Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility
Objectives To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software. Methods T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 c...
Gespeichert in:
| Veröffentlicht in: | European radiology 2022-03, Vol.32 (3), p.1506-1516 |
|---|---|
| Hauptverfasser: | , , , , , , , , , , , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 1516 |
|---|---|
| container_issue | 3 |
| container_start_page | 1506 |
| container_title | European radiology |
| container_volume | 32 |
| creator | Schurink, Niels W. van Kranen, Simon R. Roberti, Sander van Griethuysen, Joost J. M. Bogveradze, Nino Castagnoli, Francesca el Khababi, Najim Bakers, Frans C. H. de Bie, Shira H. Bosma, Gerlof P. T. Cappendijk, Vincent C. Geenen, Remy W. F. Neijenhuis, Peter A. Peterson, Gerald M. Veeken, Cornelis J. Vliegen, Roy F. A. Beets-Tan, Regina G. H. Lambregts, Doenja M. J. |
| description | Objectives
To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software.
Methods
T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 centers. Fifty-two imaging features (14 first-order/6 shape/32 higher-order) were extracted from each scan using whole-volume (expert/non-expert) and single-slice segmentations using two different software packages (PyRadiomics/CapTk). Influence of hardware, acquisition, and patient-intrinsic factors (age/gender/cTN-stage) on ADC was assessed using linear regression. Feature reproducibility was assessed between segmentation methods and software packages using the intraclass correlation coefficient.
Results
Image features differed significantly (
p
60%) caused by hardware and image acquisition differences and less so ( |
| doi_str_mv | 10.1007/s00330-021-08251-8 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_8831294</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2627261618</sourcerecordid><originalsourceid>FETCH-LOGICAL-c474t-c97396ac7ebb0cd02b3137ee527072a7b060e320085e99a8a3d1ca81f86676b93</originalsourceid><addsrcrecordid>eNp9kUtv1TAQhS1ERS-FP8ACWWLDJjC2Ez82SKiiUKmoUgtry3EmraskvthOpf57fHtLeSxYeTHfOZ4zh5BXDN4xAPU-AwgBDXDWgOYda_QTsmGt4A0D3T4lGzBCN8qY9pA8z_kGAAxr1TNyKFrZdYKJDQmXcU0eM40jvXUpuBLiQsNC53UqweNSMNGEvriJfr04pYMrjrploOUaQ6I4jnVGqyS5IcQ5-ExHdGVNWFXbFIfVhz5Mody9IAejmzK-fHiPyPeTT9-OvzRn559Pjz-eNb5VbWm8UcJI5xX2PfgBeF_3VIgdV6C4Uz1IQMEBdIfGOO3EwLzTbNRSKtkbcUQ-7H23az_jsIuQ3GS3Kcwu3dnogv17soRrexVvrdaCcdNWg7cPBin-WDEXO4fscZrcgnHNlneaa6YZlxV98w96U8-51HiWS664ZJLpSvE95VPMOeH4uAwDu2vS7pu0tUl736TdiV7_GeNR8qu6Cog9kOtoucL0--__2P4E_cSqew</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2627261618</pqid></control><display><type>article</type><title>Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility</title><source>MEDLINE</source><source>SpringerLink Journals</source><creator>Schurink, Niels W. ; van Kranen, Simon R. ; Roberti, Sander ; van Griethuysen, Joost J. M. ; Bogveradze, Nino ; Castagnoli, Francesca ; el Khababi, Najim ; Bakers, Frans C. H. ; de Bie, Shira H. ; Bosma, Gerlof P. T. ; Cappendijk, Vincent C. ; Geenen, Remy W. F. ; Neijenhuis, Peter A. ; Peterson, Gerald M. ; Veeken, Cornelis J. ; Vliegen, Roy F. A. ; Beets-Tan, Regina G. H. ; Lambregts, Doenja M. J.</creator><creatorcontrib>Schurink, Niels W. ; van Kranen, Simon R. ; Roberti, Sander ; van Griethuysen, Joost J. M. ; Bogveradze, Nino ; Castagnoli, Francesca ; el Khababi, Najim ; Bakers, Frans C. H. ; de Bie, Shira H. ; Bosma, Gerlof P. T. ; Cappendijk, Vincent C. ; Geenen, Remy W. F. ; Neijenhuis, Peter A. ; Peterson, Gerald M. ; Veeken, Cornelis J. ; Vliegen, Roy F. A. ; Beets-Tan, Regina G. H. ; Lambregts, Doenja M. J.</creatorcontrib><description>Objectives
To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software.
