Cross‐Cohort Automatic Knee MRI Segmentation With Multi‐Planar U‐Nets
Background Segmentation of medical image volumes is a time‐consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings. Purpose To evaluate the performance of the open‐source Multi‐Planar U‐Net (MPUnet), the...
Saved in:
| Published in | Journal of magnetic resonance imaging Vol. 55; no. 6; pp. 1650 - 1663 |
|---|---|
| Main Authors | , , , , |
| Format | Journal Article |
| Language | English |
| Published |
Hoboken, USA
John Wiley & Sons, Inc
01.06.2022
Wiley Subscription Services, Inc |
| Subjects | |
| Online Access | Get full text |
| ISSN | 1053-1807 1522-2586 1522-2586 |
| DOI | 10.1002/jmri.27978 |
Cover
| Abstract | Background
Segmentation of medical image volumes is a time‐consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings.
Purpose
To evaluate the performance of the open‐source Multi‐Planar U‐Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state‐of‐the‐art two‐dimensional (2D) U‐Net architecture on three clinical cohorts without extensive adaptation of the algorithms.
Study Type
Retrospective cohort study.
Subjects
A total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0–4).
Field Strength/Sequence
0.18 T, 1.0 T/1.5 T, and 3 T sagittal three‐dimensional fast‐spin echo T1w and dual‐echo steady‐state sequences.
Assessment
All models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades.
Statistical Tests
Segmentation performance differences as measured by Dice coefficients were tested with paired, two‐sided Wilcoxon signed‐rank statistics with significance threshold α = 0.05.
Results
The MPUnet performed superior or equal to KIQ and 2D U‐Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U‐Net on CCBR (0.83±0.04 vs. 0.81±0.06 and 0.82±0.05), significantly higher than KIQ and U‐Net OAI (0.86±0.03 vs. 0.84±0.04 and 0.85±0.03), and not significantly different from KIQ while significantly higher than 2D U‐Net on PROOF (0.78±0.07 vs. 0.77±0.07, P=0.10, and 0.73±0.07). The MPUnet performed significantly better on N=22 KL grade 3 CCBR scans with 0.78±0.06 vs. 0.75±0.08 for KIQ and 0.76±0.06 for 2D U‐Net.
Data Conclusion
The MPUnet matched or exceeded the performance of state‐of‐the‐art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy‐to‐use.
Level of Evidence
3
Technical Efficacy
Stage 2 |
|---|---|
| AbstractList | Segmentation of medical image volumes is a time-consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings.
To evaluate the performance of the open-source Multi-Planar U-Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state-of-the-art two-dimensional (2D) U-Net architecture on three clinical cohorts without extensive adaptation of the algorithms.
Retrospective cohort study.
A total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0-4).
0.18 T, 1.0 T/1.5 T, and 3 T sagittal three-dimensional fast-spin echo T1w and dual-echo steady-state sequences.
All models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades.
Segmentation performance differences as measured by Dice coefficients were tested with paired, two-sided Wilcoxon signed-rank statistics with significance threshold α = 0.05.
The MPUnet performed superior or equal to KIQ and 2D U-Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U-Net on CCBR (
vs.
and
), significantly higher than KIQ and U-Net OAI (
vs.
and
, and not significantly different from KIQ while significantly higher than 2D U-Net on PROOF (
vs.
,
, and
. The MPUnet performed significantly better on
KL grade 3 CCBR scans with
vs.
for KIQ and
for 2D U-Net.
The MPUnet matched or exceeded the performance of state-of-the-art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy-to-use.
