{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,5]],"date-time":"2026-06-05T21:38:07Z","timestamp":1780695487882,"version":"3.54.1"},"reference-count":22,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2020,12,1]],"date-time":"2020-12-01T00:00:00Z","timestamp":1606780800000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2020,12,1]],"date-time":"2020-12-01T00:00:00Z","timestamp":1606780800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100007210","name":"Varian Medical Systems","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007210","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Radiat Oncol"],"published-print":{"date-parts":[[2020,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n<jats:title>Background<\/jats:title>\n<jats:p>Deep learning-based delineation of organs-at-risk for radiotherapy purposes has been investigated to reduce the time-intensiveness and inter-\/intra-observer variability associated with manual delineation. We systematically evaluated ways to improve the performance and reliability of deep learning for organ-at-risk segmentation, with the salivary glands as the paradigm. Improving deep learning performance is clinically relevant with applications ranging from the initial contouring process, to on-line adaptive radiotherapy.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Methods<\/jats:title>\n<jats:p>Various experiments were designed: increasing the amount of training data (1) with original images, (2) with traditional data augmentation and (3) with domain-specific data augmentation; (4) the influence of data quality was tested by comparing training\/testing on clinical versus curated contours, (5) the effect of using several custom cost functions was explored, and (6) patient-specific Hounsfield unit windowing was applied during inference; lastly, (7) the effect of model ensembles was analyzed. Model performance was measured with geometric parameters and model reliability with those parameters\u2019 variance.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Results<\/jats:title>\n<jats:p>A positive effect was observed from increasing the (1) training set size, (2\/3) data augmentation, (6) patient-specific Hounsfield unit windowing and (7) model ensembles. The effects of the strategies on performance diminished when the base model performance was already \u2018high\u2019. The effect of combining all beneficial strategies was an increase in average S\u00f8rensen\u2013Dice coefficient of about 4% and 3% and a decrease in standard deviation of about 1% and 1% for the submandibular and parotid gland, respectively.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Conclusions<\/jats:title>\n<jats:p>A subset of the strategies that were investigated provided a positive effect on model performance and reliability. The clinical impact of such strategies would be an expected reduction in post-segmentation editing, which facilitates the adoption of deep learning for autonomous automated salivary gland segmentation.<\/jats:p>\n<\/jats:sec>","DOI":"10.1186\/s13014-020-01721-1","type":"journal-article","created":{"date-parts":[[2020,12,1]],"date-time":"2020-12-01T08:04:14Z","timestamp":1606809854000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["Strategies to improve deep learning-based salivary gland segmentation"],"prefix":"10.1186","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9177-8851","authenticated-orcid":false,"given":"Ward","family":"van Rooij","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Max","family":"Dahele","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hanne","family":"Nijhuis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Berend J.","family":"Slotman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wilko F.","family":"Verbakel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2020,12,1]]},"reference":[{"key":"1721_CR1","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1186\/1748717X-7-32","volume":"7","author":"CL Brouwer","year":"2012","unstructured":"Brouwer CL, Steenbakkers RJ, van den Heuvel E, et al. 3D variation in delineation of head and neck organs at risk. Radiat Oncol. 