno A.R. ; Al-Mamgani, Abrahim ; Waktola, Selam ; Aerts, Hugo J.W.L. ; Beets-Tan, Regina G.H. ; Castelijns, Jonas A. ; Jasperse, Bas</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Head and neck neoplasms</topic><topic>Head and Neck Neoplasms - diagnostic imaging</topic><topic>Humans</topic><topic>Machine learning</topic><topic>Magnetic Resonance Imaging</topic><topic>Oropharyngeal neoplasms</topic><topic>Oropharyngeal Neoplasms - diagnostic imaging</topic><topic>Outcome prediction</topic><topic>Radiomics</topic><topic>Retrospective Studies</topic><topic>Squamous Cell Carcinoma of Head and Neck</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bos, Paula</creatorcontrib><creatorcontrib>van den Brekel, Michiel W.M.</creatorcontrib><creatorcontrib>Taghavi, Marjaneh</creatorcontrib><creatorcontrib>Gouw, Zeno A.R.</creatorcontrib><creatorcontrib>Al-Mamgani, Abrahim</creatorcontrib><creatorcontrib>Waktola, Selam</creatorcontrib><creatorcontrib>Aerts, Hugo J.W.L.</creatorcontrib><creatorcontrib>Beets-Tan, Regina G.H.</creatorcontrib><creatorcontrib>Castelijns, Jonas A.</creatorcontrib><creatorcontrib>Jasperse, Bas</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>European journal of radiology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Bos, Paula</au><au>van den Brekel, Michiel W.M.</au><au>Taghavi, Marjaneh</au><au>Gouw, Zeno A.R.</au><au>Al-Mamgani, Abrahim</au><au>Waktola, Selam</au><au>Aerts, Hugo J.W.L.</au><au>Beets-Tan, Regina G.H.</au><au>Castelijns, Jonas A.</au><au>Jasperse, Bas</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Simple delineations cannot substitute full 3d tumor delineations for MR-based radiomics prediction of locoregional control in oropharyngeal cancer</atitle><jtitle>European journal of radiology</jtitle><addtitle>Eur J Radiol</addtitle><date>2022-03</date><risdate>2022</risdate><volume>148</volume><spage>110167</spage><epage>110167</epage><pages>110167-110167</pages><artnum>110167</artnum><issn>0720-048X</issn><eissn>1872-7727</eissn><abstract>Manual delineation of head and neck tumor contours for radiomics analyses is tedious and time consuming. This study investigates if fast or readily available tumor contours can substitute full tumor contours by an experienced observer for an MR-based radiomics model to predict locoregional control (LRC) in oropharyngeal squamous cell carcinoma (OPSCC) tumors. Radiomic features were extracted from postcontrast T1-weighted MRIs of 177 OPSCC primary tumors using six different manual delineation strategies. LRC prediction models based on recursive feature elimination combined with logistic regression were built. Models were trained and tested on data from each separate delineation. Additionally, the model derived from segmentations from the experienced reader was tested by each of the alternative delineations. Complementary, this was repeated with removal of size and shape features. Model performance was evaluated using area under the curve (AUC). Prediction performance of the experienced radiologist tumor delineation (AUC: 0.74) was superior compared to all other delineations when trained and tested (AUCs: 0.41–0.56) or trained on experienced delineations and tested (AUC: 0.56–0.67) on alternative segmentations. Removal of size and shape features considerably decreases prediction performance (AUC: 0.54). Applying the model based on expert delineations to spherical or single slice delineations makes prediction worthless since these models predict one class. Fast or readily available contours cannot substitute full expert tumor delineations in radiomics models predictive of LRC in OPSCC.</abstract><cop>Ireland</cop><pub>Elsevier B.V</pub><pmid>35086005</pmid><doi>10.1016/j.ejrad.2022.110167</doi><tpages>1</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0720-048X
ispartof European journal of radiology, 2022-03, Vol.148, p.110167-110167, Article 110167
issn 0720-048X
1872-7727
language eng
recordid cdi_proquest_miscellaneous_2623884108
source ScienceDirect Journals
subjects Head and neck neoplasms
Head and Neck Neoplasms - diagnostic imaging
Humans
Machine learning
Magnetic Resonance Imaging
Oropharyngeal neoplasms
Oropharyngeal Neoplasms - diagnostic imaging
Outcome prediction
Radiomics
Retrospective Studies
Squamous Cell Carcinoma of Head and Neck
title Simple delineations cannot substitute full 3d tumor delineations for MR-based radiomics prediction of locoregional control in oropharyngeal cancer
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T07%3A51%3A06IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Simple%20delineations%20cannot%20substitute%20full%203d%20tumor%20delineations%20for%20MR-based%20radiomics%20prediction%20of%20locoregional%20control%20in%20oropharyngeal%20cancer&rft.jtitle=European%20journal%20of%20radiology&rft.au=Bos,%20Paula&rft.date=2022-03&rft.volume=148&rft.spage=110167&rft.epage=110167&rft.pages=110167-110167&rft.artnum=110167&rft.issn=0720-048X&rft.eissn=1872-7727&rft_id=info:doi/10.1016/j.ejrad.2022.110167&rft_dat=%3Cproquest_cross%3E2623884108%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c359t-cea9d45d4c7d6f9b54e48bdb7b13eff575bc1efc81f69ff6e6ef5d8dfa953a873%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2623884108&rft_id=info:pmid/35086005&rfr_iscdi=true