{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T02:17:48Z","timestamp":1785291468103,"version":"3.55.0"},"reference-count":34,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,9,21]],"date-time":"2023-09-21T00:00:00Z","timestamp":1695254400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,9,21]],"date-time":"2023-09-21T00:00:00Z","timestamp":1695254400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Li Kashing Foundation Cross-Disciplinary Research Program","award":["2020LKSFG12B"],"award-info":[{"award-number":["2020LKSFG12B"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Imaging"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Cervical cell segmentation is a fundamental step in automated cervical cancer cytology screening. The aim of this study was to develop and evaluate a deep ensemble model for cervical cell segmentation including both cytoplasm and nucleus segmentation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The Cx22 dataset was used to develop the automated cervical cell segmentation algorithm. The U-Net, U-Net\u2009+\u2009\u2009+\u2009, DeepLabV3, DeepLabV3Plus, Transunet, and Segformer were used as candidate model architectures, and each of the first four architectures adopted two different encoders choosing from resnet34, resnet50 and denseNet121. Models were trained under two settings: trained from scratch, encoders initialized from ImageNet pre-trained models and then all layers were fine-tuned. For every segmentation task, four models were chosen as base models, and Unweighted average was adopted as the model ensemble method.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      U-Net and U-Net\u2009+\u2009\u2009+\u2009with resnet34 and denseNet121 encoders trained using transfer learning consistently performed better than other models, so they were chosen as base models. The ensemble model obtained the Dice similarity coefficient, sensitivity, specificity of 0.9535 (95% CI:0.9534\u20130.9536), 0.9621 (0.9619\u20130.9622),0.9835 (0.9834\u20130.9836) and 0.7863 (0.7851\u20130.7876), 0.9581 (0.9573\u20130.959), 0.9961 (0.9961\u20130.9962) on cytoplasm segmentation and nucleus segmentation, respectively. The Dice, sensitivity, specificity of baseline models for cytoplasm segmentation and nucleus segmentation were 0.948, 0.954, 0.9823 and 0.750, 0.713, 0.9988, respectively. Except for the specificity of cytoplasm segmentation, all metrics outperformed the best baseline models (\n                      <jats:italic>P<\/jats:italic>\n                      \u2009&lt;\u20090.05) with a moderate margin.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed algorithm achieved better performances on cervical cell segmentation than baseline models. It can be potentially used in automated cervical cancer cytology screening system.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12880-023-01096-1","type":"journal-article","created":{"date-parts":[[2023,9,21]],"date-time":"2023-09-21T08:01:51Z","timestamp":1695283311000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":34,"title":["Automated cervical cell segmentation using deep ensemble learning"],"prefix":"10.1186","volume":"23","author":[{"given":"Jie","family":"Ji","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weifeng","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuejiao","family":"Dong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruilin","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yiqun","family":"Geng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liangli","family":"Hong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,9,21]]},"reference":[{"issue":"3","key":"1096_CR1","doi-asserted-by":"publisher","first-page":"209","DOI":"10.3322\/caac.21660","volume":"71","author":"H Sung","year":"2021","unstructured":"Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209\u201349.","journal-title":"CA Cancer J Clin."