2012;7:32. https:\/\/doi.org\/10.1186\/1748717X-7-32.","journal-title":"Radiat Oncol"},{"issue":"1","key":"1721_CR2","doi-asserted-by":"publisher","first-page":"83","DOI":"10.1016\/j.radonc.2015.07.041","volume":"117","author":"CL Brouwer","year":"2015","unstructured":"Brouwer CL, Steenbakkers RJ, Bourhis J, et al. CT-based delineation of organs at risk in the head and neck region: DAHANCA, EORTC, GORTEC, HKNPCSG, NCIC CTG, NCRI, NRG Oncology and TROG consensus guidelines. Radiother Oncol. 2015;117(1):83\u201390. https:\/\/doi.org\/10.1016\/j.radonc.2015.07.041.","journal-title":"Radiother Oncol"},{"key":"1721_CR3","unstructured":"Nikolov S, Blackwell S, Mendes R, et al. Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy. 2018. arXiv:1809.04430v1."},{"issue":"30","key":"1721_CR4","doi-asserted-by":"publisher","first-page":"S58","DOI":"10.1016\/j.ijrobp.2009.06.090","volume":"76","author":"JO Deasy","year":"2010","unstructured":"Deasy JO, Moiseenko V, Marks L, et al. Radiotherapy dose-volume effects on salivary gland function. Int J Radiat Oncol Biol Phys. 2010;76(30):S58\u201363. https:\/\/doi.org\/10.1016\/j.ijrobp.2009.06.090.","journal-title":"Int J Radiat Oncol Biol Phys"},{"issue":"3","key":"1721_CR5","doi-asserted-by":"publisher","first-page":"677","DOI":"10.1016\/j.ijrobp.2019.02.040","volume":"104","author":"W Van Rooij","year":"2019","unstructured":"Van Rooij W, Dahele M, Ribeiro Brandao H, et al. Deep learning-based delineation of head and neck organs-at-risk: geometric and dosimetric evaluation. Int J Radiat Oncol Biol Phys. 2019;104(3):677\u201384. https:\/\/doi.org\/10.1016\/j.ijrobp.2019.02.040.","journal-title":"Int J Radiat Oncol Biol Phys"},{"issue":"3","key":"1721_CR6","doi-asserted-by":"publisher","first-page":"391","DOI":"10.1016\/j.tripleo.2005.10.063","volume":"102","author":"B Stimec","year":"2006","unstructured":"Stimec B, Nikolic S, Rakocevic Z, et al. Symmetry of the submandibular glands in humans: a postmortem study assessing the linear morphometric parameters. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;102(3):391\u20134.","journal-title":"Oral Surg Oral Med Oral Pathol Oral Radiol Endod"},{"key":"1721_CR7","doi-asserted-by":"crossref","unstructured":"Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. 2014. arXiv:1411.4038.","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"1721_CR8","doi-asserted-by":"crossref","unstructured":"Cicek O, Abdulkadir A, Lienkamp SS, et al. 3D U-Net: learning dense volumetric segmentation from sparse annotation. 2016. arXiv:1606.06650v1.","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"1721_CR9","first-page":"1929","volume":"15","author":"N Srivastava","year":"2014","unstructured":"Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res. 2014;15:1929\u201358.","journal-title":"J Mach Learn Res"},{"key":"1721_CR10","unstructured":"Kingma DP, Ba J. Adam: a method for stochastic optimization. 2015. arXiv:1412.6980v9."},{"key":"1721_CR11","unstructured":"Hestness J, Narang S, Ardalani N, et al. Deep learning scaling is predictable, empirically. 2017. arXiv:1712.00409v1."},{"key":"1721_CR12","first-page":"979","volume":"2017","author":"Z Hussain","year":"2017","unstructured":"Hussain Z, Gimenez F, Yi D, et al. Differential data augmentation techniques for medical imaging classification tasks. AMIA Annu Symp Proc. 2017;2017:979\u201384.","journal-title":"AMIA Annu Symp Proc"},{"key":"1721_CR13","doi-asserted-by":"publisher","unstructured":"Simard PY, Steinkraus D, Platt JC. Best practices for convolutional neural networks applied to visual document analysis. 2003. In: Seventh international conference on document analysis and recognition: proceedings. https:\/\/doi.org\/10.1109\/ICDAR.2003.1227801.","DOI":"10.1109\/ICDAR.2003.1227801"},{"key":"1721_CR14","doi-asserted-by":"publisher","first-page":"190","DOI":"10.1016\/j.radonc.2018.03.012","volume":"127","author":"AR Delaney","year":"2018","unstructured":"Delaney AR, Dahele M, Slotman BJ, et al. Is accurate contouring of salivatory and swallowing structures necessary to spare them in head and neck VMAT plans? Radiother Oncol. 2018;127:190\u20136. https:\/\/doi.org\/10.1016\/j.radonc.2018.03.012.","journal-title":"Radiother Oncol"},{"issue":"10","key":"1721_CR15","doi-asserted-by":"publisher","first-page":"993","DOI":"10.1109\/34.58871","volume":"12","author":"LK Hansen","year":"1990","unstructured":"Hansen LK, Salamon P. Neural network ensembles. Trans Pattern Anal Mach Intell. 