{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T18:02:45Z","timestamp":1790704965902,"version":"4.1.0"},"reference-count":65,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2018,5,17]],"date-time":"2018-05-17T00:00:00Z","timestamp":1526515200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Magnetic Resonance in Med"],"published-print":{"date-parts":[[2018,12]]},"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>To describe and evaluate a new segmentation method using deep convolutional neural network (CNN), 3D fully connected conditional random field (CRF), and 3D simplex deformable modeling to improve the efficiency and accuracy of knee joint tissue segmentation.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>A segmentation pipeline was built by combining a semantic segmentation CNN, 3D fully connected CRF, and 3D simplex deformable modeling. A convolutional encoder\u2010decoder network was designed as the core of the segmentation method to perform high resolution pixel\u2010wise multi\u2010class tissue classification for 12 different joint structures. The 3D fully connected CRF was applied to regularize contextual relationship among voxels within the same tissue class and between different classes. The 3D simplex deformable modeling refined the output from 3D CRF to preserve the overall shape and maintain a desirable smooth surface for joint structures. The method was evaluated on 3D fast spin\u2010echo (3D\u2010FSE) MR image data sets. Quantitative morphological metrics were used to evaluate the accuracy and robustness of the method in comparison to the ground truth data.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The proposed segmentation method provided good performance for segmenting all knee joint structures. There were 4 tissue types with high mean Dice coefficient above 0.9 including the femur, tibia, muscle, and other non\u2010specified tissues. There were 7 tissue types with mean Dice coefficient between 0.8 and 0.9 including the femoral cartilage, tibial cartilage, patella, patellar cartilage, meniscus, quadriceps and patellar tendon, and infrapatellar fat pad. There was 1 tissue type with mean Dice coefficient between 0.7 and 0.8 for joint effusion and Baker's cyst. Most musculoskeletal tissues had a mean value of average symmetric surface distance below 1\u2009mm.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>The combined CNN, 3D fully connected CRF, and 3D deformable modeling approach was well\u2010suited for performing rapid and accurate comprehensive tissue segmentation of the knee joint. The deep learning\u2010based segmentation method has promising potential applications in musculoskeletal imaging.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mrm.27229","type":"journal-article","created":{"date-parts":[[2018,5,18]],"date-time":"2018-05-18T03:05:38Z","timestamp":1526612738000},"page":"2759-2770","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":183,"title":["Deep convolutional neural network for segmentation of knee joint anatomy"],"prefix":"10.1002","volume":"80","author":[{"given":"Zhaoye","family":"Zhou","sequence":"first","affiliation":[{"name":"Department of Biomedical Engineering University of Minnesota  Minneapolis Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gengyan","family":"Zhao","sequence":"additional","affiliation":[{"name":"Departments of Radiology University of Wisconsin School of Medicine and Public Health  Madison Wisconsin"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Richard","family":"Kijowski","sequence":"additional","affiliation":[{"name":"Departments of Radiology University of Wisconsin School of Medicine and Public Health  Madison Wisconsin"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8032-6681","authenticated-orcid":false,"given":"Fang","family":"Liu","sequence":"additional","affiliation":[{"name":"Departments of Radiology University of Wisconsin School of Medicine and Public Health  Madison