Semi-Supervised and Class-Imbalanced Open Set Medical Image Recognition
Open set recognition (OSR) for medical images is vital to ensure practical safeguards in clinical applications. It demands establishing model awareness of rare and unknown conditions, rejecting what the model does not know, and thus preventing these conditions from being dangerously misclassified in...
Saved in:
Published in | IEEE access Vol. 12; pp. 122852 - 122877 |
---|---|
Main Authors | , , , , |
Format | Journal Article |
Language | English |
Published |
Piscataway
IEEE
2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Open set recognition (OSR) for medical images is vital to ensure practical safeguards in clinical applications. It demands establishing model awareness of rare and unknown conditions, rejecting what the model does not know, and thus preventing these conditions from being dangerously misclassified into any false known classes. Meanwhile, it often further requires training the OSR model based on a mixture of severely class-imbalanced and partially labeled data due to the prevalence of long-tailed data distribution and the massive cost of medical image annotations. Unfortunately, most existing OSR methods perform poorly under such complex conditions, largely because they neglect to thoroughly exploit the knowledge from the imbalanced and unlabeled data to help better distinguish the known and unknown lesion patterns. This paper proposes a novel Semi-supervised and Class-Imbalanced Open Set medical image Recognition (SCI-OSR) algorithm as a joint solution to address these issues. The major contributions are as follows. Firstly, we propose a modified open mixup approach boasting selective fine-grained lesion foreground fusion based on the learned weighted attentive masks to augment the unlabeled, small-sized, and even pseudo-unknown class data, where the generated attentive masks serve as effective candidate lesion area estimators. Secondly, our approach can more effectively extract the disentangled and discriminative feature representations based on our tailored consistency and contrastive losses, which helps more properly exploit the augmented data and aids in pushing the learned features of known and unknown classes to be well separated by large margins. Finally, a prototype-based open set classifier is established based on the extracted discriminative features of the original and augmented data. It favors producing equally low confidence scores over all non-positive classes and thus more reliably rejects the unknown query data. Through comparative experiments and ablation studies conducted on several popular medical benchmarks, we demonstrate that our proposed SCI-OSR can successfully boost OSR performances, leading to improved known class classification and unknown class rejection over other state-of-the-art methods. |
---|---|