Methods
T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 centers. Fifty-two imaging features (14 first-order/6 shape/32 higher-order) were extracted from each scan using whole-volume (expert/non-expert) and single-slice segmentations using two different software packages (PyRadiomics/CapTk). Influence of hardware, acquisition, and patient-intrinsic factors (age/gender/cTN-stage) on ADC was assessed using linear regression. Feature reproducibility was assessed between segmentation methods and software packages using the intraclass correlation coefficient.
Results
Image features differed significantly (
p
< 0.001) between centers with more substantial variations in ADC compared to T2W-MRI. In total, 64.3% of the variation in mean ADC was explained by differences in hardware and acquisition, compared to 0.4% by patient-intrinsic factors. Feature reproducibility between expert and non-expert segmentations was good to excellent (median ICC 0.89–0.90). Reproducibility for single-slice versus whole-volume segmentations was substantially poorer (median ICC 0.40–0.58). Between software packages, reproducibility was good to excellent (median ICC 0.99) for most features (first-order/shape/GLCM/GLRLM) but poor for higher-order (GLSZM/NGTDM) features (median ICC 0.00–0.41).
Conclusions
Significant variations are present in multicenter MRI data, particularly related to differences in hardware and acquisition, which will likely negatively influence subsequent analysis if not corrected for. Segmentation variations had a minor impact when using whole volume segmentations. Between software packages, higher-order features were less reproducible and caution is warranted when implementing these in prediction models.
Key Points
• Features derived from T2W-MRI and in particular ADC differ significantly between centers when performing multicenter data analysis.
• Variations in ADC are mainly (> 60%) caused by hardware and image acquisition differences and less so (< 1%) by patient- or tumor-intrinsic variations.
• Features derived using different image segmentations (expert/non-expert) were reproducible, provided that whole-volume segmentations were used. When using different feature extraction software packages with similar settings, higher-order features were less reproducible.</description><identifier>ISSN: 0938-7994</identifier><identifier>ISSN: 1432-1084</identifier><identifier>EISSN: 1432-1084</identifier><identifier>DOI: 10.1007/s00330-021-08251-8</identifier><identifier>PMID: 34655313</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer Berlin Heidelberg</publisher><subject>Cancer ; Colorectal cancer ; Computer programs ; Correlation coefficient ; Correlation coefficients ; Data analysis ; Diagnostic Radiology ; Diffusion Magnetic Resonance Imaging ; Feature extraction ; Hardware ; Humans ; Image acquisition ; Image processing ; Image Processing, Computer-Assisted ; Image segmentation ; Imaging ; Imaging Informatics and Artificial Intelligence ; Internal Medicine ; Interventional Radiology ; Magnetic Resonance Imaging ; Medical imaging ; Medicine ; Medicine & Public Health ; Neuroradiology ; Prediction models ; Radiology ; Radiomics ; Rectal Neoplasms - diagnostic imaging ; Rectum ; Reproducibility ; Reproducibility of Results ; Retrospective Studies ; Software ; Software packages ; Tumors ; Ultrasound ; Variation</subject><ispartof>European radiology, 2022-03, Vol.32 (3), p.1506-1516</ispartof><rights>The Author(s) 2021. corrected publication 2022</rights><rights>2021. The Author(s).