3 TECHNICAL EFFICACY: Stage 2. BackgroundSegmentation of medical image volumes is a time‐consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings.PurposeTo evaluate the performance of the open‐source Multi‐Planar U‐Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state‐of‐the‐art two‐dimensional (2D) U‐Net architecture on three clinical cohorts without extensive adaptation of the algorithms.Study TypeRetrospective cohort study.SubjectsA total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0–4).Field Strength/Sequence0.18 T, 1.0 T/1.5 T, and 3 T sagittal three‐dimensional fast‐spin echo T1w and dual‐echo steady‐state sequences.AssessmentAll models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades.Statistical TestsSegmentation performance differences as measured by Dice coefficients were tested with paired, two‐sided Wilcoxon signed‐rank statistics with significance threshold α = 0.05.ResultsThe MPUnet performed superior or equal to KIQ and 2D U‐Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U‐Net on CCBR (0.83±0.04 vs. 0.81±0.06 and 0.82±0.05), significantly higher than KIQ and U‐Net OAI (0.86±0.03 vs. 0.84±0.04 and 0.85±0.03), and not significantly different from KIQ while significantly higher than 2D U‐Net on PROOF (0.78±0.07 vs. 0.77±0.07, P=0.10, and 0.73±0.07). The MPUnet performed significantly better on N=22 KL grade 3 CCBR scans with 0.78±0.06 vs. 0.75±0.08 for KIQ and 0.76±0.06 for 2D U‐Net.Data ConclusionThe MPUnet matched or exceeded the performance of state‐of‐the‐art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy‐to‐use.Level of Evidence3Technical EfficacyStage 2 Segmentation of medical image volumes is a time-consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings.BACKGROUNDSegmentation of medical image volumes is a time-consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings.To evaluate the performance of the open-source Multi-Planar U-Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state-of-the-art two-dimensional (2D) U-Net architecture on three clinical cohorts without extensive adaptation of the algorithms.PURPOSETo evaluate the performance of the open-source Multi-Planar U-Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state-of-the-art two-dimensional (2D) U-Net architecture on three clinical cohorts without extensive adaptation of the algorithms.Retrospective cohort study.STUDY TYPERetrospective cohort study.A total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0-4).SUBJECTSA total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0-4).0.18 T, 1.0 T/1.5 T, and 3 T sagittal three-dimensional fast-spin echo T1w and dual-echo steady-state sequences.FIELD STRENGTH/SEQUENCE0.18 T, 1.0 T/1.5 T, and 3 T sagittal three-dimensional fast-spin echo T1w and dual-echo steady-state sequences.All models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades.ASSESSMENTAll models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades.Segmentation performance differences as measured by Dice coefficients were tested with paired, two-sided Wilcoxon signed-rank statistics with significance threshold α = 0.05.STATISTICAL TESTSSegmentation performance differences as measured by Dice coefficients were tested with paired, two-sided Wilcoxon signed-rank statistics with significance threshold α = 0.05.The MPUnet performed superior or equal to KIQ and 2D U-Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U-Net on CCBR ( 0.83±0.04 vs. 0.81±0.06 and 0.82±0.05 ), significantly higher than KIQ and U-Net OAI ( 0.86±0.03 vs. 0.84±0.04 and 0.85±0.03) , and not significantly different from KIQ while significantly higher than 2D U-Net on PROOF ( 0.78±0.07 vs. 0.77±0.07 , P=0.10 , and 0.73±0.07) . The MPUnet performed significantly better on N=22 KL grade 3 CCBR scans with 0.78±0.06 vs. 0.75±0.08 for KIQ and 0.76±0.06 for 2D U-Net.RESULTSThe MPUnet performed superior