Generalizability of Deep Learning Segmentation Algorithms for Automated Assessment of Cartilage Morphology and MRI Relaxometry
Background Deep learning (DL)‐based automatic segmentation models can expedite manual segmentation yet require resource‐intensive fine‐tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine‐tuning is not well characterized. Purpose Evaluate the gene...
Uloženo v:
| Vydáno v: | Journal of magnetic resonance imaging Ročník 57; číslo 4; s. 1029 - 1039 |
|---|---|
| Hlavní autoři: | , , , , , , , , , , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Hoboken, USA
John Wiley & Sons, Inc
01.04.2023
Wiley Subscription Services, Inc |
| Témata: | |
| ISSN: | 1053-1807, 1522-2586, 1522-2586 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | Background
Deep learning (DL)‐based automatic segmentation models can expedite manual segmentation yet require resource‐intensive fine‐tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine‐tuning is not well characterized.
Purpose
Evaluate the generalizability of DL‐based models by deploying pretrained models on independent datasets varying by MR scanner, acquisition parameters, and subject population.
Study Type
Retrospective based on prospectively acquired data.
Population
Overall test dataset: 59 subjects (26 females); Study 1: 5 healthy subjects (zero females), Study 2: 8 healthy subjects (eight females), Study 3: 10 subjects with osteoarthritis (eight females), Study 4: 36 subjects with various knee pathology (10 females).
Field Strength/Sequence
A 3‐T, quantitative double‐echo steady state (qDESS).
Assessment
Four annotators manually segmented knee cartilage. Each reader segmented one of four qDESS datasets in the test dataset. Two DL models, one trained on qDESS data and another on Osteoarthritis Initiative (OAI)‐DESS data, were assessed. Manual and automatic segmentations were compared by quantifying variations in segmentation accuracy, volume, and T2 relaxation times for superficial and deep cartilage.
Statistical Tests
Dice similarity coefficient (DSC) for segmentation accuracy. Lin's concordance correlation coefficient (CCC), Wilcoxon rank‐sum tests, root‐mean‐squared error‐coefficient‐of‐variation to quantify manual vs. automatic T2 and volume variations. Bland–Altman plots for manual vs. automatic T2 agreement. A P value < 0.05 was considered statistically significant.
Results
DSCs for the qDESS‐trained model, 0.79–0.93, were higher than those for the OAI‐DESS‐trained model, 0.59–0.79. T2 and volume CCCs for the qDESS‐trained model, 0.75–0.98 and 0.47–0.95, were higher than respective CCCs for the OAI‐DESS‐trained model, 0.35–0.90 and 0.13–0.84. Bland–Altman 95% limits of agreement for superficial and deep cartilage T2 were lower for the qDESS‐trained model, ±2.4 msec and ±4.0 msec, than the OAI‐DESS‐trained model, ±4.4 msec and ±5.2 msec.
Data Conclusion
The qDESS‐trained model may generalize well to independent qDESS datasets regardless of MR scanner, acquisition parameters, and subject population.
Evidence Level
1
Technical Efficacy
Stage 1 |
|---|---|
| AbstractList | Deep learning (DL)-based automatic segmentation models can expedite manual segmentation yet require resource-intensive fine-tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine-tuning is not well characterized.
Evaluate the generalizability of DL-based models by deploying pretrained models on independent datasets varying by MR scanner, acquisition parameters, and subject population.
Retrospective based on prospectively acquired data.
Overall test dataset: 59 subjects (26 females); Study 1: 5 healthy subjects (zero females), Study 2: 8 healthy subjects (eight females), Study 3: 10 subjects with osteoarthritis (eight females), Study 4: 36 subjects with various knee pathology (10 females).
A 3-T, quantitative double-echo steady state (qDESS).
Four annotators manually segmented knee cartilage. Each reader segmented one of four qDESS datasets in the test dataset. Two DL models, one trained on qDESS data and another on Osteoarthritis Initiative (OAI)-DESS data, were assessed. Manual and automatic segmentations were compared by quantifying variations in segmentation accuracy, volume, and T2 relaxation times for superficial and deep cartilage.
Dice similarity coefficient (DSC) for segmentation accuracy. Lin's concordance correlation coefficient (CCC), Wilcoxon rank-sum tests, root-mean-squared error-coefficient-of-variation to quantify manual vs. automatic T2 and volume variations. Bland-Altman plots for manual vs. automatic T2 agreement. A P value < 0.05 was considered statistically significant.
DSCs for the qDESS-trained model, 0.79-0.93, were higher than those for the OAI-DESS-trained model, 0.59-0.79. T2 and volume CCCs for the qDESS-trained model, 0.75-0.98 and 0.47-0.95, were higher than respective CCCs for the OAI-DESS-trained model, 0.35-0.90 and 0.13-0.84. Bland-Altman 95% limits of agreement for superficial and deep cartilage T2 were lower for the qDESS-trained model, ±2.4 msec and ±4.0 msec, than the OAI-DESS-trained model, ±4.4 msec and ±5.2 msec.
The qDESS-trained model may generalize well to independent qDESS datasets regardless of MR scanner, acquisition parameters, and subject population.
