{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T21:40:34Z","timestamp":1788990034781,"version":"build-2803163510"},"reference-count":55,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2021,10,21]],"date-time":"2021-10-21T00:00:00Z","timestamp":1634774400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012659","name":"Foundation for Innovative Research Groups of the National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100012659","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100017607","name":"Shenzhen Fundamental Research Program","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100017607","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Radio"],"abstract":"<jats:p>Glioma is a type of severe brain tumor, and its accurate segmentation is useful in surgery planning and progression evaluation. Based on different biological properties, the glioma can be divided into three partially-overlapping regions of interest, including whole tumor (WT), tumor core (TC), and enhancing tumor (ET). Recently, UNet has identified its effectiveness in automatically segmenting brain tumor from multi-modal magnetic resonance (MR) images. In this work, instead of network architecture, we focus on making use of prior knowledge (brain parcellation), training and testing strategy (joint 3D+2D), ensemble and post-processing to improve the brain tumor segmentation performance. We explore the accuracy of three UNets with different inputs, and then ensemble the corresponding three outputs, followed by post-processing to achieve the final segmentation. Similar to most existing works, the first UNet uses 3D patches of multi-modal MR images as the input. The second UNet uses brain parcellation as an additional input. And the third UNet is inputted by 2D slices of multi-modal MR images, brain parcellation, and probability maps of WT, TC, and ET obtained from the second UNet. Then, we sequentially unify the WT segmentation from the third UNet and the fused TC and ET segmentation from the first and the second UNets as the complete tumor segmentation. Finally, we adopt a post-processing strategy by labeling small ET as non-enhancing tumor to correct some false-positive ET segmentation. On one publicly-available challenge validation dataset (BraTS2018), the proposed segmentation pipeline yielded average Dice scores of 91.03\/86.44\/80.58% and average 95% Hausdorff distances of 3.76\/6.73\/2.51 mm for WT\/TC\/ET, exhibiting superior segmentation performance over other state-of-the-art methods. We then evaluated the proposed method on the BraTS2020 training data through five-fold cross validation, with similar performance having also been observed. The proposed method was finally evaluated on 10 in-house data, the effectiveness of which has been established qualitatively by professional radiologists.<\/jats:p>","DOI":"10.3389\/fradi.2021.704888","type":"journal-article","created":{"date-parts":[[2021,10,21]],"date-time":"2021-10-21T03:22:30Z","timestamp":1634786550000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":46,"title":["Brain Tumor Segmentation From Multi-Modal MR Images via Ensembling UNets"],"prefix":"10.3389","volume":"1","author":[{"given":"Yue","family":"Zhang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pinyuan","family":"Zhong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dabin","family":"Jie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiewei","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shanmei","family":"Zeng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianping","family":"Chu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yilong","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ed X.","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoying","family":"Tang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2021,10,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"803","DOI":"10.1007\/s00401-016-1545-1","article-title":"The 2016 World Health Organization classification of tumors of the central nervous system: a summary","volume":"131","author":"Louis","year":"2016","journal-title":"Acta Neuropathol"},{"key":"B2","doi-asserted-by":"publisher","first-page":"v1","DOI":"10.1093\/neuonc\/now207","article-title":"CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2009-2013","volume":"18","author":"Ostrom","year":"2016","journal-title":"Neuro Oncol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"170117","DOI":"10.1038\/sdata.2017.117","article-title":"Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features","volume":"4","author":"Bakas","year":"2017","journal-title":"Sci Data"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","article-title":"The multimodal brain tumor image segmentation benchmark (BRATS)","volume":"34","author":"Menze","year":"2014","journal-title":"IEEE Trans Med Imaging"},{"key":"B5","article-title":"Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge","author":"Bakas","year":"2018","journal-title":"arXiv preprint arXiv:181102629."