or equal to KIQ and 2D U-Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U-Net on CCBR ( 0.83±0.04 vs. 0.81±0.06 and 0.82±0.05 ), significantly higher than KIQ and U-Net OAI ( 0.86±0.03 vs. 0.84±0.04 and 0.85±0.03) , and not significantly different from KIQ while significantly higher than 2D U-Net on PROOF ( 0.78±0.07 vs. 0.77±0.07 , P=0.10 , and 0.73±0.07) . The MPUnet performed significantly better on N=22 KL grade 3 CCBR scans with 0.78±0.06 vs. 0.75±0.08 for KIQ and 0.76±0.06 for 2D U-Net.The MPUnet matched or exceeded the performance of state-of-the-art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy-to-use.DATA CONCLUSIONThe MPUnet matched or exceeded the performance of state-of-the-art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy-to-use.3 TECHNICAL EFFICACY: Stage 2.LEVEL OF EVIDENCE3 TECHNICAL EFFICACY: Stage 2. Background Segmentation of medical image volumes is a time‐consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings. Purpose To evaluate the performance of the open‐source Multi‐Planar U‐Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state‐of‐the‐art two‐dimensional (2D) U‐Net architecture on three clinical cohorts without extensive adaptation of the algorithms. Study Type Retrospective cohort study. Subjects A total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0–4). Field Strength/Sequence 0.18 T, 1.0 T/1.5 T, and 3 T sagittal three‐dimensional fast‐spin echo T1w and dual‐echo steady‐state sequences. Assessment All models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades. Statistical Tests Segmentation performance differences as measured by Dice coefficients were tested with paired, two‐sided Wilcoxon signed‐rank statistics with significance threshold α = 0.05. Results The MPUnet performed superior or equal to KIQ and 2D U‐Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U‐Net on CCBR (0.83±0.04 vs. 0.81±0.06 and 0.82±0.05), significantly higher than KIQ and U‐Net OAI (0.86±0.03 vs. 0.84±0.04 and 0.85±0.03), and not significantly different from KIQ while significantly higher than 2D U‐Net on PROOF (0.78±0.07 vs. 0.77±0.07, P=0.10, and 0.73±0.07). The MPUnet performed significantly better on N=22 KL grade 3 CCBR scans with 0.78±0.06 vs. 0.75±0.08 for KIQ and 0.76±0.06 for 2D U‐Net. Data Conclusion The MPUnet matched or exceeded the performance of state‐of‐the‐art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy‐to‐use. Level of Evidence 3 Technical Efficacy Stage 2 |
| Author | Dam, Erik B. Perslev, Mathias Pai, Akshay Runhaar, Jos Igel, Christian |
| Author_xml | – sequence: 1 givenname: Mathias orcidid: 0000-0002-0358-4692 surname: Perslev fullname: Perslev, Mathias email: map@di.ku.dk organization: University of Copenhagen – sequence: 2 givenname: Akshay surname: Pai fullname: Pai, Akshay organization: Cerebriu A/S – sequence: 3 givenname: Jos surname: Runhaar fullname: Runhaar, Jos organization: Erasmus MC, Rotterdam University – sequence: 4 givenname: Christian orcidid: 0000-0003-2868-0856 surname: Igel fullname: Igel, Christian organization: University of Copenhagen – sequence: 5 givenname: Erik B. orcidid: 0000-0002-8888-2524 surname: Dam fullname: Dam, Erik B. organization: Cerebriu A/S |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/34918423$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9kctKxTAQhoMo3jc-gBTciFDNpUnTpRy8X_GCy5KmczSHttEkRdz5CD6jT2KOVRcirmYYvn-Y-f8lNNvZDhBaI3ibYEx3Jq0z2zQvcjmDFgmnNKVcitnYY85SInG-gJa8n2CMiyLj82iBZQWRGWWL6GTkrPfvr28j-2BdSHb7YFsVjE5OOoDk7OoouYb7FroQh7ZL7kx4SM76JpiouWxUp1xyG9tzCH4FzY1V42H1qy6j2_29m9FhenpxcDTaPU01K5hMgWcAmlMpqkxrUlTA6hoUpdm4xgKo4nmFZS0IZbmqFBWcaS1qKnGdCyCELaPNYe-js089-FC2xmto4jVge19SQYgQ8Vke0Y1f6MT2rovXRUpQwiTlWaTWv6i-aqEuH51plXspv22KwNYA6KlbDsY_CMHlNINymkH5mUGE8S9Ym8G94JRp_paQQfJsGnj5Z3l5HBMZNB-w9Jp2 |