1 TECHNICAL EFFICACY: Stage 1. Deep learning (DL)-based automatic segmentation models can expedite manual segmentation yet require resource-intensive fine-tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine-tuning is not well characterized.BACKGROUNDDeep learning (DL)-based automatic segmentation models can expedite manual segmentation yet require resource-intensive fine-tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine-tuning is not well characterized.Evaluate the generalizability of DL-based models by deploying pretrained models on independent datasets varying by MR scanner, acquisition parameters, and subject population.PURPOSEEvaluate the generalizability of DL-based models by deploying pretrained models on independent datasets varying by MR scanner, acquisition parameters, and subject population.Retrospective based on prospectively acquired data.STUDY TYPERetrospective based on prospectively acquired data.Overall test dataset: 59 subjects (26 females); Study 1: 5 healthy subjects (zero females), Study 2: 8 healthy subjects (eight females), Study 3: 10 subjects with osteoarthritis (eight females), Study 4: 36 subjects with various knee pathology (10 females).POPULATIONOverall test dataset: 59 subjects (26 females); Study 1: 5 healthy subjects (zero females), Study 2: 8 healthy subjects (eight females), Study 3: 10 subjects with osteoarthritis (eight females), Study 4: 36 subjects with various knee pathology (10 females).A 3-T, quantitative double-echo steady state (qDESS).FIELD STRENGTH/SEQUENCEA 3-T, quantitative double-echo steady state (qDESS).Four annotators manually segmented knee cartilage. Each reader segmented one of four qDESS datasets in the test dataset. Two DL models, one trained on qDESS data and another on Osteoarthritis Initiative (OAI)-DESS data, were assessed. Manual and automatic segmentations were compared by quantifying variations in segmentation accuracy, volume, and T2 relaxation times for superficial and deep cartilage.ASSESSMENTFour annotators manually segmented knee cartilage. Each reader segmented one of four qDESS datasets in the test dataset. Two DL models, one trained on qDESS data and another on Osteoarthritis Initiative (OAI)-DESS data, were assessed. Manual and automatic segmentations were compared by quantifying variations in segmentation accuracy, volume, and T2 relaxation times for superficial and deep cartilage.Dice similarity coefficient (DSC) for segmentation accuracy. Lin's concordance correlation coefficient (CCC), Wilcoxon rank-sum tests, root-mean-squared error-coefficient-of-variation to quantify manual vs. automatic T2 and volume variations. Bland-Altman plots for manual vs. automatic T2 agreement. A P value < 0.05 was considered statistically significant.STATISTICAL TESTSDice similarity coefficient (DSC) for segmentation accuracy. Lin's concordance correlation coefficient (CCC), Wilcoxon rank-sum tests, root-mean-squared error-coefficient-of-variation to quantify manual vs. automatic T2 and volume variations. Bland-Altman plots for manual vs. automatic T2 agreement. A P value < 0.05 was considered statistically significant.DSCs for the qDESS-trained model, 0.79-0.93, were higher than those for the OAI-DESS-trained model, 0.59-0.79. T2 and volume CCCs for the qDESS-trained model, 0.75-0.98 and 0.47-0.95, were higher than respective CCCs for the OAI-DESS-trained model, 0.35-0.90 and 0.13-0.84. Bland-Altman 95% limits of agreement for superficial and deep cartilage T2 were lower for the qDESS-trained model, ±2.4 msec and ±4.0 msec, than the OAI-DESS-trained model, ±4.4 msec and ±5.2 msec.RESULTSDSCs for the qDESS-trained model, 0.79-0.93, were higher than those for the OAI-DESS-trained model, 0.59-0.79. T2 and volume CCCs for the qDESS-trained model, 0.75-0.98 and 0.47-0.95, were higher than respective CCCs for the OAI-DESS-trained model, 0.35-0.90 and 0.13-0.84. Bland-Altman 95% limits of agreement for superficial and deep cartilage T2 were lower for the qDESS-trained model, ±2.4 msec and ±4.0 msec, than the OAI-DESS-trained model, ±4.4 msec and ±5.2 msec.The qDESS-trained model may generalize well to independent qDESS datasets regardless of MR scanner, acquisition parameters, and subject population.DATA CONCLUSIONThe qDESS-trained model may generalize well to independent qDESS datasets regardless of MR scanner, acquisition parameters, and subject population.1 TECHNICAL EFFICACY: Stage 1.EVIDENCE LEVEL1 TECHNICAL EFFICACY: Stage 1. Background Deep learning (DL)‐based automatic segmentation models can expedite manual segmentation yet require resource‐intensive fine‐tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine‐tuning is not well characterized. Purpose Evaluate the generalizability of DL‐based models by deploying pretrained models on independent datasets varying by MR scanner, acquisition parameters, and subject population. Study Type Retrospective based on prospectively acquired data. Population