},{"key":"B6","first-page":"286","article-title":"Segmentation labels and radiomic features for the pre-operative scans of the TCGA-LGG collection","author":"Bakas","year":"2017","journal-title":"Cancer Imaging Arch"},{"key":"B7","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1016\/j.compmedimag.2014.10.009","article-title":"3D multimodal MRI brain glioma tumor and edema segmentation: a graph cut distribution matching approach","volume":"40","author":"Njeh","year":"2015","journal-title":"Comput Med Imaging Graph"},{"key":"B8","doi-asserted-by":"crossref","first-page":"5928","DOI":"10.1109\/EMBC.2019.8856668","article-title":"Simultaneous tissue classification and lateral ventricle segmentation via a 2D U-net driven by a 3D fully convolutional neural network","volume-title":"2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)","author":"Wu","year":"2019"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1449","DOI":"10.3389\/fnins.2019.01449","article-title":"Improving patch-based convolutional neural networks for MRI brain tumor segmentation by leveraging location information","volume":"13","author":"Kao","year":"2020","journal-title":"Front Neurosci"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1116","DOI":"10.1109\/TMI.2018.2878669","article-title":"HyperDense-Net: a hyper-densely connected CNN for multi-modal image segmentation","volume":"38","author":"Dolz","year":"2018","journal-title":"IEEE Trans Med Imaging"},{"key":"B11","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1109\/RBME.2019.2946868","article-title":"Automated brain tumor segmentation using multimodal brain scans: a survey based on models submitted to the BraTS 2012-2018 challenges","volume":"13","author":"Ghaffari","year":"2019","journal-title":"IEEE Rev Biomed Eng"},{"key":"B12","doi-asserted-by":"crossref","first-page":"234","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"U-net: Convolutional networks for biomedical image segmentation","volume-title":"Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2015","author":"Ronneberger","year":"2015"},{"key":"B13","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1038\/s41592-018-0261-2","article-title":"U-Net: deep learning for cell counting, detection, and morphometry","volume":"16","author":"Falk","year":"2019","journal-title":"Nat Methods"},{"key":"B14","doi-asserted-by":"publisher","first-page":"101884","DOI":"10.1016\/j.media.2020.101884","article-title":"A deep learning framework for pancreas segmentation with multi-atlas registration and 3D level-set","volume":"68","author":"Zhang","year":"2021","journal-title":"Med Image Anal"},{"key":"B15","doi-asserted-by":"publisher","first-page":"121","DOI":"10.1109\/JBHI.2020.2986926","article-title":"Multi-scale self-guided attention for medical image segmentation","volume":"25","author":"Sinha","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"key":"B16","doi-asserted-by":"crossref","first-page":"433","DOI":"10.1007\/978-3-319-66185-8_49","article-title":"Boundary-aware fully convolutional network for brain tumor segmentation","volume-title":"Medical Image Computing and Computer-Assisted Intervention? MICCAI 2017","author":"Shen","year":"2017"},{"key":"B17","first-page":"168","article-title":"Brain tumor segmentation on multimodal MR imaging using multi-level upsampling in decoder","volume-title":"International MICCAI Brainlesion Workshop","author":"Hu","year":"2018"},{"key":"B18","first-page":"63","article-title":"Automatic brain tumor segmentation using a U-net neural network","volume-title":"International MICCAI Brainlesion Workshop","author":"Caver","year":"2018"},{"key":"B19","first-page":"154","article-title":"Segmentation of brain tumors using DeepLabv3+","volume-title":"International MICCAI Brainlesion Workshop","author":"Choudhury","year":"2018"},{"key":"B20","first-page":"456","article-title":"Ensembles of densely-connected CNNs with label-uncertainty for brain tumor segmentation","volume-title":"International MICCAI Brainlesion Workshop","author":"McKinley","year":"2018"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1911","DOI":"10.1109\/JBHI.2018.2874033","article-title":"Efficient brain tumor segmentation with multiscale two-pathway-group conventional neural networks","volume":"23","author":"Razzak","year":"2018","journal-title":"IEEE J Biomed Health Inform"},{"key":"B22","doi-asserted-by":"publisher","first-page":"810","DOI":"10.3389\/fnins.2019.00810","article-title":"Brain