| CitedBy_id | crossref_primary_10_1371_journal_pone_0316611 crossref_primary_10_3390_bioengineering10020137 crossref_primary_10_1002_ima_22836 crossref_primary_10_3390_bioengineering10020207 crossref_primary_10_1002_jmri_27990 crossref_primary_10_3390_bioengineering10060726 crossref_primary_10_1016_j_ostima_2023_100161 crossref_primary_10_1002_jmri_28513 crossref_primary_10_1109_JBHI_2023_3344392 |
| Cites_doi | 10.1007/s10334-020-00889-7 10.1148/radiol.2018172322 10.1109/ICDAR.2003.1227801 10.1002/mrm.27229 10.1109/ISBI.2018.8363866 10.1016/j.media.2017.07.005 10.1016/j.joca.2007.01.013 10.1016/j.joca.2018.02.907 10.1007/978-3-030-32245-8_4 10.1109/ICCVW.2019.00057 10.1016/j.media.2011.01.007 10.1093/rheumatology/keaa522 10.1007/s10462-020-09924-4 10.1109/TPAMI.2016.2572683 10.1002/mrm.26841 10.1109/TMI.2017.2781541 10.1016/j.amjmed.2015.03.006 10.1007/978-3-642-40763-5_31 10.1148/ryai.2021200078 10.1117/1.JMI.2.2.024001 10.1007/978-3-319-24574-4_28 10.2307/1932409 10.1038/s41746-021-00440-5 10.1109/ISBI.2018.8363705 10.1038/nature14539 10.1002/mrm.27130 10.1136/ard.16.4.494 10.1146/annurev-bioeng-071516-044442 10.1016/j.media.2018.11.009 |
| ContentType | Journal Article |
| Copyright | 2021 The Authors. published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine. 2021 The Authors. Journal of Magnetic Resonance Imaging published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine. 2021. This article is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
| Copyright_xml | – notice: 2021 The Authors. published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine. – notice: 2021 The Authors. Journal of Magnetic Resonance Imaging published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine. – notice: 2021. This article is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
| DBID | 24P AAYXX CITATION CGR CUY CVF ECM EIF NPM 7QO 7TK 8FD FR3 K9. P64 7X8 |
| DOI | 10.1002/jmri.27978 |
| DatabaseName | Wiley Online Library Open Access (WRLC) CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed Biotechnology Research Abstracts Neurosciences Abstracts Technology Research Database Engineering Research Database ProQuest Health & Medical Complete (Alumni) Biotechnology and BioEngineering Abstracts MEDLINE - Academic |
| DatabaseTitle | CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) ProQuest Health & Medical Complete (Alumni) Engineering Research Database Biotechnology Research Abstracts Technology Research Database Neurosciences Abstracts Biotechnology and BioEngineering Abstracts MEDLINE - Academic |
| DatabaseTitleList | MEDLINE ProQuest Health & Medical Complete (Alumni) MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: 24P name: Wiley Online Library Open Access url: https://authorservices.wiley.com/open-science/open-access/browse-journals.html sourceTypes: Publisher – sequence: 2 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: EIF name: MEDLINE url: https://proxy.k.utb.cz/login?url=https://www.webofscience.com/wos/medline/basic-search sourceTypes: Index Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine |
| EISSN | 1522-2586 |
| EndPage | 1663 |
| ExternalDocumentID | 34918423 10_1002_jmri_27978 JMRI27978 |
| Genre | article Research Support, Non-U.S. Gov't Journal Article Research Support, N.I.H., Extramural |
| GrantInformation_xml | – fundername: Independent Research Fund Denmark funderid: 9131‐00099B – fundername: Danish Industry Foundation – fundername: NIAMS NIH HHS grantid: N01 AR022262 – fundername: NHLBI NIH HHS grantid: N01 AR022258 – fundername: NIAMS NIH HHS grantid: N01 AR022261 – fundername: NIAMS NIH HHS grantid: N01 AR022260 – fundername: NIAMS NIH HHS grantid: N01 AR022259 |