Overall test dataset: 59 subjects (26 females); Study 1: 5 healthy subjects (zero females), Study 2: 8 healthy subjects (eight females), Study 3: 10 subjects with osteoarthritis (eight females), Study 4: 36 subjects with various knee pathology (10 females). Field Strength/Sequence A 3‐T, quantitative double‐echo steady state (qDESS). Assessment Four annotators manually segmented knee cartilage. Each reader segmented one of four qDESS datasets in the test dataset. Two DL models, one trained on qDESS data and another on Osteoarthritis Initiative (OAI)‐DESS data, were assessed. Manual and automatic segmentations were compared by quantifying variations in segmentation accuracy, volume, and T2 relaxation times for superficial and deep cartilage. Statistical Tests Dice similarity coefficient (DSC) for segmentation accuracy. Lin's concordance correlation coefficient (CCC), Wilcoxon rank‐sum tests, root‐mean‐squared error‐coefficient‐of‐variation to quantify manual vs. automatic T2 and volume variations. Bland–Altman plots for manual vs. automatic T2 agreement. A P value < 0.05 was considered statistically significant. Results DSCs for the qDESS‐trained model, 0.79–0.93, were higher than those for the OAI‐DESS‐trained model, 0.59–0.79. T2 and volume CCCs for the qDESS‐trained model, 0.75–0.98 and 0.47–0.95, were higher than respective CCCs for the OAI‐DESS‐trained model, 0.35–0.90 and 0.13–0.84. Bland–Altman 95% limits of agreement for superficial and deep cartilage T2 were lower for the qDESS‐trained model, ±2.4 msec and ±4.0 msec, than the OAI‐DESS‐trained model, ±4.4 msec and ±5.2 msec. Data Conclusion The qDESS‐trained model may generalize well to independent qDESS datasets regardless of MR scanner, acquisition parameters, and subject population. Evidence Level 1 Technical Efficacy Stage 1 BackgroundDeep learning (DL)‐based automatic segmentation models can expedite manual segmentation yet require resource‐intensive fine‐tuning before deployment on new datasets. The generalizability of DL methods to new datasets without fine‐tuning is not well characterized.PurposeEvaluate the generalizability of DL‐based models by deploying pretrained models on independent datasets varying by MR scanner, acquisition parameters, and subject population.Study TypeRetrospective based on prospectively acquired data.PopulationOverall test dataset: 59 subjects (26 females); Study 1: 5 healthy subjects (zero females), Study 2: 8 healthy subjects (eight females), Study 3: 10 subjects with osteoarthritis (eight females), Study 4: 36 subjects with various knee pathology (10 females).Field Strength/SequenceA 3‐T, quantitative double‐echo steady state (qDESS).AssessmentFour annotators manually segmented knee cartilage. Each reader segmented one of four qDESS datasets in the test dataset. Two DL models, one trained on qDESS data and another on Osteoarthritis Initiative (OAI)‐DESS data, were assessed. Manual and automatic segmentations were compared by quantifying variations in segmentation accuracy, volume, and T2 relaxation times for superficial and deep cartilage.Statistical TestsDice similarity coefficient (DSC) for segmentation accuracy. Lin's concordance correlation coefficient (CCC), Wilcoxon rank‐sum tests, root‐mean‐squared error‐coefficient‐of‐variation to quantify manual vs. automatic T2 and volume variations. Bland–Altman plots for manual vs. automatic T2 agreement. A P value < 0.05 was considered statistically significant.ResultsDSCs for the qDESS‐trained model, 0.79–0.93, were higher than those for the OAI‐DESS‐trained model, 0.59–0.79. T2 and volume CCCs for the qDESS‐trained model, 0.75–0.98 and 0.47–0.95, were higher than respective CCCs for the OAI‐DESS‐trained model, 0.35–0.90 and 0.13–0.84. Bland–Altman 95% limits of agreement for superficial and deep cartilage T2 were lower for the qDESS‐trained model, ±2.4 msec and ±4.0 msec, than the OAI‐DESS‐trained model, ±4.4 msec and ±5.2 msec.Data ConclusionThe qDESS‐trained model may generalize well to independent qDESS datasets regardless of MR scanner, acquisition parameters, and subject population.Evidence Level1Technical EfficacyStage 1 |
| Author | Desai, Arjun D. MacKay, James W. Gold, Garry E. Boutin, Robert D. Crowder, Hollis A. Black, Marianne S. Rubin, Elka B. Mazzoli, Valentina Schmidt, Andrew M. Kogan, Feliks Chaudhari, Akshay S. Lu, Quin Watkins, Lauren E. Hargreaves, Brian A. |
| AuthorAffiliation | 2. Electrical Engineering, Stanford University, Palo Alto, CA, United States 8. Biomedical Data Science, Stanford University, Palo Alto, CA, United States 4. Mechanical Engineering, Stanford University, Palo Alto, CA, United States 3. Bioengineering, Stanford University, Palo Alto, CA, United States 5. Philips Healthcare North America, Gainesville, FL, United States 1. Department of Radiology, Stanford University, Palo Alto, CA, United States 6. Department of Radiology, University of Cambridge, Cambridge, United Kingdom 7. Norwich Medical School, University of East Anglia, Norwich, United Kingdom |