tumor segmentation and survival prediction using multimodal MRI scans with deep learning","volume":"13","author":"Sun","year":"2019","journal-title":"Front Neurosci"},{"key":"B23","first-page":"128","article-title":"Brain tumor segmentation and tractographic feature extraction from structural MR images for overall survival prediction","volume-title":"International MICCAI Brainlesion Workshop","author":"Kao","year":"2018"},{"key":"B24","doi-asserted-by":"publisher","first-page":"25","DOI":"10.3389\/fncom.2020.00025","article-title":"Brain tumor segmentation using an ensemble of 3D U-Nets and overall survival prediction using radiomic features","volume":"14","author":"Feng","year":"2020","journal-title":"Front Comput Neurosci"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2020.2998146","article-title":"HDC-Net: hierarchical decoupled convolution network for brain tumor segmentation","author":"Luo","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"key":"B26","first-page":"234","article-title":"No new-net","volume-title":"International MICCAI Brainlesion Workshop","author":"Isensee","year":"2018"},{"key":"B27","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1007\/s13748-021-00232-8","article-title":"Brain tumor segmentation via C-dense convolutional neural network","volume":"10","author":"Wang","year":"2021","journal-title":"Prog Artif Intell"},{"key":"B28","first-page":"311","article-title":"3D MRI brain tumor segmentation using autoencoder regularization","volume-title":"International MICCAI Brainlesion Workshop","author":"Myronenko","year":"2018"},{"key":"B29","doi-asserted-by":"publisher","first-page":"101746","DOI":"10.1016\/j.media.2020.101746","article-title":"Rubik's Cube+: a self-supervised feature learning framework for 3D medical image analysis","volume":"64","author":"Zhu","year":"2020","journal-title":"Med Image Anal"},{"key":"B30","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1016\/j.media.2015.08.001","article-title":"Automatic classification of pulmonary peri-fissural nodules in computed tomography using an ensemble of 2D views and a convolutional neural network out-of-the-box","volume":"26","author":"Ciompi","year":"2015","journal-title":"Med Image Anal"},{"key":"B31","doi-asserted-by":"crossref","first-page":"693","DOI":"10.1007\/978-3-319-66182-7_79","article-title":"A fixed-point model for pancreas segmentation in abdominal CT scans","volume-title":"Medical Image Computing and Computer Assisted Intervention? MICCAI 2017","author":"Zhou","year":"2017"},{"key":"B32","doi-asserted-by":"crossref","first-page":"520","DOI":"10.1007\/978-3-319-10404-1_65","article-title":"A new 2.5 D representation for lymph node detection using random sets of deep convolutional neural network observations","volume-title":"Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2014","author":"Roth","year":"2014"},{"key":"B33","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","author":"Isensee","year":"2020","journal-title":"Nat Methods"},{"key":"B34","doi-asserted-by":"publisher","first-page":"101641","DOI":"10.1016\/j.bspc.2019.101641","article-title":"RescueNet: An unpaired GAN for brain tumor segmentation","volume":"55","author":"Nema","year":"2020","journal-title":"Biomed Signal Process Control"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-87193-2_11","article-title":"TransBTS: multimodal brain tumor segmentation using transformer","author":"Wang","year":"2021","journal-title":"arXiv preprint arXiv:210304430"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1101\/2021.07.03.450931","article-title":"Accurate automatic glioma segmentation in brain MRI images based on CapsNet","author":"Aziz","year":"2021","journal-title":"bioRxiv"},{"key":"B37","doi-asserted-by":"publisher","first-page":"773","DOI":"10.1007\/978-3-319-10470-6_96","article-title":"Segmentation of cerebrovascular pathologies in stroke patients with spatial and shape priors","volume":"17","author":"Dalca","year":"2014","journal-title":"Med Image Comput Comput Assist Interv"},{"key":"B38","doi-asserted-by":"publisher","first-page":"107904","DOI":"10.1016\/j.patcog.2021.107904","article-title":"Brain segmentation based on multi-atlas and diffeomorphism guided 3D fully convolutional network ensembles","volume":"115","author":"Wu","year":"2021","journal-title":"Pattern Recogn"},{"key":"B39","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1016\/j.patrec.2017.11.016","article-title":"Automatic brain segmentation using artificial neural networks with shape context","volume":"101","author":"Mahbod","year":"2018","journal-title":"Pattern Recogn Lett"},{"key":"B40","doi-asserted-by":"publisher","first-page":"15","DOI":"10.3389\/fnins.2020.00015","article-title":"Shape