| GroupedDBID | --- -DZ .3N .GA .GJ .Y3 05W 0R~ 10A 1L6 1OB 1OC 1ZS 24P 31~ 33P 3O- 3SF 3WU 4.4 4ZD 50Y 50Z 51W 51X 52M 52N 52O 52P 52R 52S 52T 52U 52V 52W 52X 53G 5GY 5RE 5VS 66C 702 7PT 8-0 8-1 8-3 8-4 8-5 8UM 930 A01 A03 AAESR AAEVG AAHHS AAHQN AAIPD AAMNL AANHP AANLZ AAONW AASGY AAWTL AAXRX AAYCA AAZKR ABCQN ABCUV ABEML ABIJN ABJNI ABLJU ABOCM ABPVW ABQWH ABXGK ACAHQ ACBWZ ACCFJ ACCZN ACGFO ACGFS ACGOF ACIWK ACMXC ACPOU ACPRK ACRPL ACSCC ACXBN ACXQS ACYXJ ADBBV ADBTR ADEOM ADIZJ ADKYN ADMGS ADNMO ADOZA ADXAS ADZMN AEEZP AEGXH AEIGN AEIMD AENEX AEQDE AEUQT AEUYR AFBPY AFFPM AFGKR AFPWT AFRAH AFWVQ AFZJQ AHBTC AHMBA AIACR AIAGR AITYG AIURR AIWBW AJBDE ALAGY ALMA_UNASSIGNED_HOLDINGS ALUQN ALVPJ AMBMR AMYDB ASPBG ATUGU AVWKF AZBYB AZFZN AZVAB BAFTC BDRZF BFHJK BHBCM BMXJE BROTX BRXPI BY8 C45 CS3 D-6 D-7 D-E D-F DCZOG DPXWK DR2 DRFUL DRMAN DRSTM DU5 EBD EBS EJD EMOBN F00 F01 F04 F5P FEDTE FUBAC G-S G.N GNP GODZA H.X HBH HDBZQ HF~ HGLYW HHY HHZ HVGLF HZ~ IX1 J0M JPC KBYEO KQQ LATKE LAW LC2 LC3 LEEKS LH4 LITHE LOXES LP6 LP7 LUTES LW6 LYRES M65 MEWTI MK4 MRFUL MRMAN MRSTM MSFUL MSMAN MSSTM MXFUL MXMAN MXSTM N04 N05 N9A NF~ NNB O66 O9- OIG OVD P2P P2W P2X P2Z P4B P4D PALCI PQQKQ Q.N Q11 QB0 QRW R.K RGB RIWAO RJQFR ROL RWI RX1 RYL SAMSI SUPJJ SV3 TEORI TWZ UB1 V2E V8K V9Y W8V W99 WBKPD WHWMO WIB WIH WIJ WIK WIN WJL WOHZO WQJ WRC WUP WVDHM WXI WXSBR XG1 XV2 ZXP ZZTAW ~IA ~WT AAYXX AEYWJ AGHNM AGQPQ AGYGG CITATION AAMMB AEFGJ AGXDD AIDQK AIDYY CGR CUY CVF ECM EIF NPM 7QO 7TK 8FD FR3 K9. P64 7X8 |
| ID | FETCH-LOGICAL-c3938-e54eec5286b4cc19be3ddea224fd06e2a57b08d61237aba2653cc6d280d76e113 |
| IEDL.DBID | DR2 |
| ISSN | 1053-1807 1522-2586 |
| IngestDate | Fri Jul 11 16:26:33 EDT 2025 Fri Jul 25 12:21:13 EDT 2025 Mon Jul 21 05:45:35 EDT 2025 Tue Jul 01 03:56:52 EDT 2025 Thu Apr 24 22:59:04 EDT 2025 Wed Jan 22 16:24:11 EST 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 6 |
| Keywords | deep learning magnetic resonance imaging open-source software knee segmentation |
| Language | English |
| License | Attribution-NonCommercial-NoDerivs 2021 The Authors. Journal of Magnetic Resonance Imaging published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c3938-e54eec5286b4cc19be3ddea224fd06e2a57b08d61237aba2653cc6d280d76e113 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0002-8888-2524 0000-0002-0358-4692 0000-0003-2868-0856 |
| OpenAccessLink | https://proxy.k.utb.cz/login?url=https://onlinelibrary.wiley.com/doi/abs/10.1002%2Fjmri.27978 |
| PMID | 34918423 |
| PQID | 2662138254 |
| PQPubID | 1006400 |
| PageCount | 14 |
| ParticipantIDs | proquest_miscellaneous_2611660095 proquest_journals_2662138254 pubmed_primary_34918423 crossref_primary_10_1002_jmri_27978 crossref_citationtrail_10_1002_jmri_27978 wiley_primary_10_1002_jmri_27978_JMRI27978 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 2000 |
| PublicationDate | June 2022 2022-06-00 20220601 |
| PublicationDateYYYYMMDD | 2022-06-01 |
| PublicationDate_xml | – month: 06 year: 2022 text: June 2022 |
| PublicationDecade | 2020 |
| PublicationPlace | Hoboken, USA |
| PublicationPlace_xml | – name: Hoboken, USA – name: United States – name: Nashville |
| PublicationSubtitle | JMRI |
| PublicationTitle | Journal of magnetic resonance imaging |
| PublicationTitleAlternate | J Magn Reson Imaging |
| PublicationYear | 2022 |
| Publisher | John Wiley & Sons, Inc Wiley Subscription Services, Inc |
| Publisher_xml | – name: John Wiley & Sons, Inc – name: Wiley Subscription Services, Inc |
| References | 2015; 2 2017; 42 2021; 4 2021; 3 2019; 52 2015; 521 2020; 60 2018; 288 2018; 80 2011; 15 2015; 128 2020; 54 1957; 16 2018; 26 2007; 15 2010; 6361 2021; 34 1948; 5 2017; 39 1945; 26 2003; 2 2019 2018 2017; 19 2015 2013 2018; 37 2018; 79 e_1_2_10_23_1 e_1_2_10_24_1 e_1_2_10_21_1 e_1_2_10_22_1 e_1_2_10_20_1 Ioffe S (e_1_2_10_26_1) 2015 e_1_2_10_2_1 e_1_2_10_4_1 e_1_2_10_18_1 e_1_2_10_3_1 e_1_2_10_19_1 e_1_2_10_6_1 e_1_2_10_16_1 e_1_2_10_5_1 Seim H (e_1_2_10_7_1) 2010 e_1_2_10_17_1 e_1_2_10_8_1 e_1_2_10_14_1 e_1_2_10_15_1 e_1_2_10_12_1 e_1_2_10_35_1 e_1_2_10_9_1 e_1_2_10_13_1 e_1_2_10_34_1 e_1_2_10_10_1 e_1_2_10_33_1 e_1_2_10_11_1 e_1_2_10_32_1 e_1_2_10_30_1 Sørensen T (e_1_2_10_31_1) 1948; 5 e_1_2_10_29_1 e_1_2_10_27_1 e_1_2_10_28_1 e_1_2_10_25_1 34757652 - J Magn Reson Imaging. 2022 Jun;55(6):1664-1665. doi: 10.1002/jmri.27990. |