| AuthorAffiliation_xml | – name: 3. Bioengineering, Stanford University, Palo Alto, CA, United States – name: 8. Biomedical Data Science, Stanford University, Palo Alto, CA, United States – name: 5. Philips Healthcare North America, Gainesville, FL, United States – name: 6. Department of Radiology, University of Cambridge, Cambridge, United Kingdom – name: 7. Norwich Medical School, University of East Anglia, Norwich, United Kingdom – name: 4. Mechanical Engineering, Stanford University, Palo Alto, CA, United States – name: 2. Electrical Engineering, Stanford University, Palo Alto, CA, United States – name: 1. Department of Radiology, Stanford University, Palo Alto, CA, United States |
| Author_xml | – sequence: 1 givenname: Andrew M. orcidid: 0000-0003-3826-6122 surname: Schmidt fullname: Schmidt, Andrew M. email: aschmid2@stanford.edu organization: Stanford University – sequence: 2 givenname: Arjun D. surname: Desai fullname: Desai, Arjun D. organization: Stanford University – sequence: 3 givenname: Lauren E. orcidid: 0000-0003-3592-4089 surname: Watkins fullname: Watkins, Lauren E. organization: Stanford University – sequence: 4 givenname: Hollis A. orcidid: 0000-0003-3663-5799 surname: Crowder fullname: Crowder, Hollis A. organization: Stanford University – sequence: 5 givenname: Marianne S. orcidid: 0000-0002-2564-9071 surname: Black fullname: Black, Marianne S. organization: Stanford University – sequence: 6 givenname: Valentina orcidid: 0000-0002-6700-8424 surname: Mazzoli fullname: Mazzoli, Valentina organization: Stanford University – sequence: 7 givenname: Elka B. orcidid: 0000-0003-0175-7657 surname: Rubin fullname: Rubin, Elka B. organization: Stanford University – sequence: 8 givenname: Quin orcidid: 0000-0001-9526-1614 surname: Lu fullname: Lu, Quin organization: Philips Healthcare North America – sequence: 9 givenname: James W. orcidid: 0000-0001-7558-3800 surname: MacKay fullname: MacKay, James W. organization: University of East Anglia – sequence: 10 givenname: Robert D. surname: Boutin fullname: Boutin, Robert D. organization: Stanford University – sequence: 11 givenname: Feliks surname: Kogan fullname: Kogan, Feliks organization: Stanford University – sequence: 12 givenname: Garry E. surname: Gold fullname: Gold, Garry E. organization: Stanford University – sequence: 13 givenname: Brian A. surname: Hargreaves fullname: Hargreaves, Brian A. organization: Stanford University – sequence: 14 givenname: Akshay S. orcidid: 0000-0002-3667-6796 surname: Chaudhari fullname: Chaudhari, Akshay S. organization: Stanford University |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/35852498$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9kc-L1DAcxYusuD_04h8gAS8idE3SpE0uwjDqujKLsOo5pO23nQxpMiatWg_-7WbsuuginhLI5z3eyzvNjpx3kGWPCT4nGNMXuyGYcyqKkt_LTginNKdclEfpjnmRE4Gr4-w0xh3GWErGH2THBRecMilOsh8X4CBoa77r2lgzzsh36BXAHm1AB2dcjz5AP4Ab9Wi8Qyvb-2DG7RBR5wNaTaMf9AgtWsUIMR7Ag8Nah9FY3QO68mG_9db3M9KuRVfXl-garP7mBxjD_DC732kb4dHNeZZ9evP64_ptvnl_cblebfKGMcHzTrSFqHDZSiIx17KoGSllXTKOOa8L0VTQQoNL1mlKZVXrjjEquoo3JddY6uIse7n47qd6gLZJMVNptQ9m0GFWXhv194szW9X7L0oKJpkgyeDZjUHwnyeIoxpMbMBa7cBPUdFSkkoIQWRCn95Bd34KLtVTtEr_jgtBD9STPxPdRvk9TQKeL0ATfIwBuluEYHXYXR12V792TzC-AzdmWSy1MfbfErJIvhoL83_M1bs02qL5CSBOwh4 |
| CitedBy_id | crossref_primary_10_1016_j_acra_2025_04_072 crossref_primary_10_1007_s00117_023_01252_2 crossref_primary_10_1016_j_imed_2024_12_002 crossref_primary_10_1002_jmri_29710 crossref_primary_10_1002_jmri_29689 crossref_primary_10_1002_jmri_29647 crossref_primary_10_1016_j_joca_2023_10_005 crossref_primary_10_1186_s13018_024_04680_5 crossref_primary_10_1016_j_semarthrit_2025_152652 crossref_primary_10_1002_jmri_28393 crossref_primary_10_1007_s00256_024_04734_z crossref_primary_10_1007_s00256_025_04922_5 crossref_primary_10_1007_s10334_024_01173_8 crossref_primary_10_1002_jmri_29431 crossref_primary_10_1016_j_csm_2024_08_004 crossref_primary_10_1016_j_jocmr_2024_101082 crossref_primary_10_1109_ACCESS_2025_3569158 |
| Cites_doi | 10.1016/j.joca.2009.06.003 10.1016/j.mri.2016.12.018 10.1002/mrm.27045 10.1002/jmri.24217 10.1038/s41597-022-01255-z 10.1002/jmri.24313 10.1002/jmri.25883 10.1136/annrheumdis-2014-205310 10.1002/mrm.1910070105 10.1016/j.joca.2011.10.003 10.1097/BOR.0000000000000479 10.1007/s10462-020-09924-4 10.1002/jmri.26991 10.1016/j.joca.2021.02.563 10.1002/mrm.22036 10.1002/jmri.27331 10.1016/j.aanat.2021.151866 10.1148/ryai.2021200078 10.1002/jor.24994 10.1002/art.34453 10.1016/j.joca.2016.09.015 10.1002/acr.24539 10.1148/radiology.205.2.9356643 10.1148/radiol.2523090028 10.1002/jmri.26246 10.1016/j.media.2018.11.009 10.1007/s00330-019-06542-9 10.1177/19476035211042406 10.1016/j.joca.2011.05.004 10.1007/s10334-020-00889-7 10.1007/s10334-016-0532-9 10.1148/radiology.214.1.r00ja15259 10.1148/radiol.2018172322 10.1007/s10334-021-00934-z 10.1186/s12880-022-00793-7 |
