{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T06:08:20Z","timestamp":1790230100759,"version":"4.1.0"},"reference-count":51,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2021,2,19]],"date-time":"2021-02-19T00:00:00Z","timestamp":1613692800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["P30 CA016672"],"award-info":[{"award-number":["P30 CA016672"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2021,4]]},"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>To enable generation of high\u2010quality deep learning segmentation models from severely limited contoured cases (e.g., ~10 cases).<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Thirty head and neck computed tomography (CT) scans with well\u2010defined contours were deformably registered to 200 CT scans of the same anatomic site without contours. Acquired deformation vector fields were used to train a principal component analysis (PCA) model for each of the 30 contoured CT scans by capturing the mean deformation and most prominent variations. Each PCA model can produce an infinite number of synthetic CT scans and corresponding contours by applying random deformations. We used 300, 600, 1000, and 2000 synthetic CT scans and contours generated from one PCA model to train V\u2010Net, a 3D convolutional neural network architecture, to segment parotid and submandibular glands. We repeated the training using same numbers of training cases generated from 7, 10, 20, and 30 PCA models, with the data distributed evenly between each PCA model. Performance of the segmentation models was evaluated with Dice similarity coefficients between auto\u2010generated contours and physician\u2010drawn contours on 162 test CT scans for parotid glands and another 21 test CT scans for submandibular glands.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Dice values varied with the number of synthetic CT scans and the number of PCA models used to train the network. By using 2000 synthetic CT scans generated from 10 PCA models, we achieved Dice values of 82.8%\u00a0\u00b1\u00a06.8% for right parotid, 82.0%\u00a0\u00b1\u00a06.9% for left parotid, and 74.2%\u00a0\u00b1\u00a06.8% for submandibular glands. These results are comparable with those obtained from state\u2010of\u2010the\u2010art auto\u2010contouring approaches, including a deep learning network trained from more than 1000 contoured patients and a multi\u2010atlas algorithm from 12 well\u2010contoured atlases. Improvement was marginal when &gt;10 PCA models or &gt;2000 synthetic CT scans were used.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>We demonstrated an effective data augmentation approach to train high\u2010quality deep learning segmentation models from a limited number of well\u2010contoured patient cases.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.14728","type":"journal-article","created":{"date-parts":[[2021,1,21]],"date-time":"2021-01-21T07:03:05Z","timestamp":1611212585000},"page":"1697-1706","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":13,"title":["Training deep\u2010learning segmentation models from severely limited data"],"prefix":"10.1002","volume":"48","author":[{"given":"Yao","family":"Zhao","sequence":"first","affiliation":[{"name":"Department of Radiation Physics The University of Texas MD Anderson Cancer Center Houston TX USA"},{"name":"The University of Texas MD Anderson Graduate School of Biomedical Science Houston TX USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dong Joo","family":"Rhee","sequence":"additional","affiliation":[{"name":"Department of Radiation Physics The University of Texas MD Anderson Cancer Center Houston TX USA"},{"name":"The University of Texas MD Anderson Graduate School of Biomedical Science Houston TX USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carlos","family":"Cardenas","sequence":"additional","affiliation":[{"name":"Department of Radiation Physics The University of Texas MD Anderson Cancer Center Houston TX USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Laurence E.","family":"Court","sequence":"additional","affiliation":[{"name":"Department of Radiation Physics The University of Texas MD Anderson Cancer Center Houston TX USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinzhong","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Radiation Physics The University of Texas MD Anderson Cancer Center Houston TX USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2021,2,19]]},"reference":[{"key":"e_1_2_7_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2016.08.039"},{"key":"e_1_2_7_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2018.01.114"},{"key":"e_1_2_7_4_1","unstructured":"NikolovS BlackwellS MendesR et al.Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy. ArXiv180904430 Phys Stat. Published online September 12 2018. Accessed June 25 2020.http:\/\/arxiv.org\/abs\/1809.04430"},{"key":"e_1_2_7_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12045"},{"key":"e_1_2_7_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.semradonc.2019.02.001"},{"key":"e_1_2_7_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/MIS.2009.36"},{"key":"e_1_2_7_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2010.07.009"},{"key":"e_1_2_7_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2012.02.010"},{"key":"e_1_2_7_10_1","doi-asserted-by":"publisher","DOI":"10.1186\/s40537-019-0197-0"},{"key":"e_1_2_7_11_1","first-page":"1097","volume-title":"Advances in Neural Information Processing Systems 25","author":"Krizhevsky A","year":"2012"},{"key":"e_1_2_7_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2538465"},{"key":"e_1_2_7_13_1","doi-asserted-by":"crossref","unstructured":"OliveiraA PereiraS SilvaCA.Augmenting data when training a CNN for retinal vessel segmentation: How to warp?In:2017 IEEE 5th Portuguese Meeting on Bioengineering (ENBENG). IEEE;2017:1\u20134.https:\/\/doi.org\/10.1109\/ENBENG.2017.7889443","DOI":"10.1109\/ENBENG.2017.7889443"},{"key":"e_1_2_7_14_1","doi-asserted-by":"crossref","unstructured":"ShijieJ PingW PeiyiJ SipingH.Research on data augmentation for image classification based on convolution neural networks. In:2017 Chinese Automation Congress (CAC). IEEE;2017:4165\u20134170.https:\/\/doi.org\/10.1109\/CAC.2017.8243510","DOI":"10.1109\/CAC.2017.8243510"},{"key":"e_1_2_7_15_1","unstructured":"TaylorL NitschkeG.Improving Deep Learning using Generic Data Augmentation. ArXiv170806020 Cs Stat. Published online August 20 2017. Accessed June 25 2020.http:\/\/arxiv.org\/abs\/1708.06020."},{"key":"e_1_2_7_16_1","unstructured":"Eaton\u2010RosenZ BragmanF OurselinS CardosoMJ.Improving Data Augmentation for Medical Image Segmentation. Published online April 11 2018. Accessed June 25 2020.https:\/\/openreview.net\/forum?id=rkBBChjiG."},{"key":"e_1_2_7_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2015.2496141"},{"key":"e_1_2_7_18_1","unstructured":"HaubergS FreifeldO LarsenABL FisherJWIII HansenLK.Dreaming More Data: Class\u2010dependent Distributions over Diffeomorphisms for Learned Data Augmentation. ArXiv151002795 Cs. Published online June 30 2016. Accessed June 3 2020.http:\/\/arxiv.org\/abs\/1510.02795."},{"key":"e_1_2_7_19_1","unstructured":"RatnerAJ EhrenbergHR HussainZ DunnmonJ R\u00e9C.Learning to Compose Domain\u2010Specific Transformations for Data Augmentation. ArXiv170901643 Cs Stat. Published online September 30 2017. Accessed June 3 2020.http:\/\/arxiv.org\/abs\/1709.01643"},{"key":"e_1_2_7_20_1","doi-asserted-by":"crossref","unstructured":"CubukED ZophB ManeD VasudevanV AutoAugmentLQV.Learning Augmentation Strategies From Data. In:2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE;2019:113\u2013123.https:\/\/doi.org\/10.1109\/CVPR.2019.00020","DOI":"10.1109\/CVPR.2019.00020"},{"key":"e_1_2_7_21_1","unstructured":"ZhaoA BalakrishnanG DurandF GuttagJV DalcaAV.Data augmentation using learned transformations for one\u2010shot medical image segmentation. ArXiv190209383 Cs. Published online April 6 2019. Accessed May 21 2020.http:\/\/arxiv.org\/abs\/1902.09383."