{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T16:01:12Z","timestamp":1783958472717,"version":"3.55.0"},"reference-count":27,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2022,11,11]],"date-time":"2022-11-11T00:00:00Z","timestamp":1668124800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Oncol."],"abstract":"<jats:p>Nasopharyngeal carcinoma (NPC) is primarily treated with radiation therapy. Accurate delineation of target volumes and organs at risk is important. However, manual delineation is time-consuming, variable, and subjective depending on the experience of the radiation oncologist. This work explores the use of deep learning methods to automate the segmentation of NPC primary gross tumor volume (GTVp) in planning computer tomography (CT) images. A total of sixty-three (63) patients diagnosed with NPC were included in this study. Although a number of studies applied have shown the effectiveness of deep learning methods in medical imaging, their high performance has mainly been due to the wide availability of data. In contrast, the data for NPC is scarce and inaccessible. To tackle this problem, we propose two sequential approaches. First we propose a much simpler architecture which follows the UNet design but using 2D convolutional network for 3D segmentation. We find that this specific architecture is much more effective in the segmentation of GTV in NPC. We highlight its efficacy over other more popular and modern architecture by achieving significantly higher performance. Moreover to further improve performance, we trained the model using multi-scale dataset to create an ensemble of models. However,  the performance of the model is ultimately dependent on the availability of labelled data. Hence building on top of this proposed architecture, we employ the use of semi-supervised learning by proposing the use of a combined pre-text tasks. Specifically we use the combination of 3D rotation and 3D relative-patch location pre-texts tasks to pretrain the feature extractor. We use an additional 50 CT images of healthy patients which have no annotation or labels. By semi-supervised pretraining the feature extractor can be frozen after pretraining which essentially makes it much more efficient in terms of the number of parameters since only the decoder is trained. Finally it is not only efficient in terms of parameters but also data, which is shown when the pretrained model with only portion of the labelled training data was able to achieve very close performance to the model trained with the full labelled data.<\/jats:p>","DOI":"10.3389\/fonc.2022.980312","type":"journal-article","created":{"date-parts":[[2022,11,11]],"date-time":"2022-11-11T06:02:48Z","timestamp":1668146568000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":9,"title":["Automatic segmentation of nasopharyngeal carcinoma on CT images using efficient UNet\u20102.5D ensemble with semi\u2010supervised pretext task pretraining"],"prefix":"10.3389","volume":"12","author":[{"given":"Jansen Keith L.","family":"Domoguen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jen-Jen A.","family":"Manuel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Johanna Patricia A.","family":"Ca\u00f1al","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"suffix":"Jr","given":"Prospero C.","family":"Naval","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2022,11,11]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.32113\/wcrj_20183_1046","article-title":"Nasopharyngeal cancer in the world: Epidemiology, incidence, mortality and risk factors","volume":"5","author":"Salehiniya","year":"2018","journal-title":"World Cancer Res J"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1533033819884561","DOI":"10.1177\/1533033819884561","article-title":"The tumor target segmentation of nasopharyngeal cancer in ct images based on deep learning methods","volume":"18","author":"Li","year":"2019","journal-title":"Technol Cancer Res Treat"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2022.827991","article-title":"Multiscale local enhancement deep convolutional networks for the automated 3d segmentation of gross tumor volumes in nasopharyngeal carcinoma: A multi-institutional dataset study","volume":"12","author":"Yang","year":"2022","journal-title":"Front Oncol"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2020.06.146","article-title":"Automatic segmentation of gross target volume of nasopharynx cancer using ensemble of multiscale deep neural networks with spatial attention","volume":"438","author":"Mei","year":"2021","journal-title":"Neurocomputing"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2017.00315","article-title":"Deep deconvolutional neural network for target segmentation of nasopharyngeal cancer in planning computed tomography images","volume":"7","author":"Men","year":"2017","journal-title":"Front Oncol 7"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2020.00166","article-title":"Fully-automated segmentation of nasopharyngeal carcinoma on dual-sequence mri using convolutional neural networks","volume":"10","author":"Ye","year":"2020","journal-title":"Front