| ContentType | Journal Article |
| Copyright | 2022 International Society for Magnetic Resonance in Medicine. 2023 International Society for Magnetic Resonance in Medicine |
| Copyright_xml | – notice: 2022 International Society for Magnetic Resonance in Medicine. – notice: 2023 International Society for Magnetic Resonance in Medicine |
| DBID | AAYXX CITATION CGR CUY CVF ECM EIF NPM 7QO 7TK 8FD FR3 K9. P64 7X8 5PM |
| DOI | 10.1002/jmri.28365 |
| DatabaseName | 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 PubMed Central (Full Participant titles) |
| 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 MEDLINE - Academic ProQuest Health & Medical Complete (Alumni) |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine |
| EISSN | 1522-2586 |
| EndPage | 1039 |
| ExternalDocumentID | PMC9849481 35852498 10_1002_jmri_28365 JMRI28365 |
| Genre | article Research Support, U.S. Gov't, Non-P.H.S Research Support, Non-U.S. Gov't Journal Article Research Support, N.I.H., Extramural |
| GrantInformation_xml | – fundername: National Institutes of Health funderid: P41 EB015891; K24 AR062068; R01‐AR074492; R01 EB002524; R00 EB022634; NIH R01‐AR077604 – fundername: National Science and Engineering Graduate Fellowship – fundername: GE Healthcare – fundername: National Science Foundation funderid: GRFP‐DGE 1656518 – fundername: Wu Tsai Human Performance Alliance and Stanford Medicine Precision Health and Integrated Diagnostics – fundername: NIH HHS grantid: K24 AR062068 – fundername: NIH HHS grantid: R01-AR074492 – fundername: NIBIB NIH HHS grantid: K99 EB022634 – fundername: NIBIB NIH HHS grantid: R00 EB022634 – fundername: NIH HHS grantid: P41 EB015891 – fundername: NIAMS NIH HHS grantid: K24 AR062068 – fundername: NIH HHS grantid: R00 EB022634 – fundername: NIH HHS grantid: NIH R01-AR077604 – fundername: NIBIB NIH HHS grantid: P41 EB015891 – fundername: NIAMS NIH HHS grantid: R01 AR077604 |
| 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 AAMMB AAYXX AEFGJ AEYWJ AGHNM AGQPQ AGXDD AGYGG AIDQK AIDYY AIQQE CITATION O8X CGR CUY CVF ECM EIF NPM 7QO 7TK 8FD FR3 K9. P64 7X8 5PM |
| ID | FETCH-LOGICAL-c4485-f8d38706d91905a93b4169b645055b38c7edec064fa2297baf4428f75c65a09a3 |
| IEDL.DBID | DRFUL |
| ISICitedReferencesCount | 21 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000827105300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1053-1807 1522-2586 |
| IngestDate | Tue Nov 04 02:05:56 EST 2025 Sun Nov 09 14:02:02 EST 2025 Sat Nov 29 14:26:00 EST 2025 Mon Jul 21 06:04:45 EDT 2025 Sat Nov 29 06:08:43 EST 2025 Tue Nov 18 20:53:18 EST 2025 Wed Jan 22 16:23:10 EST 2025 |
| IsDoiOpenAccess | false |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 4 |
| Keywords | segmentation machine learning osteoarthritis cartilage qMRI |
| Language | English |
| License | 2022 International Society for Magnetic Resonance in Medicine. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c4485-f8d38706d91905a93b4169b645055b38c7edec064fa2297baf4428f75c65a09a3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0002-6700-8424 0000-0003-3663-5799 0000-0001-9526-1614 0000-0001-7558-3800 0000-0003-3592-4089 0000-0003-0175-7657 0000-0002-3667-6796 0000-0003-3826-6122 0000-0002-2564-9071 |
| OpenAccessLink | https://www.ncbi.nlm.nih.gov/pmc/articles/9849481 |
| PMID | 35852498 |
| PQID | 2785203829 |
| PQPubID | 1006400 |
| PageCount | 11 |
| ParticipantIDs | pubmedcentral_primary_oai_pubmedcentral_nih_gov_9849481 proquest_miscellaneous_2691788819 proquest_journals_2785203829 pubmed_primary_35852498 crossref_primary_10_1002_jmri_28365 crossref_citationtrail_10_1002_jmri_28365 wiley_primary_10_1002_jmri_28365_JMRI28365 |
| PublicationCentury | 2000 |
| PublicationDate | April 2023 |
| PublicationDateYYYYMMDD | 2023-04-01 |
| PublicationDate_xml | – month: 04 year: 2023 text: April 2023 |
| 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 | 2023 |
| Publisher | John Wiley & Sons, Inc Wiley Subscription Services, Inc |
| Publisher_xml | – name: John Wiley & Sons, Inc – name: Wiley Subscription Services, Inc |
| References | 2013; 3 2009; 62 2019; 52 2017; 25 2000; 214 2021; 29 2018; 288 2018; 80 2022; 22 2009; 252 2011; 19 2018; 47 2021; 13 1997; 205 2021; 54 2021; 10 2022; 240 2021; 34 2013; 38 2020; 52 2020; 30 2021 2020 2017; 38 2021; 39 2022; 9 1988; 7 2019; 49 2019 2018; 30 2022; 74 2014; 39 2016; 29 2014; 73 2012; 64 2012; 20 2009; 17 e_1_2_7_6_1 Desai AD (e_1_2_7_26_1) 2021 e_1_2_7_5_1 e_1_2_7_4_1 e_1_2_7_3_1 e_1_2_7_9_1 e_1_2_7_8_1 Chaudhari AS (e_1_2_7_23_1) 2020 e_1_2_7_19_1 e_1_2_7_18_1 Desai AD (e_1_2_7_33_1) 2019 e_1_2_7_17_1 e_1_2_7_16_1 e_1_2_7_40_1 e_1_2_7_2_1 e_1_2_7_15_1 e_1_2_7_41_1 e_1_2_7_14_1 e_1_2_7_12_1 e_1_2_7_11_1 e_1_2_7_10_1 e_1_2_7_27_1 e_1_2_7_28_1 e_1_2_7_29_1 Desai AD (e_1_2_7_13_1) 2019 e_1_2_7_30_1 e_1_2_7_25_1 e_1_2_7_31_1 e_1_2_7_24_1 e_1_2_7_32_1 e_1_2_7_22_1 e_1_2_7_34_1 e_1_2_7_21_1 e_1_2_7_35_1 e_1_2_7_20_1 e_1_2_7_36_1 e_1_2_7_37_1 e_1_2_7_38_1 Matzat SJ (e_1_2_7_7_1) 2013; 3 e_1_2_7_39_1 35959715 - J Magn Reson Imaging. 2023 Apr;57(4):1040-1041 |