</rights><rights>The Author(s) 2021. corrected publication 2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>The Author(s) 2021, corrected publication 2022</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c474t-c97396ac7ebb0cd02b3137ee527072a7b060e320085e99a8a3d1ca81f86676b93</citedby><cites>FETCH-LOGICAL-c474t-c97396ac7ebb0cd02b3137ee527072a7b060e320085e99a8a3d1ca81f86676b93</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s00330-021-08251-8$$EPDF$$P50$$Gspringer$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s00330-021-08251-8$$EHTML$$P50$$Gspringer$$Hfree_for_read</linktohtml><link.rule.ids>230,314,776,780,881,27901,27902,41464,42533,51294</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/34655313$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Schurink, Niels W.</creatorcontrib><creatorcontrib>van Kranen, Simon R.</creatorcontrib><creatorcontrib>Roberti, Sander</creatorcontrib><creatorcontrib>van Griethuysen, Joost J. M.</creatorcontrib><creatorcontrib>Bogveradze, Nino</creatorcontrib><creatorcontrib>Castagnoli, Francesca</creatorcontrib><creatorcontrib>el Khababi, Najim</creatorcontrib><creatorcontrib>Bakers, Frans C. H.</creatorcontrib><creatorcontrib>de Bie, Shira H.</creatorcontrib><creatorcontrib>Bosma, Gerlof P. T.</creatorcontrib><creatorcontrib>Cappendijk, Vincent C.</creatorcontrib><creatorcontrib>Geenen, Remy W. F.</creatorcontrib><creatorcontrib>Neijenhuis, Peter A.</creatorcontrib><creatorcontrib>Peterson, Gerald M.</creatorcontrib><creatorcontrib>Veeken, Cornelis J.</creatorcontrib><creatorcontrib>Vliegen, Roy F. A.</creatorcontrib><creatorcontrib>Beets-Tan, Regina G. H.</creatorcontrib><creatorcontrib>Lambregts, Doenja M. J.</creatorcontrib><title>Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility</title><title>European radiology</title><addtitle>Eur Radiol</addtitle><addtitle>Eur Radiol</addtitle><description>Objectives
To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software.
Methods
T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 centers. Fifty-two imaging features (14 first-order/6 shape/32 higher-order) were extracted from each scan using whole-volume (expert/non-expert) and single-slice segmentations using two different software packages (PyRadiomics/CapTk). Influence of hardware, acquisition, and patient-intrinsic factors (age/gender/cTN-stage) on ADC was assessed using linear regression. Feature reproducibility was assessed between segmentation methods and software packages using the intraclass correlation coefficient.
Results
Image features differed significantly (
p
< 0.001) between centers with more substantial variations in ADC compared to T2W-MRI. In total, 64.3% of the variation in mean ADC was explained by differences in hardware and acquisition, compared to 0.4% by patient-intrinsic factors. Feature reproducibility between expert and non-expert segmentations was good to excellent (median ICC 0.89–0.90). Reproducibility for single-slice versus whole-volume segmentations was substantially poorer (median ICC 0.40–0.58). Between software packages, reproducibility was good to excellent (median ICC 0.99) for most features (first-order/shape/GLCM/GLRLM) but poor for higher-order (GLSZM/NGTDM) features (median ICC 0.00–0.41).
Conclusions
Significant variations are present in multicenter MRI data, particularly related to differences in hardware and acquisition, which will likely negatively influence subsequent analysis if not corrected for. Segmentation variations had a minor impact when using whole volume segmentations. Between software packages, higher-order features were less reproducible and caution is warranted when implementing these in prediction models.
Key Points
• Features derived from T2W-MRI and in particular ADC differ significantly between centers when performing multicenter data analysis.
• Variations in ADC are mainly (> 60%) caused by hardware and image acquisition differences and less so (< 1%) by patient- or tumor-intrinsic variations.