AbstractList | Open set recognition (OSR) for medical images is vital to ensure practical safeguards in clinical applications. It demands establishing model awareness of rare and unknown conditions, rejecting what the model does not know, and thus preventing these conditions from being dangerously misclassified into any false known classes. Meanwhile, it often further requires training the OSR model based on a mixture of severely class-imbalanced and partially labeled data due to the prevalence of long-tailed data distribution and the massive cost of medical image annotations. Unfortunately, most existing OSR methods perform poorly under such complex conditions, largely because they neglect to thoroughly exploit the knowledge from the imbalanced and unlabeled data to help better distinguish the known and unknown lesion patterns. This paper proposes a novel Semi-supervised and Class-Imbalanced Open Set medical image Recognition (SCI-OSR) algorithm as a joint solution to address these issues. The major contributions are as follows. Firstly, we propose a modified open mixup approach boasting selective fine-grained lesion foreground fusion based on the learned weighted attentive masks to augment the unlabeled, small-sized, and even pseudo-unknown class data, where the generated attentive masks serve as effective candidate lesion area estimators. Secondly, our approach can more effectively extract the disentangled and discriminative feature representations based on our tailored consistency and contrastive losses, which helps more properly exploit the augmented data and aids in pushing the learned features of known and unknown classes to be well separated by large margins. Finally, a prototype-based open set classifier is established based on the extracted discriminative features of the original and augmented data. It favors producing equally low confidence scores over all non-positive classes and thus more reliably rejects the unknown query data. Through comparative experiments and ablation studies conducted on several popular medical benchmarks, we demonstrate that our proposed SCI-OSR can successfully boost OSR performances, leading to improved known class classification and unknown class rejection over other state-of-the-art methods. |
Author | Xiao, Xingxing Xu, Yiqian Zhao, Rui-Wei Feng, Rui Wang, Ruofan |
Author_xml | – sequence: 1 givenname: Yiqian surname: Xu fullname: Xu, Yiqian organization: School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai, China – sequence: 2 givenname: Ruofan surname: Wang fullname: Wang, Ruofan organization: School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai, China – sequence: 3 givenname: Rui-Wei orcidid: 0000-0002-8498-5761 surname: Zhao fullname: Zhao, Rui-Wei organization: Fudan University, Shanghai, China – sequence: 4 givenname: Xingxing surname: Xiao fullname: Xiao, Xingxing email: xiaoxingxing@fudan.edu.cn organization: Institute for Humanities and Social Science Data, Fudan University, Shanghai, China – sequence: 5 givenname: Rui orcidid: 0000-0002-4747-0574 surname: Feng fullname: Feng, Rui organization: School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai, China |
BookMark | eNpNkU9r20AQxZfgQtI0nyA5CHqWu3-klfZohJsYHAJRcl5Go1mzRta6u3ah3z5KFErmMsNj3m8G3ne2GMNIjN0KvhSCm1-rplm37VJyWSxVUchSmwt2JYU2uSqVXnyZL9lNSns-VT1JZXXF7ls6-Lw9Hyn-9Yn6DMY-awZIKd8cOhhgxEl8OtKYtXTKHqn3CEO2OcCOsmfCsBv9yYfxB_vmYEh089mv2evv9UvzkG-f7jfNapujLLjIu4rXyoBQZLiuoSJHZKqahEShnCGoi7KUPZeKNCDIWnPUDlAYJw0oVNdsM3P7AHt7jP4A8Z8N4O2HEOLOQjx5HMh2RlSqU7VTFRaG00Rx3GFXd4jIsZxYP2fWMYY_Z0onuw_nOE7vWyW4NNoYo6ctNW9hDClFcv-vCm7fA7BzAPY9APsZwOS6m12eiL44tCqErtQbyU-CZA |