},{"key":"e_1_2_7_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101552"},{"key":"e_1_2_7_23_1","doi-asserted-by":"crossref","unstructured":"ChuquicusmaMJM HusseinS BurtJ BagciU.How to fool radiologists with generative adversarial networks? A visual turing test for lung cancer diagnosis. In:2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). IEEE;2018:240\u2013244.https:\/\/doi.org\/10.1109\/ISBI.2018.8363564","DOI":"10.1109\/ISBI.2018.8363564"},{"key":"e_1_2_7_24_1","doi-asserted-by":"crossref","unstructured":"MoradiM MadaniA KarargyrisA Syeda\u2010MahmoodTF.Chest x\u2010ray generation and data augmentation for cardiovascular abnormality classification. In: Angelini ED Landman BA ed.Medical Imaging 2018: Image Processing. SPIE;2018:57.https:\/\/doi.org\/10.1117\/12.2293971","DOI":"10.1117\/12.2293971"},{"key":"e_1_2_7_25_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-68612-7_71"},{"key":"e_1_2_7_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2018.09.013"},{"key":"e_1_2_7_27_1","first-page":"2234","volume-title":"Advances in Neural Information Processing Systems 29","author":"Salimans T","year":"2016"},{"key":"e_1_2_7_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/0169-7439(87)80084-9"},{"key":"e_1_2_7_29_1","doi-asserted-by":"publisher","DOI":"10.1120\/jacmp.v17i1.5888"},{"key":"e_1_2_7_30_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/50\/24\/009"},{"key":"e_1_2_7_31_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3481506"},{"key":"e_1_2_7_32_1","doi-asserted-by":"publisher","DOI":"10.1006\/cviu.1995.1004"},{"key":"e_1_2_7_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/0262-8856(94)90060-4"},{"key":"e_1_2_7_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2004.07.677"},{"key":"e_1_2_7_35_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3301594"},{"key":"e_1_2_7_36_1","unstructured":"MilletariF NavabN AhmadiS\u2010A.V\u2010Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation. ArXiv160604797 Cs. Published online June 15 2016. Accessed June 25 2020.http:\/\/arxiv.org\/abs\/1606.04797."},{"key":"e_1_2_7_37_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF02439713"},{"key":"e_1_2_7_38_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2013.01.002"},{"key":"e_1_2_7_39_1","unstructured":"KingmaDP AdamBJ.A Method for Stochastic Optimization. ArXiv14126980 Cs. Published online January 29 2017. Accessed June 25 2020.http:\/\/arxiv.org\/abs\/1412.6980."},{"key":"e_1_2_7_40_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13814"},{"key":"e_1_2_7_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2016.11.016"},{"key":"e_1_2_7_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2572683"},{"key":"e_1_2_7_43_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aa94ba"},{"key":"e_1_2_7_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ctro.2019.04.001"},{"key":"e_1_2_7_45_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12406"},{"key":"e_1_2_7_46_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13090"},{"key":"e_1_2_7_47_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2013.03.003"},{"key":"e_1_2_7_48_1","unstructured":"YangJ ZhangY ZhangL DongL.Automatic Segmentation of Parotids from CT Scans Using Multiple Atlases.Med Image Anal Clin Gd Chall. Published online January 1 2010."},{"key":"e_1_2_7_49_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2008.05.008"},{"key":"e_1_2_7_50_1","first-page":"1","article-title":"Retrospective validation and clinical implementation of automated contouring of organs at risk in the head and neck: a step toward automated radiation treatment planning for low\u2010 and middle\u2010income countries","volume":"4","author":"McCarroll RE","year":"2018","journal-title":"J Glob Oncol"},{"key":"e_1_2_7_51_1","unstructured":"BowlesC ChenL GuerreroR et al.GAN Augmentation: Augmenting Training Data using Generative Adversarial Networks. ArXiv181010863 Cs. Published online October 25 2018. Accessed November 11 2020.http:\/\/arxiv.org\/abs\/1810.10863."},{"key":"e_1_2_7_52_1","doi-asserted-by":"publisher","DOI":"10.1109\/34.927467"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14728","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.14728","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14728","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,29]],"date-time":"2023-08-29T21:19:56Z","timestamp":1693343996000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.14728"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,2,19]]},"references-count":51,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2021,4]]}},"alternative-id":["10.1002\/mp.14728"],"URL":"https:\/\/doi.org\/10.1002\/mp.14728","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,2,19]]},"assertion":[{"value":"2020-08-31","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-01-13","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-02-19","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}