Wisconsin"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2018,5,17]]},"reference":[{"key":"e_1_2_6_2_1","doi-asserted-by":"publisher","DOI":"10.1002\/art.1780380817"},{"key":"e_1_2_6_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0033-8389(03)00161-1"},{"key":"e_1_2_6_4_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2262012190"},{"key":"e_1_2_6_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1063-4584(03)00160-2"},{"key":"e_1_2_6_6_1","doi-asserted-by":"publisher","DOI":"10.1002\/art.11483"},{"key":"e_1_2_6_7_1","doi-asserted-by":"publisher","DOI":"10.1186\/ar1875"},{"key":"e_1_2_6_8_1","doi-asserted-by":"publisher","DOI":"10.7326\/0003-4819-134-7-200104030-00007"},{"key":"e_1_2_6_9_1","first-page":"1330","article-title":"Knee effusions, popliteal cysts, and synovial thickening: association with knee pain in osteoarthritis","volume":"28","author":"Hill CL","year":"2001","journal-title":"J Rheumatol."},{"key":"e_1_2_6_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1063-4584(03)00080-3"},{"key":"e_1_2_6_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2006.02.025"},{"key":"e_1_2_6_12_1","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2014-205310"},{"key":"e_1_2_6_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2003.11.003"},{"key":"e_1_2_6_14_1","doi-asserted-by":"publisher","DOI":"10.1136\/ard.2006.066183"},{"key":"e_1_2_6_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2011.05.004"},{"key":"e_1_2_6_16_1","doi-asserted-by":"publisher","DOI":"10.1177\/1941738116663922"},{"key":"e_1_2_6_17_1","doi-asserted-by":"publisher","DOI":"10.1136\/ard.2008.099200"},{"key":"e_1_2_6_18_1","doi-asserted-by":"publisher","DOI":"10.1136\/ard.2008.089904"},{"key":"e_1_2_6_19_1","doi-asserted-by":"publisher","DOI":"10.1136\/ard.2008.099762"},{"key":"e_1_2_6_20_1","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2013-204157"},{"key":"e_1_2_6_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2012.09.011"},{"key":"e_1_2_6_22_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2015142201"},{"key":"e_1_2_6_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2014.06.026"},{"key":"e_1_2_6_24_1","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2015-207602"},{"key":"e_1_2_6_25_1","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2013-204052"},{"key":"e_1_2_6_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2007.11.003"},{"key":"e_1_2_6_27_1","doi-asserted-by":"publisher","DOI":"10.1186\/ar4292"},{"key":"e_1_2_6_28_1","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2474-14-3"},{"key":"e_1_2_6_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2016.03.015"},{"key":"e_1_2_6_30_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-011-2234-z"},{"key":"e_1_2_6_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2016.12.009"},{"key":"e_1_2_6_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2015.08.003"},{"key":"e_1_2_6_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2016.06.008"},{"key":"e_1_2_6_34_1","doi-asserted-by":"publisher","DOI":"10.1002\/acr.23326"},{"key":"e_1_2_6_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2017.02.789"},{"key":"e_1_2_6_36_1","doi-asserted-by":"publisher","DOI":"10.1002\/acr.20588"},{"key":"e_1_2_6_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2012.04.009"},{"key":"e_1_2_6_38_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2010.10.023"},{"key":"e_1_2_6_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2004.09.008"},{"key":"e_1_2_6_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2009.2024743"},{"key":"e_1_2_6_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2012.2186612"},{"issue":"2","key":"e_1_2_6_42_1","first-page":"246","article-title":"Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network","volume":"16","author":"Prasoon A","year":"2013","journal-title":"Med Image Comput Comput Assist Interv."},{"key":"e_1_2_6_43_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.26841"},{"key":"e_1_2_6_44_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2017170700"},{"key":"e_1_2_6_45_1","first-page":"1556","article-title":"Very deep convolutional networks for large\u2010scale image recognition","volume":"1409","author":"Simonyan K","year":"2014","journal-title":"arXiv e\u2010prints arXiv"},{"key":"e_1_2_6_46_1","first-page":"03167","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume":"1502","author":"Ioffe S","year":"2015","journal-title":"arXiv e\u2010prints arXiv"},{"key":"e_1_2_6_47_1","first-page":"807","volume-title":"Proceedings of the 27th International Conference on Machine Learning, Haifa, Israel","author":"Nair V","year":"2010"},{"key":"e_1_2_6_48_1","unstructured":"HeK ZhangX RenS SunJ.Identity mappings in deep residual networks. arXiv e\u2010prints arXiv2016:1603.05027."