000138551 001__ 138551
000138551 005__ 20240321220447.0
000138551 0247_ $$2doi$$a10.1109/TMI.2016.2528120
000138551 0247_ $$2pmid$$apmid:26891484
000138551 0247_ $$2ISSN$$a0278-0062
000138551 0247_ $$2ISSN$$a1558-0062
000138551 0247_ $$2ISSN$$a1558-254X
000138551 0247_ $$2altmetric$$aaltmetric:5637754
000138551 037__ $$aDZNE-2020-04873
000138551 041__ $$aEnglish
000138551 082__ $$a620
000138551 1001_ $$0P:(DE-2719)2810881$$aAlbarqouni, Shadi$$b0$$eFirst author
000138551 245__ $$aAggNet: Deep Learning From Crowds for Mitosis Detection in Breast Cancer Histology Images.
000138551 260__ $$aNew York, NY$$bIEEE$$c2016
000138551 264_1 $$2Crossref$$3print$$bInstitute of Electrical and Electronics Engineers (IEEE)$$c2016-05-01
000138551 3367_ $$2DRIVER$$aarticle
000138551 3367_ $$2DataCite$$aOutput Types/Journal article
000138551 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1586966677_28672
000138551 3367_ $$2BibTeX$$aARTICLE
000138551 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000138551 3367_ $$00$$2EndNote$$aJournal Article
000138551 520__ $$aThe lack of publicly available ground-truth data has been identified as the major challenge for transferring recent developments in deep learning to the biomedical imaging domain. Though crowdsourcing has enabled annotation of large scale databases for real world images, its application for biomedical purposes requires a deeper understanding and hence, more precise definition of the actual annotation task. The fact that expert tasks are being outsourced to non-expert users may lead to noisy annotations introducing disagreement between users. Despite being a valuable resource for learning annotation models from crowdsourcing, conventional machine-learning methods may have difficulties dealing with noisy annotations during training. In this manuscript, we present a new concept for learning from crowds that handle data aggregation directly as part of the learning process of the convolutional neural network (CNN) via additional crowdsourcing layer (AggNet). Besides, we present an experimental study on learning from crowds designed to answer the following questions. 1) Can deep CNN be trained with data collected from crowdsourcing? 2) How to adapt the CNN to train on multiple types of annotation datasets (ground truth and crowd-based)? 3) How does the choice of annotation and aggregation affect the accuracy? Our experimental setup involved Annot8, a self-implemented web-platform based on Crowdflower API realizing image annotation tasks for a publicly available biomedical image database. Our results give valuable insights into the functionality of deep CNN learning from crowd annotations and prove the necessity of data aggregation integration.
000138551 536__ $$0G:(DE-HGF)POF3-341$$a341 - Molecular Signaling (POF3-341)$$cPOF3-341$$fPOF III$$x0
000138551 588__ $$aDataset connected to CrossRef, PubMed,
000138551 650_2 $$2MeSH$$aBreast Neoplasms: diagnostic imaging
000138551 650_2 $$2MeSH$$aCrowdsourcing: methods
000138551 650_2 $$2MeSH$$aFemale
000138551 650_2 $$2MeSH$$aHistocytochemistry
000138551 650_2 $$2MeSH$$aHumans
000138551 650_2 $$2MeSH$$aImage Interpretation, Computer-Assisted: methods
000138551 650_2 $$2MeSH$$aInternet
000138551 650_2 $$2MeSH$$aMachine Learning
000138551 650_2 $$2MeSH$$aMitosis: physiology
000138551 650_2 $$2MeSH$$aNeural Networks, Computer
000138551 650_2 $$2MeSH$$aVideo Games
000138551 7001_ $$aBaur, Christoph$$b1
000138551 7001_ $$aAchilles, Felix$$b2
000138551 7001_ $$aBelagiannis, Vasileios$$b3
000138551 7001_ $$aDemirci, Stefanie$$b4
000138551 7001_ $$aNavab, Nassir$$b5
000138551 77318 $$2Crossref$$3journal-article$$a10.1109/tmi.2016.2528120$$b : Institute of Electrical and Electronics Engineers (IEEE), 2016-05-01$$n5$$p1313-1321$$tIEEE Transactions on Medical Imaging$$v35$$x0278-0062$$y2016
000138551 773__ $$0PERI:(DE-600)2068206-2$$a10.1109/TMI.2016.2528120$$gVol. 35, no. 5, p. 1313 - 1321$$n5$$p1313-1321$$q35:5<1313 - 1321$$tIEEE transactions on medical imaging$$v35$$x0278-0062$$y2016
000138551 909CO $$ooai:pub.dzne.de:138551$$pVDB
000138551 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)2810881$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b0$$kDZNE
000138551 9131_ $$0G:(DE-HGF)POF3-341$$1G:(DE-HGF)POF3-340$$2G:(DE-HGF)POF3-300$$aDE-HGF$$bForschungsbereich Gesundheit$$lErkrankungen des Nervensystems$$vMolecular Signaling$$x0
000138551 9141_ $$y2016
000138551 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline
000138551 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bIEEE T MED IMAGING : 2017
000138551 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS
000138551 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search
000138551 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC
000138551 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List
000138551 915__ $$0StatID:(DE-HGF)0110$$2StatID$$aWoS$$bScience Citation Index
000138551 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection
000138551 915__ $$0StatID:(DE-HGF)0111$$2StatID$$aWoS$$bScience Citation Index Expanded
000138551 915__ $$0StatID:(DE-HGF)1110$$2StatID$$aDBCoverage$$bCurrent Contents - Clinical Medicine
000138551 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology
000138551 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bIEEE T MED IMAGING : 2017
000138551 9201_ $$0I:(DE-2719)1013012$$kAG Alamoudi$$lCryo-electron microscopy and tomography$$x0
000138551 980__ $$ajournal
000138551 980__ $$aVDB
000138551 980__ $$aI:(DE-2719)1013012
000138551 980__ $$aUNRESTRICTED