CODEN | IAECCG |
Cites_doi | 10.1109/CVPR.2019.00019 10.1109/CVPR.2019.00414 10.1109/ISBI.2018.8363547 10.1109/TPAMI.2019.2914680 10.1109/ICASSP49357.2023.10096305 10.1109/CVPR52729.2023.02309 10.1016/j.dib.2020.105474 10.3390/app14020923 10.1109/CVPR.2016.319 10.1007/978-3-319-46478-7_31 10.1609/aaai.v37i1.25191 10.1016/j.procs.2023.10.295 10.1038/sdata.2018.161 10.1109/TII.2023.3316180 10.1038/s41592-023-01886-z 10.1109/CVPR.2016.90 10.1109/CVPRW50498.2020.00374 10.1016/j.media.2021.102274 10.1007/978-3-030-59861-7_62 10.1109/CVPR52729.2023.01162 10.1109/ISBI52829.2022.9761412 10.1007/978-3-030-01231-1_38 10.1109/CVPR46437.2021.01515 10.1007/978-3-031-16431-6_69 10.1038/s41586-023-05881-4 10.1109/ICCV48922.2021.00689 10.1109/TCYB.2021.3124231 10.1007/978-3-030-32226-7_90 10.1016/j.media.2019.03.009 10.1016/j.compbiomed.2023.107840 10.48550/arxiv.1710.09412 10.5244/C.31.42 10.1007/978-3-030-69538-5_29 10.1109/CVPR42600.2020.01269 10.1109/TNNLS.2017.2732482 10.1016/j.media.2024.103107 10.1609/aaai.v38i6.28367 10.1007/978-3-030-87196-3_19 10.1007/978-3-030-01216-8_12 10.1016/j.jbi.2022.104227 10.1016/j.cell.2018.02.010 10.3390/biomedicines11030887 10.1109/TPAMI.2023.3277881 10.1145/3065386 10.1007/978-3-031-43993-3_53 10.1007/11538059_91 10.1145/3422622 10.1016/j.knosys.2021.108021 10.1016/S2589-7500(20)30292-2 10.1609/aaai.v35i2.16255 10.1109/TCSVT.2023.3268680 10.1109/CVPR.2018.00459 10.1007/978-3-030-32245-8_67 10.1109/JPROC.2021.3054390 10.1109/ICCV51070.2023.01766 10.1109/TPAMI.2020.2981604 10.1007/s13735-021-00218-1 10.1109/CVPR.2019.00264 10.1109/CVPR42600.2020.01349 10.1016/j.patcog.2023.109385 10.1109/CVPR52688.2022.00676 10.1109/JBHI.2018.2824327 10.2991/978-94-6463-366-5_9 10.1609/aaai.v33i01.33013714 10.1109/ICCV.2017.74 10.1016/j.ejrad.2021.110069 10.1109/ICCV48922.2021.01310 10.1109/TPAMI.2021.3106743 10.1109/CVPR.2019.00949 10.1109/ICASSP40776.2020.9053994 10.1007/978-3-030-58580-8_30 10.1109/CVPRW50498.2020.00380 10.1109/CVPR52688.2022.01182 10.1007/0-387-34471-3 10.1109/CVPR.2018.00366 10.1109/TPAMI.2020.3045079 10.1148/radiol.220182 10.1109/ICME55011.2023.00053 10.1007/978-3-030-01219-9_18 10.1016/j.engappai.2022.105373 10.1016/S2589-7500(23)00126-7 10.1007/978-3-319-46478-7_29 10.1109/TMI.2022.3176050 10.1016/j.patcog.2023.110133 10.1016/j.media.2021.101985 10.1613/jair.953 10.1109/TAI.2024.3379113 10.1109/TCSVT.2024.3359319 10.1016/j.neunet.2018.07.011 10.1109/TNNLS.2020.2995319 10.1109/BIBM52615.2021.9669560 10.11591/ijece.v14i3.pp2583-2591 10.1109/CVPR.2016.173 10.1109/ICCV.2017.324 10.1007/978-3-031-16452-1_22 10.1109/ICCV.2019.00612 10.1109/CVPR.2019.00241 10.17632/rscbjbr9sj.3 10.1038/s41591-021-01312-x 10.1007/978-3-031-16434-7_10 10.1109/TPAMI.2023.3268118 10.1109/WACV56688.2023.00640 |
ContentType | Journal Article |
Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 |
Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 |
DBID | 97E ESBDL RIA RIE AAYXX CITATION 7SC 7SP 7SR 8BQ 8FD JG9 JQ2 L7M L~C L~D DOA |
DOI | 10.1109/ACCESS.2024.3442569 |
DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Xplore CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Engineered Materials Abstracts METADEX Technology Research Database Materials Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional DOAJ Directory of Open Access Journals |
DatabaseTitle | CrossRef Materials Research Database Engineered Materials Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace METADEX Computer and Information Systems Abstracts Professional |
DatabaseTitleList | Materials Research Database |
Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: RIE name: IEEE Xplore url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering |
EISSN | 2169-3536 |
EndPage | 122877 |
ExternalDocumentID | oai_doaj_org_article_b9173b38f37c490eac1f0fcb8bccc0c5 10_1109_ACCESS_2024_3442569 10634167 |
Genre | orig-research |
GrantInformation_xml | – fundername: Science and Technology Commission of Shanghai Municipality grantid: 23511100602; 21511100500 funderid: 10.13039/501100003399 – fundername: National Natural Science Foundation of China grantid: 62172101 funderid: 10.13039/501100001809 |
GroupedDBID | 0R~ 4.4 5VS 6IK 97E AAJGR ACGFS ADBBV ALMA_UNASSIGNED_HOLDINGS BCNDV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD ESBDL GROUPED_DOAJ IFIPE IPLJI JAVBF KQ8 M43 M~E O9- OCL OK1 RIA RIE RIG RNS AAYXX CITATION 7SC 7SP 7SR 8BQ 8FD JG9 JQ2 L7M L~C L~D |
ID | FETCH-LOGICAL-c2401-b70839a13e9068a7efee978e12c13f9ea84552d023e6aca2860c6fac19f29a3c3 |
IEDL.DBID | RIE |
ISSN | 2169-3536 |
IngestDate | Tue Sep 17 03:05:05 EDT 2024 Thu Sep 12 00:44:04 EDT 2024 Wed Sep 18 12:54:26 EDT 2024 Wed Sep 18 05:50:32 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c2401-b70839a13e9068a7efee978e12c13f9ea84552d023e6aca2860c6fac19f29a3c3 |
ORCID | 0000-0002-4747-0574 0000-0002-8498-5761 |
OpenAccessLink | https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/document/10634167 |
PQID | 3102969996 |
PQPubID | 4845423 |
PageCount | 26 |
ParticipantIDs | doaj_primary_oai_doaj_org_article_b9173b38f37c490eac1f0fcb8bccc0c5 ieee_primary_10634167 crossref_primary_10_1109_ACCESS_2024_3442569 proquest_journals_3102969996 |
PublicationCentury | 2000 |
PublicationDate | 20240000 2024-00-00 20240101 2024-01-01 |
PublicationDateYYYYMMDD | 2024-01-01 |
PublicationDate_xml | – year: 2024 text: 20240000 |
PublicationDecade | 2020 |
PublicationPlace | Piscataway |
PublicationPlace_xml | – name: Piscataway |
PublicationTitle | IEEE access |
PublicationTitleAbbrev | Access |
PublicationYear | 2024 |
Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
References | ref57 ref59 ref58 ref53 ref52 ref55 ref54 Dhamija (ref11) ref51 ref50 ref46 ref48 ref47 ref42 ref41 ref44 Lee (ref18) ref43 ref49 Cao (ref68) ref8 ref9 ref4 ref3 ref6 ref5 ref100 ref101 ref40 Verma (ref33) ref34 ref37 ref30 ref32 Laine (ref35) 2016 ref39 ref38 Liu (ref17) ref24 ref23 ref26 ref25 ref20 ref22 ref21 Shu (ref71) ref27 ref29 Byrd (ref77) Sensoy (ref45) ref13 ref12 ref15 ref14 ref97 ref96 Prabhu (ref7) 2019 ref99 ref124 ref10 Kim (ref86) ref98 ref16 ref19 Sohn (ref106) Pacheco (ref28) 2019 ref93 ref92 ref95 ref94 ref91 ref90 ref89 Combalia (ref119) 2019 ref85 ref88 ref87 Ren (ref72) ref82 Tarvainen (ref36) ref81 ref83 ref80 ref79 ref108 ref78 ref109 ref107 ref75 ref74 ref102 ref76 ref103 ref2 ref1 Snell (ref56) Takahashi (ref84) Berthelot (ref104) ref111 ref70 ref112 ref73 ref110 Deng (ref31) ref67 ref117 ref69 ref118 ref115 ref63 ref116 ref66 ref113 ref65 ref114 Zhang (ref64) Berthelot (ref105) Van der Maaten (ref125) 2008; 9 ref60 ref122 ref123 ref62 ref120 ref61 ref121 |