• Features derived using different image segmentations (expert/non-expert) were reproducible, provided that whole-volume segmentations were used. When using different feature extraction software packages with similar settings, higher-order features were less reproducible.</description><subject>Cancer</subject><subject>Colorectal cancer</subject><subject>Computer programs</subject><subject>Correlation coefficient</subject><subject>Correlation coefficients</subject><subject>Data analysis</subject><subject>Diagnostic Radiology</subject><subject>Diffusion Magnetic Resonance Imaging</subject><subject>Feature extraction</subject><subject>Hardware</subject><subject>Humans</subject><subject>Image acquisition</subject><subject>Image processing</subject><subject>Image Processing, Computer-Assisted</subject><subject>Image segmentation</subject><subject>Imaging</subject><subject>Imaging Informatics and Artificial Intelligence</subject><subject>Internal Medicine</subject><subject>Interventional Radiology</subject><subject>Magnetic Resonance Imaging</subject><subject>Medical imaging</subject><subject>Medicine</subject><subject>Medicine & Public Health</subject><subject>Neuroradiology</subject><subject>Prediction models</subject><subject>Radiology</subject><subject>Radiomics</subject><subject>Rectal Neoplasms - diagnostic imaging</subject><subject>Rectum</subject><subject>Reproducibility</subject><subject>Reproducibility of Results</subject><subject>Retrospective Studies</subject><subject>Software</subject><subject>Software packages</subject><subject>Tumors</subject><subject>Ultrasound</subject><subject>Variation</subject><issn>0938-7994</issn><issn>1432-1084</issn><issn>1432-1084</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>C6C</sourceid><sourceid>EIF</sourceid><sourceid>BENPR</sourceid><recordid>eNp9kUtv1TAQhS1ERS-FP8ACWWLDJjC2Ez82SKiiUKmoUgtry3EmraskvthOpf57fHtLeSxYeTHfOZ4zh5BXDN4xAPU-AwgBDXDWgOYda_QTsmGt4A0D3T4lGzBCN8qY9pA8z_kGAAxr1TNyKFrZdYKJDQmXcU0eM40jvXUpuBLiQsNC53UqweNSMNGEvriJfr04pYMrjrploOUaQ6I4jnVGqyS5IcQ5-ExHdGVNWFXbFIfVhz5Mody9IAejmzK-fHiPyPeTT9-OvzRn559Pjz-eNb5VbWm8UcJI5xX2PfgBeF_3VIgdV6C4Uz1IQMEBdIfGOO3EwLzTbNRSKtkbcUQ-7H23az_jsIuQ3GS3Kcwu3dnogv17soRrexVvrdaCcdNWg7cPBin-WDEXO4fscZrcgnHNlneaa6YZlxV98w96U8-51HiWS664ZJLpSvE95VPMOeH4uAwDu2vS7pu0tUl736TdiV7_GeNR8qu6Cog9kOtoucL0--__2P4E_cSqew</recordid><startdate>20220301</startdate><enddate>20220301</enddate><creator>Schurink, Niels W.</creator><creator>van Kranen, Simon R.</creator><creator>Roberti, Sander</creator><creator>van Griethuysen, Joost J. M.</creator><creator>Bogveradze, Nino</creator><creator>Castagnoli, Francesca</creator><creator>el Khababi, Najim</creator><creator>Bakers, Frans C. H.</creator><creator>de Bie, Shira H.</creator><creator>Bosma, Gerlof P. T.</creator><creator>Cappendijk, Vincent C.</creator><creator>Geenen, Remy W. F.</creator><creator>Neijenhuis, Peter A.</creator><creator>Peterson, Gerald M.</creator><creator>Veeken, Cornelis J.</creator><creator>Vliegen, Roy F. A.</creator><creator>Beets-Tan, Regina G. H.</creator><creator>Lambregts, Doenja M. J.</creator><general>Springer Berlin Heidelberg</general><general>Springer Nature B.V</general><scope>C6C</scope><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>3V.</scope><scope>7QO</scope><scope>7RV</scope><scope>7X7</scope><scope>7XB</scope><scope>88E</scope><scope>8AO</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>8FH</scope><scope>8FI</scope><scope>8FJ</scope><scope>8FK</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BBNVY</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>BHPHI</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>FR3</scope><scope>FYUFA</scope><scope>GHDGH</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>K9.</scope><scope>KB0</scope><scope>LK8</scope><scope>M0S</scope><scope>M1P</scope><scope>M7P</scope><scope>NAPCQ</scope><scope>P5Z</scope><scope>P62</scope><scope>P64</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>7X8</scope><scope>5PM</scope></search><sort><creationdate>20220301</creationdate><title>Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility</title><author>Schurink, Niels W. ; van Kranen, Simon R. ; Roberti, Sander ; van Griethuysen, Joost J. M. ; Bogveradze, Nino ; Castagnoli, Francesca ; el Khababi, Najim ; Bakers, Frans C. H. ; de Bie, Shira H. ; Bosma, Gerlof P. T. ; Cappendijk, Vincent C. ; Geenen, Remy W. F. ; Neijenhuis, Peter A. ; Peterson, Gerald M. ; Veeken, Cornelis J. ; Vliegen, Roy F. A. ; Beets-Tan, Regina G. H. ; Lambregts, Doenja M. J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c474t-c97396ac7ebb0cd02b3137ee527072a7b060e320085e99a8a3d1ca81f86676b93</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Cancer</topic><topic>Colorectal