| References_xml | – volume: 54 start-page: 2445 year: 2021 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: 29 start-page: 849 year: 2021 end-page: 858 article-title: Assessment of quantitative [ F]sodium fluoride PET measures of knee subchondral bone perfusion and mineralization in osteoarthritic and healthy subjects publication-title: Osteoarthr Cartil – volume: 13 start-page: 747S year: 2021 end-page: 756S article-title: Open source software for automatic subregional assessment of knee cartilage degradation using quantitative T2 relaxometry and deep learning publication-title: Cartilage – volume: 54 start-page: 357 year: 2021 end-page: 371 article-title: Prospective deployment of deep learning in MRI: A framework for important considerations, challenges, and recommendations for best practices publication-title: J Magn Reson Imaging – volume: 29 start-page: 207 year: 2016 end-page: 221 article-title: Segmentation of joint and musculoskeletal tissue in the study of arthritis publication-title: MAGMA – volume: 30 start-page: 2231 year: 2020 end-page: 2240 article-title: Time‐saving opportunities in knee osteoarthritis: T mapping and structural imaging of the knee using a single 5‐min MRI scan publication-title: Eur Radiol – volume: 22 start-page: 69 year: 2022 article-title: Transfer learning for medical image classification: A literature review publication-title: BMC Med Imaging – volume: 20 start-page: 13 year: 2012 end-page: 21 article-title: The diagnostic performance of MRI in osteoarthritis: A systematic review and meta‐analysis publication-title: Osteoarthr Cartil – volume: 19 start-page: 990 year: 2011 end-page: 1002 article-title: Evolution of semi‐quantitative whole joint assessment of knee OA: MOAKS (MRI Osteoarthritis Knee Score) publication-title: Osteoarthr Cartil – year: 2021 – volume: 205 start-page: 546 year: 1997 end-page: 550 article-title: Spatial variation of T2 in human articular cartilage publication-title: Radiology – volume: 38 start-page: 991 year: 2013 end-page: 1008 article-title: Quantitative MRI of articular cartilage and its clinical applications publication-title: J Magn Reson Imaging – volume: 7 start-page: 35 year: 1988 end-page: 42 article-title: A new steady‐state imaging sequence for simultaneous acquisition of two MR images with clearly different contrasts publication-title: Magn Reson Med – volume: 214 start-page: 259 year: 2000 end-page: 266 article-title: Human articular cartilage: Influence of aging and early symptomatic degeneration on the spatial variation of T2‐‐preliminary findings at 3 T publication-title: Radiology – volume: 25 start-page: 513 year: 2017 end-page: 520 article-title: Cluster analysis of quantitative MRI T and T relaxation times of cartilage identifies differences between healthy and ACL‐injured individuals at 3T publication-title: Osteoarthr Cartil – volume: 64 start-page: 1697 year: 2012 end-page: 1707 article-title: Osteoarthritis: A disease of the joint as an organ publication-title: Arthritis Rheum – volume: 52 start-page: 1321 year: 2020 end-page: 1339 article-title: Rapid knee MRI acquisition and analysis techniques for imaging osteoarthritis publication-title: J Magn Reson Imaging – volume: 49 start-page: 400 year: 2019 end-page: 410 article-title: 3D convolutional neural networks for detection and severity staging of meniscus and PFJ cartilage morphological degenerative changes in osteoarthritis and anterior cruciate ligament subjects publication-title: J Magn Reson Imaging – volume: 62 start-page: 544 year: 2009 end-page: 549 article-title: Rapid estimation of cartilage T2 based on double echo at steady state (DESS) with 3 tesla publication-title: Magn Reson Med – volume: 34 start-page: 859 year: 2021 end-page: 875 article-title: Automatic knee cartilage and bone segmentation using multi‐stage convolutional neural networks: Data from the osteoarthritis initiative publication-title: MAGMA – volume: 38 start-page: 63 year: 2017 end-page: 70 article-title: A simple analytic method for estimating T2 in the knee from DESS publication-title: Magn Reson Imaging – volume: 10 issue: 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: 47 start-page: 1328 year: 2018 end-page: 1341 article-title: Five‐minute knee MRI for simultaneous morphometry and T relaxometry of cartilage and meniscus and for semiquantitative radiological assessment using double‐echo in steady‐state at 3T publication-title: J Magn Reson Imaging – volume: 73 start-page: 1289 year: 2014 end-page: 1300 article-title: Imaging research results from the Osteoarthritis Initiative (OAI): A review and lessons learned 10 years after start of enrolment publication-title: Ann Rheum Dis – volume: 240 year: 2022 article-title: Local MRI‐based measures of thigh adipose tissue derived from fully automated deep convolutional neural network‐based