References_xml | – ident: ref74 doi: 10.1109/CVPR.2019.00019 – start-page: 1195 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref36 article-title: Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results contributor: fullname: Tarvainen – ident: ref58 doi: 10.1109/CVPR.2019.00414 – ident: ref123 doi: 10.1109/ISBI.2018.8363547 – ident: ref69 doi: 10.1109/TPAMI.2019.2914680 – ident: ref47 doi: 10.1109/ICASSP49357.2023.10096305 – ident: ref40 doi: 10.1109/CVPR52729.2023.02309 – ident: ref118 doi: 10.1016/j.dib.2020.105474 – ident: ref112 doi: 10.3390/app14020923 – ident: ref115 doi: 10.1109/CVPR.2016.319 – ident: ref49 doi: 10.1007/978-3-319-46478-7_31 – ident: ref90 doi: 10.1609/aaai.v37i1.25191 – start-page: 21464 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref17 article-title: Energy-based out-of-distribution detection contributor: fullname: Liu – ident: ref111 doi: 10.1016/j.procs.2023.10.295 – start-page: 4080 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref56 article-title: Prototypical Siamese networks for few-shot learning contributor: fullname: Snell – ident: ref124 doi: 10.1038/sdata.2018.161 – start-page: 3183 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref45 article-title: Evidential deep learning to quantify classification uncertainty contributor: fullname: Sensoy – ident: ref61 doi: 10.1109/TII.2023.3316180 – ident: ref9 doi: 10.1038/s41592-023-01886-z – ident: ref42 doi: 10.1109/CVPR.2016.90 – ident: ref29 doi: 10.1109/CVPRW50498.2020.00374 – ident: ref8 doi: 10.1016/j.media.2021.102274 – start-page: 1 volume-title: Proc. Med. Imag. Deep Learn. ident: ref31 article-title: Medical open set recognition via intra-class clustering contributor: fullname: Deng – ident: ref26 doi: 10.1007/978-3-030-59861-7_62 – start-page: 1567 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref68 article-title: Learning imbalanced datasets with label-distribution-aware margin loss contributor: fullname: Cao – ident: ref92 doi: 10.1109/CVPR52729.2023.01162 – ident: ref30 doi: 10.1109/ISBI52829.2022.9761412 – ident: ref20 doi: 10.1007/978-3-030-01231-1_38 – ident: ref24 doi: 10.1109/CVPR46437.2021.01515 – ident: ref38 doi: 10.1007/978-3-031-16431-6_69 – ident: ref63 doi: 10.1038/s41586-023-05881-4 – ident: ref52 doi: 10.1109/ICCV48922.2021.00689 – ident: ref91 doi: 10.1109/TCYB.2021.3124231 – ident: ref99 doi: 10.1007/978-3-030-32226-7_90 – ident: ref94 doi: 10.1016/j.media.2019.03.009 – ident: ref95 doi: 10.1016/j.compbiomed.2023.107840 – ident: ref32 doi: 10.48550/arxiv.1710.09412 – ident: ref19 doi: 10.5244/C.31.42 – ident: ref21 doi: 10.1007/978-3-030-69538-5_29 – ident: ref102 doi: 10.1109/CVPR42600.2020.01269 – ident: ref67 doi: 10.1109/TNNLS.2017.2732482 – ident: ref96 doi: 10.1016/j.media.2024.103107 – ident: ref48 