cancer</topic><topic>Computer programs</topic><topic>Correlation coefficient</topic><topic>Correlation coefficients</topic><topic>Data analysis</topic><topic>Diagnostic Radiology</topic><topic>Diffusion Magnetic Resonance Imaging</topic><topic>Feature extraction</topic><topic>Hardware</topic><topic>Humans</topic><topic>Image acquisition</topic><topic>Image processing</topic><topic>Image Processing, Computer-Assisted</topic><topic>Image segmentation</topic><topic>Imaging</topic><topic>Imaging Informatics and Artificial Intelligence</topic><topic>Internal Medicine</topic><topic>Interventional Radiology</topic><topic>Magnetic Resonance Imaging</topic><topic>Medical imaging</topic><topic>Medicine</topic><topic>Medicine & Public Health</topic><topic>Neuroradiology</topic><topic>Prediction models</topic><topic>Radiology</topic><topic>Radiomics</topic><topic>Rectal Neoplasms - diagnostic imaging</topic><topic>Rectum</topic><topic>Reproducibility</topic><topic>Reproducibility of Results</topic><topic>Retrospective Studies</topic><topic>Software</topic><topic>Software packages</topic><topic>Tumors</topic><topic>Ultrasound</topic><topic>Variation</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Schurink, Niels W.</creatorcontrib><creatorcontrib>van Kranen, Simon R.</creatorcontrib><creatorcontrib>Roberti, Sander</creatorcontrib><creatorcontrib>van Griethuysen, Joost J. M.</creatorcontrib><creatorcontrib>Bogveradze, Nino</creatorcontrib><creatorcontrib>Castagnoli, Francesca</creatorcontrib><creatorcontrib>el Khababi, Najim</creatorcontrib><creatorcontrib>Bakers, Frans C. H.</creatorcontrib><creatorcontrib>de Bie, Shira H.</creatorcontrib><creatorcontrib>Bosma, Gerlof P. T.</creatorcontrib><creatorcontrib>Cappendijk, Vincent C.</creatorcontrib><creatorcontrib>Geenen, Remy W. F.</creatorcontrib><creatorcontrib>Neijenhuis, Peter A.</creatorcontrib><creatorcontrib>Peterson, Gerald M.</creatorcontrib><creatorcontrib>Veeken, Cornelis J.</creatorcontrib><creatorcontrib>Vliegen, Roy F. A.</creatorcontrib><creatorcontrib>Beets-Tan, Regina G. H.</creatorcontrib><creatorcontrib>Lambregts, Doenja M. J.</creatorcontrib><collection>Springer Nature OA Free Journals</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>Biotechnology Research Abstracts</collection><collection>Nursing & Allied Health Database</collection><collection>Health & Medical Collection</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Medical Database (Alumni Edition)</collection><collection>ProQuest Pharma Collection</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Natural Science Collection</collection><collection>Hospital Premium Collection</collection><collection>Hospital Premium Collection (Alumni Edition)</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>Advanced Technologies & Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>Biological Science Collection</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>Natural Science Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>Engineering Research Database</collection><collection>Health Research Premium Collection</collection><collection>Health Research Premium Collection (Alumni)</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Nursing & Allied Health Database (Alumni Edition)</collection><collection>ProQuest Biological Science Collection</collection><collection>Health & Medical Collection (Alumni Edition)</collection><collection>Medical Database</collection><collection>Biological Science Database</collection><collection>Nursing & Allied Health Premium</collection><collection>Advanced Technologies & Aerospace Database</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>European radiology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Schurink, Niels W.</au><au>van Kranen, Simon R.</au><au>Roberti, Sander</au><au>van Griethuysen, Joost J. M.</au><au>Bogveradze, Nino</au><au>Castagnoli, Francesca</au><au>el Khababi, Najim</au><au>Bakers, Frans C. H.</au><au>de Bie, Shira H.</au><au>Bosma, Gerlof P. T.</au><au>Cappendijk, Vincent C.</au><au>Geenen, Remy W. F.</au><au>Neijenhuis, Peter A.</au><au>Peterson, Gerald M.</au><au>Veeken, Cornelis J.</au><au>Vliegen, Roy F. A.</au><au>Beets-Tan, Regina G. H.</au><au>Lambregts, Doenja M. J.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility</atitle><jtitle>European radiology</jtitle><stitle>Eur Radiol</stitle><addtitle>Eur Radiol</addtitle><date>2022-03-01</date><risdate>2022</risdate><volume>32</volume><issue>3</issue><spage>1506</spage><epage>1516</epage><pages>1506-1516</pages><issn>0938-7994</issn><issn>1432-1084</issn><eissn>1432-1084</eissn><abstract>Objectives
To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software.