segmentation show a comparable responsiveness to bidirectional change in body weight as from quality controlled manual segmentation publication-title: Ann Anat – year: 2019 article-title: Technical considerations for semantic segmentation in MRI using convolutional neural networks publication-title: arXiv – volume: 39 start-page: 972 year: 2014 end-page: 977 article-title: Quantitative measurement of femoral condyle cartilage in the knee by MRI: Validation study by multireaders publication-title: J Magn Reson Imaging – volume: 30 start-page: 160 year: 2018 end-page: 167 article-title: Epidemiology of osteoarthritis: Literature update publication-title: Curr Opin Rheumatol – volume: 3 start-page: 162 year: 2013 end-page: 174 article-title: Quantitative MRI techniques of cartilage composition publication-title: Quant Imaging Med Surg – year: 2020 – volume: 74 start-page: 929 year: 2022 end-page: 936 article-title: A deep learning automated segmentation algorithm accurately detects differences in longitudinal cartilage thickness loss ‐ data from the FNIH biomarkers study of the osteoarthritis initiative publication-title: Arthritis Care Res (Hoboken) – 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 – volume: 252 start-page: 486 year: 2009 end-page: 495 article-title: Knee joint: Comprehensive assessment with 3D isotropic resolution fast spin‐echo MR imaging‐‐diagnostic performance compared with that of conventional MR imaging at 3.0 T publication-title: Radiology – 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: 9 start-page: 152 year: 2022 article-title: fastMRI+, clinical pathology annotations for knee and brain fully sampled magnetic resonance imaging data publication-title: Sci Data – volume: 17 start-page: 1589 year: 2009 end-page: 1597 article-title: Intra‐ and inter‐observer reproducibility of volume measurement of knee cartilage segmented from the OAI MR image set using a novel semi‐automated segmentation method publication-title: Osteoarthr Cartil – volume: 39 start-page: 2340 year: 2021 end-page: 2352 article-title: Characterizing the transient response of knee cartilage to running: Decreases in cartilage T of female recreational runners publication-title: J Orthop Res – year: 2019 – 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: MAGMA – volume: 80 start-page: 529 year: 2018 end-page: 537 article-title: Simultaneous bilateral‐knee MR imaging publication-title: Magn Reson Med – ident: e_1_2_7_41_1 doi: 10.1016/j.joca.2009.06.003 – ident: e_1_2_7_30_1 doi: 10.1016/j.mri.2016.12.018 – ident: e_1_2_7_27_1 doi: 10.1002/mrm.27045 – ident: e_1_2_7_40_1 doi: 10.1002/jmri.24217 – ident: e_1_2_7_39_1 doi: 10.1038/s41597-022-01255-z – ident: e_1_2_7_6_1 doi: 10.1002/jmri.24313 – ident: e_1_2_7_29_1 doi: 10.1002/jmri.25883 – ident: e_1_2_7_10_1 doi: 10.1136/annrheumdis-2014-205310 – ident: e_1_2_7_28_1 doi: 10.1002/mrm.1910070105 – ident: e_1_2_7_5_1 doi: 10.1016/j.joca.2011.10.003 – volume-title: Thirty‐fifth Conference on Neural Information Processing Systems Datasets and Benchmarks (Round 2) year: 2021 ident: e_1_2_7_26_1 – ident: e_1_2_7_3_1 doi: 10.1097/BOR.0000000000000479 – ident: e_1_2_7_14_1 doi: 10.1007/s10462-020-09924-4 – ident: e_1_2_7_4_1 doi: 10.1002/jmri.26991 – ident: e_1_2_7_25_1 doi: 10.1016/j.joca.2021.02.563 – ident: e_1_2_7_31_1 doi: 10.1002/mrm.22036 – ident: e_1_2_7_21_1 doi: 10.1002/jmri.27331 – ident: e_1_2_7_38_1 doi: 10.1016/j.aanat.2021.151866 – ident: e_1_2_7_15_1 doi: 10.1148/ryai.2021200078 – ident: e_1_2_7_24_1 doi: 10.1002/jor.24994 – ident: e_1_2_7_2_1 doi: 10.1002/art.34453 – volume: 3 start-page: 162 year: 2013 ident: e_1_2_7_7_1 article-title: Quantitative MRI techniques of cartilage composition publication-title: Quant Imaging Med Surg – volume-title: Proceedings of the 27th Annual Meeting of ISMRM year: 2019 ident: e_1_2_7_33_1 – ident: e_1_2_7_34_1 doi: 10.1016/j.joca.2016.09.015 – year: 2019 ident: e_1_2_7_13_1 article-title: Technical considerations for semantic segmentation in MRI using convolutional neural networks publication-title: arXiv – volume-title: Proceedings of the 27th Annual Meeting of ISMRM (virtual) year: 2020 ident: e_1_2_7_23_1 – ident: e_1_2_7_16_1 doi: 10.1002/acr.24539 – ident: e_1_2_7_8_1 doi: 10.1148/radiology.205.2.9356643 – ident: e_1_2_7_37_1 doi: 10.1148/radiol.2523090028 – ident: e_1_2_7_12_1 doi: 10.1002/jmri.26246 – ident: e_1_2_7_17_1 doi: 10.1016/j.media.2018.11.009 – ident: e_1_2_7_32_1 doi: 10.1007/s00330-019-06542-9 – ident: e_1_2_7_35_1 doi: 10.1177/19476035211042406 – ident: e_1_2_7_36_1 doi: 10.1016/j.joca.2011.05.004 – ident: e_1_2_7_20_1 doi: 10.1007/s10334-020-00889-7 – ident: e_1_2_7_11_1 doi: 10.1007/s10334-016-0532-9 – ident: e_1_2_7_9_1 doi: 10.1148/radiology.214.1.r00ja15259 – ident: e_1_2_7_18_1 doi: 10.1148/radiol.2018172322 – ident: e_1_2_7_19_1 doi: 10.1007/s10334-021-00934-z – ident: e_1_2_7_22_1 doi: 10.1186/s12880-022-00793-7 – reference: 35959715 - J Magn Reson Imaging. 2023 Apr;57(4):1040-1041 |
