{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T23:12:21Z","timestamp":1790809941576,"version":"4.1.0"},"reference-count":45,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2018,12,17]],"date-time":"2018-12-17T00:00:00Z","timestamp":1545004800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2019,2]]},"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>Radiation therapy (RT) is a common treatment option for head and neck (HaN) cancer. An important step involved in RT planning is the delineation of organs\u2010at\u2010risks (OARs) based on HaN computed tomography (CT). However, manually delineating OARs is time\u2010consuming as each slice of CT images needs to be individually examined and a typical CT consists of hundreds of slices. Automating OARs segmentation has the benefit of both reducing the time and improving the quality of RT planning. Existing anatomy autosegmentation algorithms use primarily atlas\u2010based methods, which require sophisticated atlas creation and cannot adequately account for anatomy variations among patients. In this work, we propose an end\u2010to\u2010end, atlas\u2010free three\u2010dimensional (3D) convolutional deep learning framework for fast and fully automated whole\u2010volume HaN anatomy segmentation.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Our deep learning model, called AnatomyNet, segments OARs from head and neck CT images in an end\u2010to\u2010end fashion, receiving whole\u2010volume HaN CT images as input and generating masks of all OARs of interest in one shot. AnatomyNet is built upon the popular 3D U\u2010net architecture, but extends it in three important ways: (a) a new encoding scheme to allow autosegmentation on whole\u2010volume CT images instead of local patches or subsets of slices, (b) incorporating 3D squeeze\u2010and\u2010excitation residual blocks in encoding layers for better feature representation, and (c) a new loss function combining Dice scores and focal loss to facilitate the training of the neural model. These features are designed to address two main challenges in deep learning\u2010based HaN segmentation: (a) segmenting small anatomies (i.e., optic chiasm and optic nerves) occupying only a few slices, and (b) training with inconsistent data annotations with missing ground truth for some anatomical structures.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We collected 261 HaN CT images to train AnatomyNet and used MICCAI Head and Neck Auto Segmentation Challenge 2015 as a benchmark dataset to evaluate the performance of AnatomyNet. The objective is to segment nine anatomies: brain stem, chiasm, mandible, optic nerve left, optic nerve right, parotid gland left, parotid gland right, submandibular gland left, and submandibular gland right. Compared to previous state\u2010of\u2010the\u2010art results from the MICCAI 2015 competition, AnatomyNet increases Dice similarity coefficient by 3.3% on average. AnatomyNet takes about 0.12\u00a0s to fully segment a head and neck CT image of dimension 178\u00a0\u00d7\u00a0302\u00a0\u00d7\u00a0225, significantly faster than previous methods. In addition, the model is able to process whole\u2010volume CT images and delineate all OARs in one pass, requiring little pre\u2010 or postprocessing.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>Deep learning models offer a feasible solution to the problem of delineating OARs from CT images. We demonstrate that our proposed model can improve segmentation accuracy and simplify the autosegmentation pipeline. With this method, it is possible to delineate OARs of a head and neck CT within a fraction of a second.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.13300","type":"journal-article","created":{"date-parts":[[2018,11,27]],"date-time":"2018-11-27T15:34:43Z","timestamp":1543332883000},"page":"576-589","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":462,"title":["AnatomyNet: Deep learning for fast and fully automated whole\u2010volume segmentation of head and neck anatomy"],"prefix":"10.1002","volume":"46","author":[{"given":"Wentao","family":"Zhu","sequence":"first","affiliation":[{"name":"Department of Computer Science University of California Irvine CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yufang","family":"Huang","sequence":"additional","affiliation":[{"name":"Lenovo Research Beijing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang","family":"Zeng","sequence":"additional","affiliation":[{"name":"Deepvoxel Inc Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuming","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Shanghai General Hospital Shanghai Jiao Tong University School of Medicine Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yong","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Shanghai General Hospital Shanghai Jiao Tong University School of Medicine Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhen","family":"Qian","sequence":"additional","affiliation":[{"name":"Tencent Medical AI Lab Palo Alto CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nan","family":"Du","sequence":"additional","affiliation":[{"name":"Tencent Medical AI Lab Palo Alto CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Fan","sequence":"additional","affiliation":[{"name":"Tencent Medical AI Lab Palo Alto CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaohui","family":"Xie","sequence":"additional","affiliation":[{"name":"Department of Computer Science University of California Irvine CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2018,12,17]]},"reference":[{"key":"e_1_2_8_2_1","first-page":"87","article-title":"Global cancer statistics, 2012","volume":"65","author":"Torre LA","year":"2015","journal-title":"CA"},{"key":"e_1_2_8_3_1","first-page":"434","article-title":"Atlas\u2010based auto\u2010segmentation of head and neck CT images","volume":"11","author":"Han X","year":"2008","journal-title":"Med Image Comput Comput Assist