1990;12(10):993\u20131001. https:\/\/doi.org\/10.1109\/34.58871.","journal-title":"Trans Pattern Anal Mach Intell"},{"key":"1721_CR16","doi-asserted-by":"publisher","DOI":"10.1117\/1.JMI.6.1.011005","author":"A H\u00e4nsch","year":"2018","unstructured":"H\u00e4nsch A, Schwier M, Gass T, et al. Evaluation of deep learning methods for parotid gland segmentation. J Med Imaging. 2018. https:\/\/doi.org\/10.1117\/1.JMI.6.1.011005.","journal-title":"J Med Imaging"},{"key":"1721_CR17","unstructured":"Goodfellow IJ, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks. 2014. arXiv:1406.266."},{"key":"1721_CR18","doi-asserted-by":"crossref","unstructured":"Frid-Adar M, Klang E, Amitai M, et al. Synthetic data augmentation using GAN for improved liver lesion classification. 2018. arXiv:1801.02385.","DOI":"10.1109\/ISBI.2018.8363576"},{"key":"1721_CR19","unstructured":"Brainlesion: glioma, multiple sclerosis, stroke, traumatic brain injuries, p. 462."},{"key":"1721_CR20","doi-asserted-by":"crossref","unstructured":"Tajbakhsh N, Jeyaseelan L, Li Q, et al. Embracing imperfect datasets: a review of deep learning solutions for medical image segmentation. 2019. arXiv:1908.10454v1.","DOI":"10.1016\/j.media.2020.101693"},{"key":"1721_CR21","doi-asserted-by":"publisher","first-page":"S281","DOI":"10.1016\/S0167-8140(18)30840-5","volume":"127","author":"A H\u00e4nsch","year":"2018","unstructured":"H\u00e4nsch A, Gass T, Morgas T, et al. Parotid gland segmentation with deep learning using clinical vs. curated training data. Radiother Oncol. 2018;127:S281\u20132. https:\/\/doi.org\/10.1016\/S0167-8140(18)30840-5.","journal-title":"Radiother Oncol"},{"key":"1721_CR22","unstructured":"Kervadec H, Bouchtiba J, Desrosiers C, et al. Boundary loss for highly unbalanced segmentation. 2019. arXiv:1907.08612."}],"container-title":["Radiation Oncology"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/link.springer.com\/content\/pdf\/10.1186\/s13014-020-01721-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/link.springer.com\/article\/10.1186\/s13014-020-01721-1\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/link.springer.com\/content\/pdf\/10.1186\/s13014-020-01721-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2020,12,1]],"date-time":"2020-12-01T08:33:08Z","timestamp":1606811588000},"score":1,"resource":{"primary":{"URL":"https:\/\/ro-journal.biomedcentral.com\/articles\/10.1186\/s13014-020-01721-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,12]]},"references-count":22,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,12]]}},"alternative-id":["1721"],"URL":"https:\/\/doi.org\/10.1186\/s13014-020-01721-1","relation":{},"ISSN":["1748-717X"],"issn-type":[{"value":"1748-717X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,12]]},"assertion":[{"value":"2 June 2020","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 November 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 December 2020","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"On 16 November 2018 the Medical Ethics Review Committee of VU University Medical Center decided that the Medical Research Involving Human Subjects Act does not apply to our research. The Medical Ethics Review Committee of VU University Medical Center is registered with the US Office for Human Research Protections (OHRP) as IRB00002991. The FWA number assigned to VU University Medical Center is FWA00017598.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"On 16 November 2018 the Medical Ethics Review Committee of VU University Medical Center decided that the Medical Research Involving Human Subjects Act does not apply to our research. The Medical Ethics Review Committee of VU University Medical Center is registered with the US Office for Human Research Protections (OHRP) as IRB00002991. The FWA number assigned to VU University Medical Center is FWA00017598.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The Department of Radiation Oncology has a research collaboration with Varian Medical Systems, Palo Alto, CA, and Dahele, Slotman and Verbakel have received honoraria\/travel support from Varian Medical Systems.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"272"}}