},{"issue":"1","key":"1096_CR2","doi-asserted-by":"publisher","first-page":"7","DOI":"10.3322\/caac.21590","volume":"70","author":"RL Siegel","year":"2020","unstructured":"Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CACancer J Clin. 2020;70(1):7\u201330.","journal-title":"CACancer J Clin."},{"key":"1096_CR3","doi-asserted-by":"crossref","unstructured":"Cen L-P, Ji J, Lin J-W, Ju S-T, Lin H-J, Li T-P, et al. Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks. Nat Commun. 2021;12(1):1\u201313.\u00a0https:\/\/www.nature.com\/articles\/s41467-021-25138-w.","DOI":"10.1038\/s41467-021-25138-w"},{"key":"1096_CR4","doi-asserted-by":"crossref","unstructured":"Wang J, Ji J, Zhang M, Lin J-W, Zhang G, Gong W, et al. Automated explainable multidimensional deep learning platform of retinal images for retinopathy of prematurity screening. JAMA Netw Open. 2021;4(5):1\u201312.\u00a0https:\/\/jamanetwork.com\/journals\/jamanetworkopen\/fullarticle\/2779454.","DOI":"10.1001\/jamanetworkopen.2021.8758"},{"key":"1096_CR5","doi-asserted-by":"crossref","unstructured":"Tang Y-W, Ji J, Lin J-W, Wang J, Wang Y, Liu Z, et al. Automatic detection of peripheral retinal lesions from ultrawide-field fundus images using deep learning. Asia Pac J Ophthalmol. 2023;12(3):284\u201392.\u00a0https:\/\/journals.lww.com\/apjoo\/fulltext\/2023\/05000\/automatic_detection_of_peripheral_retinal_lesions.4.aspx.","DOI":"10.1097\/APO.0000000000000599"},{"key":"1096_CR6","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.106194","volume":"150","author":"G Liu","year":"2022","unstructured":"Liu G, Ding Q, Luo H, Sha M, Li X, Ju M. Cx22: A new publicly available dataset for deep learning-based segmentation of cervical cytology images. Comput Biol Med. 2022;150: 106194.","journal-title":"Comput Biol Med"},{"key":"1096_CR7","doi-asserted-by":"crossref","unstructured":"Zhou Y, Chen H, Xu J, Dou Q, Heng P-A, editors. IRNet: Instance Relation Network for Overlapping Cervical Cell Segmentation. Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI 2019; 2019 2019\/\/; Cham: Springer International Publishing.","DOI":"10.1007\/978-3-030-32239-7_71"},{"key":"1096_CR8","doi-asserted-by":"publisher","first-page":"53709","DOI":"10.1109\/ACCESS.2018.2871153","volume":"6","author":"Y Liu","year":"2018","unstructured":"Liu Y, Zhang P, Song Q, Li A, Zhang P, Gui Z. Automatic segmentation of cervical nuclei based on deep learning and a conditional random field. IEEE Access. 2018;6:53709\u201321.","journal-title":"IEEE Access"},{"issue":"4","key":"1096_CR9","doi-asserted-by":"publisher","first-page":"1261","DOI":"10.1109\/TIP.2015.2389619","volume":"24","author":"Z Lu","year":"2015","unstructured":"Lu Z, Carneiro G, Bradley AP. An improved joint optimization of multiple level set functions for the segmentation of overlapping cervical cells. IEEE Trans Image Process. 2015;24(4):1261\u201372.","journal-title":"IEEE Trans Image Process"},{"issue":"2","key":"1096_CR10","doi-asserted-by":"publisher","first-page":"441","DOI":"10.1109\/JBHI.2016.2519686","volume":"21","author":"Z Lu","year":"2017","unstructured":"Lu Z, Carneiro G, Bradley AP, Ushizima D, Nosrati MS, Bianchi AGC, et al. Evaluation of three algorithms for the segmentation of overlapping cervical cells. IEEE J Biomed Health Inform. 2017;21(2):441\u201350.","journal-title":"IEEE J Biomed Health Inform"},{"key":"1096_CR11","doi-asserted-by":"crossref","unstructured":"Plissiti ME, Dimitrakopoulos P, Sfikas G, Nikou C, Krikoni O, Charchanti A, editors. Sipakmed: a new dataset for feature and image based classification of normal and pathological cervical cells in Pap Smear Images. 