| References_xml | – volume: 288 start-page: 177 year: 2018 end-page: 185 article-title: Use of 2D U‐net convolutional neural networks for automated cartilage and meniscus segmentation of knee MR imaging data to determine relaxometry and morphometry publication-title: Radiology – start-page: 234 year: 2015 end-page: 241 – volume: 26 start-page: 680 year: 2018 end-page: 688 article-title: Knee menisci segmentation using convolutional neural networks: Data from the Osteoarthritis Initiative publication-title: Osteoarthritis Cartilage – volume: 80 start-page: 1219 year: 2018 end-page: 1232 article-title: Cartilage cavity—An MRI marker of cartilage lesions in knee OA with data from CCBR, OAI, and PROOF publication-title: Magn Reson Med – volume: 39 start-page: 640 year: 2017 end-page: 651 article-title: Fully convolutional networks for semantic segmentation publication-title: IEEE Trans Pattern Anal Mach Intell – volume: 3 year: 2021 article-title: The international workshop on osteoarthritis imaging knee MRI segmentation challenge: A multi‐institute evaluation and analysis framework on a standardized dataset publication-title: Radiol Artif Intell – volume: 26 start-page: 297 year: 1945 end-page: 302 article-title: Measures of the amount of ecologic association between species publication-title: Ecology – volume: 19 start-page: 221 year: 2017 end-page: 248 article-title: Deep learning in medical image analysis publication-title: Annu Rev Biomed Eng – start-page: 851 year: 2018 end-page: 854 – volume: 16 start-page: 494 year: 1957 end-page: 502 article-title: Radiological assessment of osteo‐arthrosis publication-title: Ann Rheum Dis – start-page: 1538 year: 2018 end-page: 1541 – volume: 54 start-page: 2445 year: 2020 end-page: 2494 article-title: From classical to deep learning: Review on cartilage and bone segmentation techniques in knee osteoarthritis research publication-title: Artif Intell Rev – volume: 2 year: 2015 article-title: Automatic segmentation of high‐ and low‐field knee MRIs using knee image quantification with data from the osteoarthritis initiative publication-title: J Med Imaging – volume: 79 start-page: 2379 year: 2018 end-page: 2391 article-title: Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging publication-title: Magn Reson Med – volume: 128 start-page: 888 year: 2015 end-page: 895 article-title: Prevention of knee osteoarthritis in overweight females: The first preventive randomized controlled trial in osteoarthritis publication-title: Am J Med – volume: 521 start-page: 436 year: 2015 end-page: 444 article-title: Deep learning publication-title: Nature – start-page: 246 year: 2013 end-page: 253 – volume: 15 start-page: 438 year: 2011 end-page: 448 article-title: Automatic graph‐cut based segmentation of bones from knee magnetic resonance images for osteoarthritis research publication-title: Med Image Anal – volume: 2 start-page: 958 year: 2003 end-page: 963 – volume: 15 start-page: 808 year: 2007 end-page: 818 article-title: Automatic morphometric cartilage quantification in the medial tibial plateau from MRI for osteoarthritis grading publication-title: Osteoarthritis Cartilage – volume: 42 start-page: 60 year: 2017 end-page: 88 article-title: A survey on deep learning in medical image analysis publication-title: Med Image Anal – volume: 80 start-page: 2759 year: 2018 end-page: 2770 article-title: Deep convolutional neural network for segmentation of knee joint anatomy publication-title: Magn Reson Med – volume: 4 start-page: 72 issue: 72 year: 2021 article-title: U‐Sleep: Resilient high‐frequency sleep staging publication-title: npj Digit Med – volume: 52 start-page: 109 year: 2019 end-page: 118 