doi: 10.1609/aaai.v38i6.28367 – start-page: 786 volume-title: Proc. Asian Conf. Mach. Learn. ident: ref84 article-title: RICAP: Random image cropping and patching data augmentation for deep CNNs contributor: fullname: Takahashi – ident: ref103 doi: 10.1007/978-3-030-87196-3_19 – ident: ref81 doi: 10.1007/978-3-030-01216-8_12 – start-page: 1 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref11 article-title: Reducing network agnostophobia contributor: fullname: Dhamija – ident: ref13 doi: 10.1016/j.jbi.2022.104227 – ident: ref121 doi: 10.1016/j.cell.2018.02.010 – ident: ref12 doi: 10.3390/biomedicines11030887 – start-page: 1 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref104 article-title: MixMatch: A holistic approach to semi-supervised learning contributor: fullname: Berthelot – ident: ref57 doi: 10.1109/TPAMI.2023.3277881 – ident: ref41 doi: 10.1145/3065386 – ident: ref44 doi: 10.1007/978-3-031-43993-3_53 – ident: ref76 doi: 10.1007/11538059_91 – ident: ref60 doi: 10.1145/3422622 – start-page: 1 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref77 article-title: What is the effect of importance weighting in deep learning? contributor: fullname: Byrd – ident: ref97 doi: 10.1016/j.knosys.2021.108021 – start-page: 5275 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref86 article-title: Puzzle Mix: Exploiting saliency and local statistics for optimal mixup contributor: fullname: Kim – start-page: 1 volume-title: Proc. Int. Conf. Represent. Learn. ident: ref18 article-title: Training confidence-calibrated classifiers for detecting out-of-distribution samples contributor: fullname: Lee – ident: ref4 doi: 10.1016/S2589-7500(20)30292-2 – ident: ref88 doi: 10.1609/aaai.v35i2.16255 – ident: ref93 doi: 10.1109/TCSVT.2023.3268680 – ident: ref83 doi: 10.1109/CVPR.2018.00459 – ident: ref37 doi: 10.1007/978-3-030-32245-8_67 – ident: ref1 doi: 10.1109/JPROC.2021.3054390 – start-page: 1919 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref71 article-title: Meta-weight-net: Learning an explicit mapping for sample weighting contributor: fullname: Shu – ident: ref73 doi: 10.1109/ICCV51070.2023.01766 – ident: ref10 doi: 10.1109/TPAMI.2020.2981604 – ident: ref2 doi: 10.1007/s13735-021-00218-1 – ident: ref39 doi: 10.1109/CVPR.2019.00264 – ident: ref23 doi: 10.1109/CVPR42600.2020.01349 – ident: ref53 doi: 10.1016/j.patcog.2023.109385 – ident: ref78 doi: 10.1109/CVPR52688.2022.00676 – ident: ref120 doi: 10.1109/JBHI.2018.2824327 – ident: ref113 doi: 10.2991/978-94-6463-366-5_9 – ident: ref34 doi: 10.1609/aaai.v33i01.33013714 – start-page: 1724 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Workshops (CVPR) ident: ref64 article-title: Recognize anything: A strong image tagging model contributor: fullname: Zhang – year: 2019 ident: ref7 article-title: Open set medical diagnosis publication-title: arXiv:1910.02830 contributor: fullname: Prabhu – ident: ref116 doi: 10.1109/ICCV.2017.74 – ident: ref14 doi: 10.1016/j.ejrad.2021.110069 – ident: ref46 doi: 10.1109/ICCV48922.2021.01310 – ident: ref22 doi: 10.1109/TPAMI.2021.3106743 – ident: ref66 doi: 10.1109/CVPR.2019.00949 – ident: ref87 doi: 10.1109/ICASSP40776.2020.9053994 – year: 2016 ident: ref35 article-title: Temporal ensembling for semi-supervised learning publication-title: arXiv:1610.02242 contributor: fullname: Laine – ident: ref117 doi: 10.1007/978-3-030-58580-8_30 – year: 2019 ident: ref119 article-title: BCN20000: Dermoscopic lesions in the wild publication-title: arXiv:1908.02288 contributor: fullname: Combalia – ident: ref27 doi: 10.1109/CVPRW50498.2020.00380 – ident: ref89 doi: 10.1109/CVPR52688.2022.01182 – start-page: 1 volume-title: Proc. Int. Conf. Learn. Represent. ident: ref105 article-title: ReMixMatch: Semi-supervised learning with distribution matching and augmentation anchoring contributor: fullname: Berthelot – ident: ref43 doi: 10.1007/0-387-34471-3 – ident: ref50 doi: 10.1109/CVPR.2018.00366 – ident: ref51 doi: 10.1109/TPAMI.2020.3045079 – ident: ref6 doi: 10.1148/radiol.220182 – ident: ref25 doi: 10.1109/ICME55011.2023.00053 – ident: ref82 doi: 10.1007/978-3-030-01219-9_18 – ident: ref101 doi: 10.1016/j.engappai.2022.105373 – start-page: 4334 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref72 article-title: Learning to reweight examples for robust deep learning contributor: fullname: Ren – ident: ref5 doi: 10.1016/S2589-7500(23)00126-7 – ident: ref80 doi: 10.1007/978-3-319-46478-7_29 – ident: ref100 doi: 10.1109/TMI.2022.3176050 – ident: ref54 doi: 10.1016/j.patcog.2023.110133 – ident: ref110 doi: 10.1016/j.media.2021.101985 – ident: ref75 doi: 10.1613/jair.953 – ident: ref62 doi: 10.1109/TAI.2024.3379113 – ident: ref55 doi: 10.1109/TCSVT.2024.3359319 – start-page: 6438 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref33 article-title: Manifold mixup: Better representations by interpolating hidden states contributor: fullname: Verma – ident: ref79 doi: 10.1016/j.neunet.2018.07.011 – ident: ref98 doi: 10.1109/TNNLS.2020.2995319 – ident: ref15 doi: 10.1109/BIBM52615.2021.9669560 – volume: 9 start-page: 2579 issue: 11 year: 2008 ident: ref125 article-title: Visualizing data using t-SNE publication-title: J. Mach. Learn. Res. contributor: fullname: Van der Maaten – ident: ref114 doi: 10.11591/ijece.v14i3.pp2583-2591 – year: 2019 ident: ref28 article-title: Skin cancer detection based on deep learning and entropy to detect outlier samples publication-title: arXiv:1909.04525 contributor: fullname: Pacheco – ident: ref16 doi: 10.1109/CVPR.2016.173 – ident: ref70 doi: 10.1109/ICCV.2017.324 – ident: ref109 doi: 10.1007/978-3-031-16452-1_22 – ident: ref85 doi: 10.1109/ICCV.2019.00612 – ident: ref59 doi: 10.1109/CVPR.2019.00241 – ident: ref122 doi: 10.17632/rscbjbr9sj.3 – ident: ref3 doi: 10.1038/s41591-021-01312-x – ident: ref108 doi: 10.1007/978-3-031-16434-7_10 – ident: ref65 doi: 10.1109/TPAMI.2023.3268118 – ident: ref107 doi: 10.1109/WACV56688.2023.00640 – start-page: 596 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref106 article-title: FixMatch: Simplifying semi-supervised learning with consistency and confidence contributor: fullname: Sohn |
SSID | ssj0000816957 |
Score | 2.3406997 |
Snippet | Open set recognition (OSR) for medical images is vital to ensure practical safeguards in clinical applications. It demands establishing model awareness of rare... |