},{"key":"e_1_2_6_49_1","first-page":"695","volume-title":"Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, DC","author":"He X","year":"2004"},{"key":"e_1_2_6_50_1","unstructured":"Kr\u00e4henb\u00fchlP KoltunV.Efficient inference in fully connected CRFs with Gaussian edge potentials. arXiv e\u2010prints arXive2012:1210.5644."},{"key":"e_1_2_6_51_1","doi-asserted-by":"publisher","DOI":"10.1145\/37402.37422"},{"key":"e_1_2_6_52_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1008157432188"},{"key":"e_1_2_6_53_1","unstructured":"LongJ ShelhamerE DarrellT.Fully convolutional networks for semantic segmentation. arXiv e\u2010prints arXiv2014;1411.4038."},{"key":"e_1_2_6_54_1","unstructured":"AbadiM AgarwalA BarhamP et al.TensorFlow: large\u2010scale machine learning on heterogeneous distributed systems. arXiv e\u2010prints arXiv2016:1603.04467."},{"key":"e_1_2_6_55_1","unstructured":"Fran\u00e7oisChollet.Keras. GitHub 2015.https:\/\/github.com\/fchollet\/keras. Published March 27 2015. Updated February 13 2018. Accessed March 27 2015."},{"key":"e_1_2_6_56_1","first-page":"207","article-title":"Segmentation of knee images: a grand challenge","author":"Heimann T","year":"2010","journal-title":"Med Image Comput Comput Assist Interv."},{"key":"e_1_2_6_57_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Delving deep into rectifiers: surpassing human\u2010level performance on imagenet classification. arXiv e\u2010prints arXiv2015;1502.08152.","DOI":"10.1109\/ICCV.2015.123"},{"key":"e_1_2_6_58_1","unstructured":"KingmaDP BaJ.Adam: a method for stochastic optimization. arXiv e\u2010prints arXiv2014:1412.6980."},{"key":"e_1_2_6_59_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2016.10.004"},{"key":"e_1_2_6_60_1","doi-asserted-by":"crossref","unstructured":"MilletariF NavabN AhmadiSA.V\u2010Net: fully convolutional neural networks for volumetric medical image segmentation. arXiv e\u2010prints arXiv2016:1606.04797.","DOI":"10.1109\/3DV.2016.79"},{"key":"e_1_2_6_61_1","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek\u00d6 AbdulkadirA LienkampSS BroxT RonnebergerO.3D U\u2010Net: learning dense volumetric segmentation from sparse annotation. arXiv e\u2010prints arXiv2016:1606.06650.","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"e_1_2_6_62_1","unstructured":"ChristPF EttlingerF Gr\u00fcnF et al.Automatic liver and tumor segmentation of CT and MRI volumes using cascaded fully convolutional neural networks. arXiv e\u2010prints arXiv2017:1702.05970"},{"key":"e_1_2_6_63_1","unstructured":"ChenLC PapandreouG KokkinosI MurphyK YuilleAL.Semantic image segmentation with deep convolutional nets and fully connected CRFs. arXiv e\u2010prints arXiv2014:1412.7062"},{"key":"e_1_2_6_64_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.24901"},{"key":"e_1_2_6_65_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.25689"},{"key":"e_1_2_6_66_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2014.2366792"}],"container-title":["Magnetic Resonance in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fmrm.27229","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mrm.27229","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,15]],"date-time":"2023-09-15T14:25:16Z","timestamp":1694787916000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/mrm.27229"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,5,17]]},"references-count":65,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2018,12]]}},"alternative-id":["10.1002\/mrm.27229"],"URL":"https:\/\/doi.org\/10.1002\/mrm.27229","archive":["Portico"],"relation":{},"ISSN":["0740-3194","1522-2594"],"issn-type":[{"value":"0740-3194","type":"print"},{"value":"1522-2594","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,5,17]]},"assertion":[{"value":"2017-12-12","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2018-03-31","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2018-05-17","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}