article-title: Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the osteoarthritis initiative publication-title: Med Image Anal – volume: 37 start-page: 1103 year: 2018 end-page: 1113 article-title: Learning‐based cost functions for 3‐D and 4‐D multi‐surface multi‐object segmentation of knee MRI: Data from the osteoarthritis initiative publication-title: IEEE Trans Med Imaging – volume: 60 start-page: 1392 year: 2020 end-page: 1399 article-title: Association between meniscal volume and development of knee osteoarthritis publication-title: Rheumatology – start-page: 448 year: 2015 end-page: 456 – volume: 5 start-page: 1 year: 1948 end-page: 34 article-title: A method of establishing groups of equal amplitude in plant sociology based on similarity publication-title: Kong Dansk Vidensk Selsk Biol Skr – volume: 34 start-page: 337 year: 2021 end-page: 354 article-title: Accuracy and longitudinal reproducibility of quantitative femorotibial cartilage measures derived from automated U‐Net‐based segmentation of two different MRI contrasts: Data from the osteoarthritis initiative healthy reference cohort publication-title: Magn Reson Mater Phys Biol Med – start-page: 450 year: 2019 end-page: 459 – start-page: 30 year: 2019 end-page: 38 – volume: 6361 start-page: 215 year: 2010 end-page: 223 – ident: e_1_2_10_13_1 doi: 10.1007/s10334-020-00889-7 – ident: e_1_2_10_12_1 doi: 10.1148/radiol.2018172322 – ident: e_1_2_10_27_1 doi: 10.1109/ICDAR.2003.1227801 – ident: e_1_2_10_15_1 doi: 10.1002/mrm.27229 – ident: e_1_2_10_19_1 doi: 10.1109/ISBI.2018.8363866 – ident: e_1_2_10_33_1 – ident: e_1_2_10_3_1 doi: 10.1016/j.media.2017.07.005 – start-page: 215 volume-title: Proceedings ‐ Medical Image Analysis for the Clinic: a Grand Challenge in Conjunction with MICCAI year: 2010 ident: e_1_2_10_7_1 – ident: e_1_2_10_29_1 doi: 10.1016/j.joca.2007.01.013 – ident: e_1_2_10_16_1 doi: 10.1016/j.joca.2018.02.907 – ident: e_1_2_10_24_1 doi: 10.1007/978-3-030-32245-8_4 – ident: e_1_2_10_14_1 doi: 10.1109/ICCVW.2019.00057 – volume: 5 start-page: 1 year: 1948 ident: e_1_2_10_31_1 article-title: A method of establishing groups of equal amplitude in plant sociology based on similarity publication-title: Kong Dansk Vidensk Selsk Biol Skr – ident: e_1_2_10_8_1 doi: 10.1016/j.media.2011.01.007 – start-page: 448 volume-title: Proceedings ‐ 32nd International Conference on Machine Learning (ICML) year: 2015 ident: e_1_2_10_26_1 – ident: e_1_2_10_34_1 doi: 10.1093/rheumatology/keaa522 – ident: e_1_2_10_4_1 doi: 10.1007/s10462-020-09924-4 – ident: e_1_2_10_11_1 doi: 10.1109/TPAMI.2016.2572683 – ident: e_1_2_10_21_1 doi: 10.1002/mrm.26841 – ident: e_1_2_10_6_1 doi: 10.1109/TMI.2017.2781541 – ident: e_1_2_10_30_1 doi: 10.1016/j.amjmed.2015.03.006 – ident: e_1_2_10_20_1 doi: 10.1007/978-3-642-40763-5_31 – ident: e_1_2_10_22_1 doi: 10.1148/ryai.2021200078 – ident: e_1_2_10_9_1 doi: 10.1117/1.JMI.2.2.024001 – ident: e_1_2_10_23_1 doi: 10.1007/978-3-319-24574-4_28 – ident: e_1_2_10_32_1 doi: 10.2307/1932409 – ident: e_1_2_10_35_1 doi: 10.1038/s41746-021-00440-5 – ident: e_1_2_10_18_1 doi: 10.1109/ISBI.2018.8363705 – ident: e_1_2_10_25_1 – ident: e_1_2_10_10_1 doi: 10.1038/nature14539 – ident: e_1_2_10_5_1 doi: 10.1002/mrm.27130 – ident: e_1_2_10_28_1 doi: 10.1136/ard.16.4.494 – ident: e_1_2_10_2_1 doi: 10.1146/annurev-bioeng-071516-044442 – ident: e_1_2_10_17_1 doi: 10.1016/j.media.2018.11.009 – reference: 34757652 - J Magn Reson Imaging. 2022 Jun;55(6):1664-1665. doi: 10.1002/jmri.27990. |
| SSID | ssj0009945 |
| Score | 2.4278436 |
| Snippet | Background
Segmentation of medical image volumes is a time‐consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is... Segmentation of medical image volumes is a time-consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear... BackgroundSegmentation of medical image volumes is a time‐consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is... |