information improves the cross-cohort performance of deep learning-based segmentation of the hippocampus","volume":"14","author":"Brusini","year":"2020","journal-title":"Front Neurosci"},{"key":"B41","doi-asserted-by":"publisher","first-page":"526","DOI":"10.1109\/JBHI.2020.2996783","article-title":"MI-UNet: Multi-inputs UNet incorporating brain parcellation for stroke lesion segmentation from T1-weighted magnetic resonance images","volume":"25","author":"Zhang","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"key":"B42","doi-asserted-by":"publisher","first-page":"2663","DOI":"10.1109\/TMI.2018.2845918","article-title":"H-DenseUNet: hybrid densely connected UNet for liver and tumor segmentation from CT volumes","volume":"37","author":"Li","year":"2018","journal-title":"IEEE Trans Med Imaging"},{"key":"B43","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1007\/978-3-030-32245-8_13","article-title":"HD-net: Hybrid discriminative network for prostate segmentation in MR images","volume-title":"Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI 2019","author":"Jia","year":"2019"},{"key":"B44","first-page":"301","article-title":"A joint 3D+2D fully convolutional framework for subcortical segmentation","volume-title":"Med. Image Comput. Comput. Assist. Interv","author":"Wu","year":"2019"},{"key":"B45","first-page":"402","article-title":"Ensembles of multiple scales, losses and models for brain tumor segmentation and overall survival time prediction task","volume-title":"International MICCAI Brainlesion Workshop","author":"Ren","year":"2018"},{"key":"B46","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1016\/j.ins.2015.10.018","article-title":"Single seed delineation of brain tumor using multi-thresholding","volume":"330","author":"Banerjee","year":"2016","journal-title":"Inf Syst"},{"key":"B47","first-page":"231","article-title":"Two-stage cascaded U-Net: 1st place solution to brats challenge 2019 segmentation task","volume-title":"International MICCAI Brainlesion Workshop","author":"Jiang","year":"2019"},{"key":"B48","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-72087-2_11","article-title":"nnU-Net for brain tumor segmentation","author":"Isensee","year":"2020","journal-title":"arXiv preprint arXiv:201100848"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-72087-2_6","article-title":"H2NF-Net for brain tumor segmentation using multimodal MR imaging: 2nd place solution to BraTS challenge 2020","author":"Jia","year":"2020","journal-title":"Segmentation Task"},{"key":"B50","doi-asserted-by":"publisher","first-page":"101828","DOI":"10.1016\/j.compmedimag.2020.101828","article-title":"Analyzing magnetic resonance imaging data from glioma patients using deep learning","volume":"88","author":"Menze","year":"2021","journal-title":"Comput Med Imaging Graph"},{"key":"B51","doi-asserted-by":"publisher","first-page":"S173","DOI":"10.1016\/j.neuroimage.2008.10.055","article-title":"Bayesian analysis of neuroimaging data in FSL","volume":"45","author":"Woolrich","year":"2009","journal-title":"Neuroimage"},{"key":"B52","doi-asserted-by":"publisher","first-page":"S208","DOI":"10.1016\/j.neuroimage.2004.07.051","article-title":"Advances in functional and structural MR image analysis and implementation as FSL","volume":"23","author":"Smith","year":"2004","journal-title":"Neuroimage"},{"key":"B53","doi-asserted-by":"publisher","first-page":"782","DOI":"10.1016\/j.neuroimage.2011.09.015","article-title":"FSL","volume":"62","author":"Jenkinson","year":"2012","journal-title":"Neuroimage"},{"key":"B54","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1007\/s11060-018-03025-7","article-title":"Comparative analysis of the diffusion kurtosis imaging and diffusion tensor imaging in grading gliomas, predicting tumour cell proliferation and IDH-1 gene mutation status","volume":"141","author":"Zhao","year":"2019","journal-title":"J Neurooncol"},{"key":"B55","first-page":"3","article-title":"Current trends in the surgical management and treatment of adult glioblastoma","author":"Young","year":"2015","journal-title":"Ann Transl Med"}],"container-title":["Frontiers in Radiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2021.704888\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,10,21]],"date-time":"2021-10-21T03:22:47Z","timestamp":1634786567000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2021.704888\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,10,21]]},"references-count":55,"alternative-id":["10.3389\/fradi.2021.704888"],"URL":"https:\/\/doi.org\/10.3389\/fradi.2021.704888","relation":{},"ISSN":["2673-8740"],"issn-type":[{"value":"2673-8740","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,10,21]]},"article-number":"704888"}}