Interv"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4871620"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12197"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2010.11.017"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2007.11.030"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4871623"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2008.01.018"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2009.08.013"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4810971"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4901270"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2015.2442753"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4927567"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/60\/16\/6547"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2007.01.038"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3515459"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3654160"},{"key":"e_1_2_8_19_1","doi-asserted-by":"crossref","unstructured":"LeavensC VikT SchulzH et\u00a0al.Validation of automatic landmark identification for atlas\u2010based segmentation for radiation treatment planning of the head\u2010and\u2010neck region. In:SPIE;2008.","DOI":"10.1117\/12.769710"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20180095"},{"key":"e_1_2_8_21_1","doi-asserted-by":"crossref","unstructured":"TamC YangX TianS JiangX BeitlerJ LiS.Automated delineation of organs\u2010at\u2010risk in head and neck ct images using multi\u2010output support vector regression. In:SPIE;2018.","DOI":"10.1117\/12.2292556"},{"key":"e_1_2_8_22_1","doi-asserted-by":"crossref","unstructured":"WuX UdupaJK TongY et\u00a0al.Auto\u2010contouring via automatic anatomy recognition of organs at risk in head and neck cancer on CT images. In:SPIE;2018.","DOI":"10.1117\/12.2293946"},{"key":"e_1_2_8_23_1","doi-asserted-by":"crossref","unstructured":"TongY UdupaJK WuX et\u00a0al.Hierarchical model\u2010based object localization for auto\u2010contouring in head and neck radiation therapy planning. In:SPIE;2018.","DOI":"10.1117\/12.2294042"},{"key":"e_1_2_8_24_1","unstructured":"PednekarGV UdupaJK McLaughlinDJ et\u00a0al.Image quality and segmentation. In:SPIE;2018."},{"key":"e_1_2_8_25_1","doi-asserted-by":"crossref","unstructured":"WangZ WeiL WangL GaoY ChenW ShenD.Hierarchical vertex regression\u2010based segmentation of head and neck CT images for radiotherapy planning. In:IEEE TIP;2018.","DOI":"10.1109\/TIP.2017.2768621"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46723-8_19"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12045"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12837"},{"key":"e_1_2_8_30_1","doi-asserted-by":"crossref","unstructured":"H\u00e4nschA SchwierM GassT et\u00a0al.Comparison of different deep learning approaches for parotid gland segmentation from ct images. In:SPIE;2018.","DOI":"10.1117\/12.2292962"},{"key":"e_1_2_8_31_1","doi-asserted-by":"crossref","unstructured":"HuJ ShenL SunG.Squeeze\u2010and\u2010excitation networks. In:IEEE CVPR;2018.","DOI":"10.1109\/CVPR.2018.00745"},{"key":"e_1_2_8_32_1","doi-asserted-by":"crossref","unstructured":"SalehiSSM ErdogmusD GholipourA.Tversky loss function for image segmentation using 3D fully convolutional deep networks. In:International Workshop on MLMI;2017.","DOI":"10.1007\/978-3-319-67389-9_44"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2006.880587"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-67558-9_28"},{"key":"e_1_2_8_35_1","doi-asserted-by":"crossref","unstructured":"LinTY GoyalP GirshickR HeK DollarP.Focal loss for dense object detection. In:CVPR;2017.","DOI":"10.1109\/ICCV.2017.324"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-66179-7_69"},{"key":"e_1_2_8_37_1","unstructured":"2018 Springer Cham K Wong M Moradi H Tang T Syeda\u2010Mahmood 3D segmentation with exponential logarithmic loss for highly unbalanced object sizesMICCAI 612 619"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-013-9622-7"},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-017-10371-5"},{"key":"e_1_2_8_40_1","doi-asserted-by":"crossref","unstructured":"ZhuW LiuC FanW XieX.Deeplung: deep 3D dual path nets for automated pulmonary nodule detection and classification. In:IEEE WACV;2018.","DOI":"10.1109\/WACV.2018.00079"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-00934-2_90"},{"key":"e_1_2_8_42_1","unstructured":"MaasAL HannunAY NgAY.Rectifier nonlinearities improve neural network acoustic models. In:Proceedings of ICML Vol. 30;2013:3."},{"key":"e_1_2_8_43_1","first-page":"26","article-title":"Lecture 6.5\u2010rmsprop: divide the gradient by a running average of its recent magnitude","volume":"4","author":"Tieleman T","year":"2012","journal-title":"COURSERA: Neural networks Mach Learn"},{"key":"e_1_2_8_44_1","doi-asserted-by":"crossref","unstructured":"LinT\u2010Y Doll\u00e1rP GirshickR HeK HariharanB BelongieS.Feature pyramid networks for object detection. In:CVPR. Vol. 1;2017:4.","DOI":"10.1109\/CVPR.2017.106"},{"key":"e_1_2_8_45_1","unstructured":"NikolovS BlackwellS MendesR et\u00a0al.Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy.arXiv preprint arXiv:1809.04430;2018."},{"key":"e_1_2_8_46_1","doi-asserted-by":"crossref","unstructured":"ZhuW XiangX TranTD HagerGD XieX.Adversarial deep structured nets for mass segmentation from mammograms. In:IEEE ISBI;2018.","DOI":"10.1109\/ISBI.2018.8363704"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fmp.13300","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.13300","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.13300","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.13300","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,5]],"date-time":"2023-09-05T15:40:00Z","timestamp":1693928400000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.13300"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,12,17]]},"references-count":45,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2019,2]]}},"alternative-id":["10.1002\/mp.13300"],"URL":"https:\/\/doi.org\/10.1002\/mp.13300","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,12,17]]},"assertion":[{"value":"2018-08-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2018-11-07","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2018-12-17","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}