| SourceID | proquest pubmed crossref wiley |
| SourceType | Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 1650 |
| SubjectTerms | Aged Algorithms Arthritis Biomedical materials Body weight Cohort Studies deep learning Demography Female Females Field strength Humans Image processing Image segmentation Knee Knee - diagnostic imaging Knee - pathology Knee Joint - diagnostic imaging Knee Joint - pathology knee segmentation Magnetic resonance imaging Magnetic Resonance Imaging - methods Male Medical imaging Middle Aged open‐source software Osteoarthritis Osteoarthritis, Knee - diagnostic imaging Osteoarthritis, Knee - pathology Overweight Patients Performance evaluation Retrospective Studies Statistical analysis Statistical tests Tuning |
| Title | Cross‐Cohort Automatic Knee MRI Segmentation With Multi‐Planar U‐Nets |
| URI | https://onlinelibrary.wiley.com/doi/abs/10.1002%2Fjmri.27978 https://www.ncbi.nlm.nih.gov/pubmed/34918423 https://www.proquest.com/docview/2662138254 https://www.proquest.com/docview/2611660095 |
| Volume | 55 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3NTtwwEB4Bh6oXWtpS0gJyRS8gZYl_YidSL2hbREGLEO0KLlUUO16WtmSr3eylpz5Cn5EnYexks6IgpHKLlLHseGbsb-zMNwDvY5mogYkHIUbLSSi0pGEiTBQWWgg1QIBMqctG7h3Lg744PI_PF-DDLBem5odoD9ycZ_j12jl4rie7c9LQ71fjyw5TGAXhAky5dMT5H0_n3FFp6isUI37gIU0i1XKTst1509u70R2IeRux-i1n_xl8mw22_tPkR2da6Y75_Q-P42O_5jksN1iU7NXGswILtnwBT3rNbftLOOq6QV__-dsdDRGlk71pNfIMr-SotJb0Tj-TL_biqsleKsnZZTUkPqMX27hqSPmY9PHx2FaTV9Df__S1exA2xRdCw53abCysNTFLpBbG0FRbjithjjv-oIikZXmsdJQUjr1F5TpnMubGyIIlUaGkpZSvwlI5Ku0aEG1FnqZoEhZjP8SbmjnaOK4Sig1VVASwPVNCZhpmclcg42dWcyqzzM1O5mcngK1W9lfNx3Gv1PpMl1njk5MMoQhzjIuxCOBd-xq9yV2R5KUdTZ0MpVI63BnA69oG2m64SDEcZjyAHa_JB_rPDlEF_unN_wi_hafMZVf4Q551WKrGU7uBmKfSm7DIxMmmt_AbCVf_CA |
| linkProvider | Wiley-Blackwell |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1fb9MwED-xIW28IP6TUcCIvYCUNXYcO3mcKqZuXSu0rWJvUey4bNOWTm36zkfgM_JJuHOyVBMIiTdLuYulO__5ne37HcBuolI9s8ksxGg5DaVRPEyljcLSSKlnCJA5p2zk8UQNp_LoPDlv3-ZQLkzDD9EduNHM8Os1TXA6kO6vWUOvbhaXe0JjGLQBDyUCc3rRJ-TXNedu5msUI4KIQ55GumMnFf217v396A-QeR-z-k3n4Ak8btEi22_c-xQeuOoZbI3b-_DnMBrQT3_9-DmYXyCOZvureu45WNmoco6NTw7Zqft-0-YXVezbZX3BfM4t6lC9omLBpticuHr5AqYHX84Gw7AtjxDamAzrEumcTUSqjLSWZ8bFuFYVuCfPykg5USTaRGlJ_Cq6MAXaJ7ZWlSKNSq0c5_FL2KzmlXsNzDhZZBk6zWF0hojQCCJ2i3XKUVFHZQCf7oyU25Y7nEpYXOcN67HIyaC5N2gAHzvZ24Yx469SvTtb5-2sWeYIFgRxIiYygA_dZxzvdIlRVG6-IhnOlSJkGMCrxkddN7HMMGAVcQCfvdP-0X9-hC7wrZ3_EX4P28Oz8XF-fDgZvYFHgnIh_JFMDzbrxcq9RYRSm3d-HP4GnrLhGA |
| linkToPdf | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3NbtQwEB6VIlVcWijQpi1gBBeQso0dx04kLtXCqu2yK1RY0UsVxY7T_2y1zV566iPwjDwJYyebVQFVglukjGXHM2N_Y2e-AXgbiVgWOip8jJZjnytB_ZjrwM8V57JAgEypzUYeDMXuiO8fRocL8GGWC1PzQ7QHbtYz3HptHfwqL7bnpKFnl5PTDpMYBT2Ah1wglLCQ6GBOHpUkrkQxAojQp3EgW3JStj1ve3c7-gNj3oWsbs_prcDRbLT1rybnnWmlOvrmNyLH__2cx7DcgFGyU1vPE1gw5SosDZrr9qfQ79pB_7z90R2fIEwnO9Nq7CheSb80hgwO9shXc3zZpC-V5PtpdUJcSi-2seWQsgkZ4ePQVNfPYNT79K276zfVF3wdWr2ZiBujIxYLxbWmiTIhLoUZbvlFHgjDskiqIM4tfYvMVMZEFGotchYHuRSG0vA5LJbj0qwDUYZnSYI2YTD4Q8CpmOWNC2VMsaEMcg_ezZSQ6oaa3FbIuEhrUmWW2tlJ3ex48KaVvaoJOf4qtTXTZdo45XWKWIRZysWIe_C6fY3uZO9IstKMp1aGUiEs8PRgrbaBtpuQJxgPs9CD906T9_Sf7qMK3NPGvwi_gqUvH3vp571hfxMeMZtp4Q58tmCxmkzNC8Q_lXrpzPwX8s8BDg |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Cross-Cohort+Automatic+Knee+MRI+Segmentation+With+Multi-Planar+U-Nets&rft.jtitle=Journal+of+magnetic+resonance+imaging&rft.au=Perslev%2C+Mathias&rft.au=Pai%2C+Akshay&rft.au=Runhaar%2C+Jos&rft.au=Igel%2C+Christian&rft.date=2022-06-01&rft.eissn=1522-2586&rft.volume=55&rft.issue=6&rft.spage=1650&rft_id=info:doi/10.1002%2Fjmri.27978&rft_id=info%3Apmid%2F34918423&rft.externalDocID=34918423 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1053-1807&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1053-1807&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1053-1807&client=summon |