2018 25th IEEE International Conference on Image Processing (ICIP); 2018 7\u201310 Oct. 2018.","DOI":"10.1109\/ICIP.2018.8451588"},{"key":"1096_CR12","doi-asserted-by":"crossref","unstructured":"Rezende MT, Silva R, Bernardo FdO, Tobias AHG, Oliveira PHC, Machado TM, et al. Cric searchable image database as a public platform for conventional pap smear cytology data. Sci Data. 2021;8(1):151.","DOI":"10.1038\/s41597-021-00933-8"},{"key":"1096_CR13","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2021.104649","volume":"136","author":"MM Rahaman","year":"2021","unstructured":"Rahaman MM, Li C, Yao Y, Kulwa F, Wu X, Li X, et al. DeepCervix: a deep learning-based framework for the classification of cervical cells using hybrid deep feature fusion techniques. Comput Biol Med. 2021;136: 104649.","journal-title":"Comput Biol Med"},{"key":"1096_CR14","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.348","volume":"7","author":"AR Bhatt","year":"2021","unstructured":"Bhatt AR, Ganatra A, Kotecha K. Cervical cancer detection in pap smear whole slide images using convNet with transfer learning and progressive resizing. PeerJ Computer Science. 2021;7: e348.","journal-title":"PeerJ Computer Science"},{"issue":"2","key":"1096_CR15","doi-asserted-by":"publisher","first-page":"121","DOI":"10.1159\/000381842","volume":"59","author":"R Nayar","year":"2015","unstructured":"Nayar R, Wilbur DC. The Pap test and Bethesda 2014. Acta Cytol. 2015;59(2):121\u201332.","journal-title":"Acta Cytol"},{"key":"1096_CR16","doi-asserted-by":"publisher","first-page":"223","DOI":"10.1016\/j.compbiomed.2019.03.011","volume":"108","author":"J Zhang","year":"2019","unstructured":"Zhang J, Liu Z, Du B, He J, Li G, Chen D. Binary tree-like network with two-path Fusion Attention Feature for cervical cell nucleus segmentation. Comput Biol Med. 2019;108:223\u201333.","journal-title":"Comput Biol Med"},{"key":"1096_CR17","unstructured":"Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. ArXiv e-prints [Internet]. 2015; 1505. Available from: http:\/\/adsabs.harvard.edu\/abs\/2015arXiv150504597R.\u00a0Accessed 1 Apr 2023."},{"key":"1096_CR18","unstructured":"Zhou Z, Mahfuzur Rahman Siddiquee M, Tajbakhsh N, Liang J. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. ArXiv e-prints [Internet]. 2018; 1807. Available from: http:\/\/adsabs.harvard.edu\/abs\/2018arXiv180710165Z.\u00a0Accessed 15 Apr 2023."},{"key":"1096_CR19","doi-asserted-by":"crossref","unstructured":"Huang H, Lin L, Tong R, Hu H, Zhang Q, Iwamoto Y, et al. UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation. 2020:[arXiv:2004.08790 p.]. Available from: https:\/\/ui.adsabs.harvard.edu\/abs\/2020arXiv200408790H.\u00a0Accessed 15 Apr 2023.","DOI":"10.1109\/ICASSP40776.2020.9053405"},{"key":"1096_CR20","doi-asserted-by":"publisher","first-page":"411","DOI":"10.1007\/978-1-4899-7488-4_293","volume-title":"Encyclopedia of Biometrics","author":"Z-H Zhou","year":"2015","unstructured":"Zhou Z-H. Ensemble Learning. In: Li SZ, Jain AK, editors. Encyclopedia of Biometrics. Boston, MA: Springer, US; 2015. p. 411\u20136."},{"key":"1096_CR21","unstructured":"Chen L-C, Papandreou G, Schroff F, Adam H. Rethinking atrous convolution for semantic image segmentation. ArXiv e-prints [Internet]. 2017; 1706. Available from: http:\/\/adsabs.harvard.edu\/abs\/2017arXiv170605587C.\u00a0Accessed 1 Apr 2023."},{"key":"1096_CR22","doi-asserted-by":"crossref","unstructured":"Chen L-C, Zhu Y, Papandreou G, Schroff F, Adam H. Encoder-Decoder with atrous separable convolution for semantic image segmentation. arXiv e-prints [Internet]. 2018. Available from: https:\/\/ui.adsabs.harvard.edu\/#abs\/2018arXiv180202611C.\u00a0Accessed 1 Apr 2023.","DOI":"10.1007\/978-3-030-01234-2_49"},{"key":"1096_CR23","unstructured":"Chen J, Lu Y, Yu Q, Luo X, Adeli E, Wang Y, et al. TransUNet: transformers make strong encoders for medical image segmentation. 2021 :[arXiv:2102.04306 p.]. Available from: https:\/\/ui.adsabs.harvard.edu\/abs\/2021arXiv210204306C.\u00a0Accessed 1 Apr 2023."},{"key":"1096_CR24","unstructured":"Xie E, Wang W, Yu Z, Anandkumar A, Alvarez JM, Luo P. SegFormer: Simple and efficient design for semantic segmentation with transformers. 2021:[arXiv:2105.15203 p.]