Oncol"},{"key":"B7","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2018.00745","article-title":"Squeeze-and-excitation networks","author":"Hu","year":"2018"},{"key":"B8","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-66185-8_29","article-title":"Semi-supervised Learning for Network-Based Cardiac MR Image Segmentation","volume-title":"Medical Image Computing and Computer-Assisted Intervention \u2212 MICCAI 2017: 20th International Conference","author":"Bai","year":""},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-020-03407-7","article-title":"Dnetunet: a semi-supervised cnn of medical image segmentation for super-computing ai service","volume":"77","author":"Tseng","year":"2021","journal-title":"J Supercomputing"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2016.09.030","article-title":"Semi-supervised learning and graph cuts for consensus based medical image segmentation","volume":"63","author":"Mahapatra","year":"2017","journal-title":"Pattern recognition"},{"key":"B11","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-32226-7_90","article-title":"Semi-supervised medical image segmentation via learning consistency under transformations","volume-title":"International conference on medical image computing and computer-assisted intervention","author":"Bortsova","year":"2019"},{"key":"B12","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-32248-9_51","article-title":"Multi-task attention-based\u00b4 semi-supervised learning for medical image segmentation","volume-title":"International conference on medical image computing and computer-assisted intervention","author":"Chen","year":"2019"},{"key":"B13","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR42600.2020.00401","article-title":"Focalmix: Semi-supervised learning for 3d medical image detection","author":"Wang","year":"2020"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.03.009","article-title":"Not-so-supervised: A survey of semi-supervised, multi-instance, and transfer learning in medical image analysis","volume":"54","author":"Cheplygina","year":"2019","journal-title":"Med image Anal"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2006.03829","article-title":"3d self-supervised methods for medical imaging","volume":"33","author":"Taleb","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"B16","article-title":"A simple framework for contrastive learning of visual representations","volume-title":"International conference on machine learning","author":"Chen","year":"2020"},{"key":"B17","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-87193-2_36","volume-title":"A Spatial Guided Self-supervised Clustering Network for Medical Image Segmentation","author":"Ahn","year":""},{"key":"B18","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-87196-3_4","volume-title":"Imbalance-Aware Self-supervised Learning for 3D Radiomic Representations","author":"Li","year":""},{"key":"B19","volume-title":"3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation","author":"\u00c7i\u00e7ek","year":""},{"key":"B20","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","volume-title":"International conference on machine learning","author":"Ioffe","year":"2015"},{"key":"B21","article-title":"Empirical evaluation of rectified activations in convolutional network","author":"Xu","year":"2015"},{"key":"B22","doi-asserted-by":"publisher","first-page":"199","DOI":"10.1136\/jamia.1997.0040199","article-title":"Understanding and using dicom, the data interchange standard for biomedical imaging","volume":"4","author":"Bidgood","year":"1997","journal-title":"J Am Med Inf Assoc"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2021.704850","article-title":"Multi-task deep supervision on attention r2u-net for brain tumor segmentation","volume":"11","author":"Ma","year":"2021","journal-title":"Front Oncol"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12593","article-title":"Esophagus segmentation in ct via 3d fully convolutional neural network and random walk","volume":"44","author":"Fechter","year":"2017","journal-title":"Med Phys"},{"key":"B25","doi-asserted-by":"publisher","first-page":"6980","DOI":"10.48550\/arXiv.1412.6980","article-title":"Adam: A method for stochastic optimization","volume":"1412","author":"Kingma","year":"2014","journal-title":"arXiv preprint arXiv"},{"key":"B26","volume-title":"Models genesis: Generic autodidactic models for 3d medical image analysis","author":"Zhou","year":"2019"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01008-z","article-title":"Nnu-net: A self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat Methods"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2022.980312\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,11,11]],"date-time":"2022-11-11T06:02:56Z","timestamp":1668146576000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2022.980312\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,11]]},"references-count":27,"alternative-id":["10.3389\/fonc.2022.980312"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2022.980312","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,11,11]]},"article-number":"980312"}}