| SSID | ssj0009945 |
| Score | 2.5245414 |
| Snippet | Background
Deep learning (DL)‐based automatic segmentation models can expedite manual segmentation yet require resource‐intensive fine‐tuning before deployment... Deep learning (DL)-based automatic segmentation models can expedite manual segmentation yet require resource-intensive fine-tuning before deployment on new... BackgroundDeep learning (DL)‐based automatic segmentation models can expedite manual segmentation yet require resource‐intensive fine‐tuning before deployment... |
| SourceID | pubmedcentral proquest pubmed crossref wiley |
| SourceType | Open Access Repository Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 1029 |
| SubjectTerms | Accuracy Algorithms Arthritis Cartilage Cartilage diseases Cartilage, Articular - pathology Coefficient of variation Correlation coefficient Correlation coefficients Data acquisition Datasets Deep Learning Female Females Field strength Humans Knee Machine learning Magnetic resonance imaging Magnetic Resonance Imaging - methods Mathematical models Osteoarthritis Osteoarthritis, Knee - pathology Parameters Population studies qMRI Retrospective Studies Scanners Segmentation Statistical analysis Statistical tests Tuning Variation |
| Title | Generalizability of Deep Learning Segmentation Algorithms for Automated Assessment of Cartilage Morphology and MRI Relaxometry |
| URI | https://onlinelibrary.wiley.com/doi/abs/10.1002%2Fjmri.28365 https://www.ncbi.nlm.nih.gov/pubmed/35852498 https://www.proquest.com/docview/2785203829 https://www.proquest.com/docview/2691788819 https://pubmed.ncbi.nlm.nih.gov/PMC9849481 |
| Volume | 57 |
| WOSCitedRecordID | wos000827105300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVWIB databaseName: Wiley Online Library Full Collection 2020 customDbUrl: eissn: 1522-2586 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0009945 issn: 1053-1807 databaseCode: DRFUL dateStart: 19990101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3db9MwED9tHUK88P0RGJMRvICULbGT2JZ4qTYqQHRCg0l9i5zY7oraZEpTxF742zknaUo1hIR4i-RLYvnu7N_Z598BvOI8MzLXzKc2s37EmfYV2rFvQqG5sTJPYtUUm-Cnp2IykZ934O36LkzLD9FvuDnPaOZr5-AqWx5tSEO_LarZIS6OSbwLexQNNx7A3snZ6PzThnRXNkWKEUIwPxQB7-lJ6dHm7e0F6RrKvJ4s-TuIbVah0Z3_6_9duN2hTzJszeUe7JjiPtwcd-frD-Bnx0LtUr1c0uwVKS05MeaSdDysU_LFTBfdfaWCDOfTsprVF4slQfBLhqu6RARsNBn2hJ_uC8fOPuc4c5FxiXptdvKJKjQZn30gLh3vR7kwdXX1EM5H774ev_e7Cg1-jmFd7FuhmTso1RJxRawkyxDfySyJEFfFGRM5N9rkiHqsolTyTNkIwx3LY2cCgVTsEQyKsjBPgJicW6MirSJBoyxMFOU2l4FiVoeB4JkHr9dqSvOOvtxV0ZinLfEyTd2Aps2AevCyl71sSTv-KLW_1nbaOe4ypVzENGCCSg9e9M3ocu4cRRWmXKFMgjGuEIilPHjcGkf_G4bhF0a0wgO-ZTa9gKPz3m4pZhcNrbcUjqon9OBNYzZ_6Xn6EdXTPD39F-FncIsiJGtzzvdhUFcr8xxu5N_r2bI6gF0-EQedC_0CsuQi3Q |
| linkProvider | Wiley-Blackwell |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1bb9MwFD6CDQEv3C-BAUbwAlJY4lxsP1Yb1QZthcYm7S1yYrsrapMpTRF74bdznHgp1RAS4i2ST272d-zv2MefAd4ylmtRqMinJjd-zCLlS8Sxr0OumDaiSBPZHjbBJhN-eiq-uNwcuxem04foJ9ysZ7T9tXVwOyG9u1YN_baoZx9wdEyT67AdI44Q4Nv7R8OT0Vp1V7SnFCOHiPyQB6zXJ6W767s3R6QrNPNqtuTvLLYdhoZ3__MH7sEdxz_JoAPMfbimywdwc-xW2B_CT6dDbZO9bNrsBakM2df6nDgl1in5qqcLt2OpJIP5tKpnzdliSZD-ksGqqZADa0UGveSnfcKeRegc-y4yrrBl27l8IktFxkeHxCbk_agWuqkvHsHJ8OPx3oHvzmjwCwzsEt9wFdmlUiWQWSRSRDkyPJGnMTKrJI94wbTSBfIeIykVLJcmxoDHsMSCIBAyegxbZVXqp0B0wYyWsZIxp3EeppIyU4hARkaFAWe5B-8u2ykrnIC5PUdjnnXSyzSzFZq1FerBm972vJPt-KPVzmVzZ851lxllPKFBxKnw4HVfjE5nV1JkqasV2qQY5XKObMqDJx06-tdEGIBhTMs9YBu46Q2soPdmSTk7a4W9BbdiPaEH71vc_OXLs0_YPO3Vs38xfgW3Do7Ho2x0OPn8HG5TJGxdFtIObDX1Sr-AG8X3ZrasXzpP-gUnJSXm |
| linkToPdf | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3db9MwED9BhyZe-P4IDDCCF5CyJc6H7cdqpWKwVtNg0t4iJ7a7ojap0haxF_52zomXUg0hId4i-fLlu7N_Z59_B_CWsVyLQkU-NbnxYxYpX6Id-zrkimkjijSRTbEJNh7z83Nx4nJz7FmYlh-iW3CzntGM19bB9UKZgw1r6Ld5Pd3H2TFNbsJObKvI9GBncDo8O96w7oqmSjFiiMgPecA6flJ6sLl7e0a6BjOvZ0v-jmKbaWh49z9_4B7ccfiT9FuDuQ83dPkAdkduh_0h_HQ81DbZy6bNXpLKkIHWC-KYWCfki57M3YmlkvRnk6qeri7mS4Lwl_TXqwoxsFak31F-2iccWgud4dhFRhVqtlnLJ7JUZHR6RGxC3o9qrlf15SM4G374evjRdzUa_AIDu8Q3XEV2q1QJRBaJFFGOCE_kaYzIKskjXjCtdIG4x0hKBculiTHgMSyxRhAIGT2GXlmV-ikQXTCjZaxkzGmch6mkzBQikJFRYcBZ7sG7Kz1lhSMwt3U0ZllLvUwz26FZ06EevOlkFy1txx-l9q7UnTnXXWaU8YQGEafCg9ddMzqd3UmRpa7WKJNilMs5oikPnrTW0b0mwgAMY1ruAduym07AEnpvt5TTi4bYW3BL1hN68L6xm798efYJ1dNcPfsX4VewezIYZsdH48_P4TZFvNYmIe1Bb1Wv9Qu4VXxfTZf1S-dIvwB-IiVh |
| 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=Generalizability+of+Deep+Learning+Segmentation+Algorithms+for+Automated+Assessment+of+Cartilage+Morphology+and+MRI+Relaxometry&rft.jtitle=Journal+of+magnetic+resonance+imaging&rft.au=Schmidt%2C+Andrew+M.&rft.au=Desai%2C+Arjun+D.&rft.au=Watkins%2C+Lauren+E.&rft.au=Crowder%2C+Hollis&rft.date=2023-04-01&rft.issn=1053-1807&rft.eissn=1522-2586&rft.volume=57&rft.issue=4&rft.spage=1029&rft.epage=1039&rft_id=info:doi/10.1002%2Fjmri.28365&rft_id=info%3Apmid%2F35852498&rft.externalDocID=PMC9849481 |
| 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 |