. Available from: https:\/\/ui.adsabs.harvard.edu\/abs\/2021arXiv210515203X.\u00a0Accessed 15 Apr 2023."},{"key":"1096_CR25","unstructured":"Oktay O, Schlemper J, Le Folgoc L, Lee M, Heinrich M, Misawa K, et al. Attention U-Net: learning where to look for the pancreas. ArXiv e-prints [Internet]. 2018. Available from: https:\/\/ui.adsabs.harvard.edu\/#abs\/2018arXiv180403999O.\u00a0Accessed 1 Apr 2023."},{"key":"1096_CR26","doi-asserted-by":"crossref","unstructured":"Zahangir Alom M, Hasan M, Yakopcic C, Taha TM, Asari VK. Recurrent residual convolutional neural network based on U-Net (R2U-Net) for medical image segmentation. ArXiv e-prints [Internet]. 2018; 1802. Available from: http:\/\/adsabs.harvard.edu\/abs\/2018arXiv180206955Z.\u00a0Accessed 1 Apr 2023.","DOI":"10.1109\/NAECON.2018.8556686"},{"key":"1096_CR27","unstructured":"Transformer S. Swin-Transformer-Semantic-Segmentation 2020 [Available from: https:\/\/github.com\/SwinTransformer\/Swin-Transformer-Semantic-Segmentation.\u00a0Accessed 10 Oct 2022."},{"issue":"4","key":"1096_CR28","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1249","volume":"8","author":"O Sagi","year":"2018","unstructured":"Sagi O, Rokach L. Ensemble learning: a survey. WIREs Data Min Knowl Discovery. 2018;8(4): e1249.","journal-title":"WIREs Data Min Knowl Discovery"},{"key":"1096_CR29","doi-asserted-by":"crossref","unstructured":"Jadon S. A survey of loss functions for semantic segmentation. 2020:[arXiv:2006.14822 p.]. Available from: https:\/\/ui.adsabs.harvard.edu\/abs\/2020arXiv200614822J.\u00a0Accessed 15 May 2021.","DOI":"10.1109\/CIBCB48159.2020.9277638"},{"key":"1096_CR30","unstructured":"Kingma DP, Ba J. Adam: A Method for Stochastic Optimization. arXiv e-prints. 2014:arXiv:1412.6980.\u00a0Accessed 6 Nov 2022."},{"key":"1096_CR31","unstructured":"Zhang MR, Lucas J, Hinton G, Ba J. Lookahead Optimizer: k steps forward, 1 step back. arXiv e-prints [Internet]. 2019. Available from: https:\/\/ui.adsabs.harvard.edu\/abs\/2019arXiv190708610Z.\u00a0Accessed 1 Apr 2023."},{"key":"1096_CR32","unstructured":"Micikevicius P, Narang S, Alben J, Diamos G, Elsen E, Garcia D, et al. Mixed Precision Training. arXiv e-prints. 2017:arXiv:1710.03740."},{"key":"1096_CR33","unstructured":"Guo C, Pleiss G, Sun Y, Weinberger KQ. On calibration of modern neural networks. Proceedings of the 34th International Conference on Machine Learning - Volume 70; Sydney, NSW, Australia. 3305518: JMLR.org; 2017. p. 1321\u201330."},{"issue":"1","key":"1096_CR34","doi-asserted-by":"publisher","first-page":"45","DOI":"10.4103\/0301-4738.37595","volume":"56","author":"R Parikh","year":"2008","unstructured":"Parikh R, Mathai A, Parikh S, Chandra Sekhar G, Thomas R. Understanding and using sensitivity, specificity and predictive values. Indian J Ophthalmol. 2008;56(1):45\u201350.","journal-title":"Indian J Ophthalmol"}],"container-title":["BMC Medical Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-023-01096-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12880-023-01096-1\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-023-01096-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,17]],"date-time":"2023-11-17T15:13:40Z","timestamp":1700234020000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedimaging.biomedcentral.com\/articles\/10.1186\/s12880-023-01096-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,21]]},"references-count":34,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["1096"],"URL":"https:\/\/doi.org\/10.1186\/s12880-023-01096-1","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-3102519\/v1","asserted-by":"object"}]},"ISSN":["1471-2342"],"issn-type":[{"value":"1471-2342","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,21]]},"assertion":[{"value":"24 June 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 September 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"21 September 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"137"}}