Methods
T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 centers. Fifty-two imaging features (14 first-order/6 shape/32 higher-order) were extracted from each scan using whole-volume (expert/non-expert) and single-slice segmentations using two different software packages (PyRadiomics/CapTk). Influence of hardware, acquisition, and patient-intrinsic factors (age/gender/cTN-stage) on ADC was assessed using linear regression. Feature reproducibility was assessed between segmentation methods and software packages using the intraclass correlation coefficient.
Results
Image features differed significantly (
p
< 0.001) between centers with more substantial variations in ADC compared to T2W-MRI. In total, 64.3% of the variation in mean ADC was explained by differences in hardware and acquisition, compared to 0.4% by patient-intrinsic factors. Feature reproducibility between expert and non-expert segmentations was good to excellent (median ICC 0.89–0.90). Reproducibility for single-slice versus whole-volume segmentations was substantially poorer (median ICC 0.40–0.58). Between software packages, reproducibility was good to excellent (median ICC 0.99) for most features (first-order/shape/GLCM/GLRLM) but poor for higher-order (GLSZM/NGTDM) features (median ICC 0.00–0.41).
Conclusions
Significant variations are present in multicenter MRI data, particularly related to differences in hardware and acquisition, which will likely negatively influence subsequent analysis if not corrected for. Segmentation variations had a minor impact when using whole volume segmentations. Between software packages, higher-order features were less reproducible and caution is warranted when implementing these in prediction models.
Key Points
• Features derived from T2W-MRI and in particular ADC differ significantly between centers when performing multicenter data analysis.
• Variations in ADC are mainly (> 60%) caused by hardware and image acquisition differences and less so (< 1%) by patient- or tumor-intrinsic variations.
• Features derived using different image segmentations (expert/non-expert) were reproducible, provided that whole-volume segmentations were used. When using different feature extraction software packages with similar settings, higher-order features were less reproducible.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><pmid>34655313</pmid><doi>10.1007/s00330-021-08251-8</doi><tpages>11</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0938-7994 |
| ispartof | European radiology, 2022-03, Vol.32 (3), p.1506-1516 |
| issn | 0938-7994 1432-1084 1432-1084 |
| language | eng |
| recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_8831294 |
| source | MEDLINE; SpringerLink Journals |
| subjects | Cancer Colorectal cancer Computer programs Correlation coefficient Correlation coefficients Data analysis Diagnostic Radiology Diffusion Magnetic Resonance Imaging Feature extraction Hardware Humans Image acquisition Image processing Image Processing, Computer-Assisted Image segmentation Imaging Imaging Informatics and Artificial Intelligence Internal Medicine Interventional Radiology Magnetic Resonance Imaging Medical imaging Medicine Medicine & Public Health Neuroradiology Prediction models Radiology Radiomics Rectal Neoplasms - diagnostic imaging Rectum Reproducibility Reproducibility of Results Retrospective Studies Software Software packages Tumors Ultrasound Variation |
| title | Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T04%3A32%3A06IST&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=Sources%20of%20variation%20in%20multicenter%20rectal%20MRI%20data%20and%20their%20effect%20on%20radiomics%20feature%20reproducibility&rft.jtitle=European%20radiology&rft.au=Schurink,%20Niels%20W.&rft.date=2022-03-01&rft.volume=32&rft.issue=3&rft.spage=1506&rft.epage=1516&rft.pages=1506-1516&rft.issn=0938-7994&rft.eissn=1432-1084&rft_id=info:doi/10.1007/s00330-021-08251-8&rft_dat=%3Cproquest_pubme%3E2627261618%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2627261618&rft_id=info:pmid/34655313&rfr_iscdi=true |