SourceID | doaj proquest crossref ieee |
SourceType | Open Website Aggregation Database Publisher |
StartPage | 122852 |
SubjectTerms | Ablation Algorithms Biomedical imaging class imbalance Data augmentation Data models Diseases Feature extraction Image recognition Lesions Masks medical image analysis Medical imaging Open set recognition semi-supervised Semisupervised learning Training |
SummonAdditionalLinks | – databaseName: DOAJ Directory of Open Access Journals dbid: DOA link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07T8MwELZQJxgQjyIKBXlgJNSOE8ceS0VpkWCgVOpmOc5ZYmha0fb_c05SFImBhdWK4tx3vldy-Y6QO5tyAB-7CFwmo8QrhjaXQATSFjYVXkMafnB-fZOTefKySBetUV-hJ6ymB66BG-RYT4hcKC8yl2iGfoJ75l2ucuccczV7KU9bxVTlgxWXOs0amiHO9GA4GqFEWBDGyYNI8KSGFudWKKoY-5sRK7_8chVsxifkuMkS6bB-ulNyAOUZOWpxB56T5xksP6PZbh2MfQMFtWVBqxGX0XSZh4ZFh4uhX4TOYEubDzJ0ukQHQt_3bUOrskvm46eP0SRqpiJEDqMvj_IMsyZtuQDNpLIZeAAsBYHHjgdsrUrSNC4wFiPezsZKMic9wqZ9rK1w4oJ0ylUJl4SyQB4nBQMR60TibUSG0nPwEhQavuuR-z1AZl2TX5iqaGDa1HiagKdp8OyRxwDiz6WBubpaQH2aRp_mL332SDeooLWfxEArsx7p73ViGjPbGMxNYy1DzXb1H3tfk8MgT_2GpU86268d3GDOsc1vq-P1DZMZ0oM priority: 102 providerName: Directory of Open Access Journals |
Title | Semi-Supervised and Class-Imbalanced Open Set Medical Image Recognition |
URI | https://ieeexplore.ieee.org/document/10634167 https://www.proquest.com/docview/3102969996/abstract/ https://doaj.org/article/b9173b38f37c490eac1f0fcb8bccc0c5 |
Volume | 12 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1LT9wwEB4Bp_bQJ1WXUpRDj2TrR-LER7oqhUrlwILEzXImY6lCm0Wwe-HXd-x40aoVUm-RFSe2P8_LngfAF19LoqCwJGxMWYVWMM1VVJLxva91sFTHAOdfF-bsuvp5U9_kYPUUC0NEyfmMpvEx3eX3S1zHozKmcMNM1zS7sNsKNQZrPR2oxAoStm5yZiEp7NeT2YwnwTagqqa64s0ZvZq3pE9K0p-rqvzDipN8OX0NF5uRjW4lt9P1qpvi419JG_976G_gVdY0i5Nxa7yFHRrewcut_IPv4cecFr_L-fouMowH6gs_9EUqk1meL7ro9IjcGH1OijmtinypU5wvmAkVlxvXo-WwD9en369mZ2WurFAiS3BZdg1rXtZLTVaY1jcUiNicJKlQRnx8W9W16lmeM2boVWsEmuBR2qCs16g_wN6wHOgjFCImoDNakFa2MvwZ3fDkJQVDLTMPnMDxZsXd3ZhAwyXDQ1g3AuQiQC4DNIFvEZWnV2P269TAq-kyMbmObUzd6TboBisrWHbIIAJ2bYeIAusJ7EcEtv43Lv4EDjcgu0yqD471W2VNtPsOnun2CV7EIY4HL4ewt7pf02dWRVbdUTLhj9JG_AOMINvu |
link.rule.ids | 315,786,790,802,870,2115,4043,27956,27957,27958,55109 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3NT9swFH8acBg7DLaB1o1tOey4dE6cOPERqrF2gx5WkLhZzsuzhFBTNNrL_vo9Oy6qQEjcIufL9s_vy34fAF9tmRG5HFPCSqWFqwXTXEEpKdvaUjpNpQ9wPp-q8WXx66q8isHqIRaGiILzGQ39ZTjLbxe48ltlTOGKma6qtmCHBb3QfbjW_ZaKryGhyyrmFuL7349HIx4GW4F5MZQFL0_v17whf0Ka_lhX5REzDhLmdA-m6771jiU3w9WyGeK_B2kbn935fXgddc3kuF8cb-AFdW_h1UYGwnfwc0bz63S2uvUs447axHZtEgplppN5490ekRu910kyo2USj3WSyZzZUPJn7Xy06A7g8vTHxWicxtoKKbIMz9KmYt1L20ySFqq2FTkiNigpyzHzCNm6KMu8ZYnOqKHNayVQOYuZdrm2EuUhbHeLjt5DInwKOiUFyVwXij8jKx58Rk5RzewDB_BtPePmtk-hYYLpIbTpATIeIBMBGsCJR-X-UZ__OjTwbJpITqZhK1M2snaywkILlh6ZEw6bukFEgeUADjwCG__rJ38AR2uQTSTWO8Mabq6Vt_w-PPHaF3g5vjg_M2eT6e-PsOu722_DHMH28u-KPrFismw-h-X4Hx_73k8 |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Semi-Supervised+and+Class-Imbalanced+Open+Set+Medical+Image+Recognition&rft.jtitle=IEEE+access&rft.au=Xu%2C+Yiqian&rft.au=Wang%2C+Ruofan&rft.au=Zhao%2C+Rui-Wei&rft.au=Xiao%2C+Xingxing&rft.date=2024&rft.issn=2169-3536&rft.eissn=2169-3536&rft.volume=12&rft.spage=122852&rft.epage=122877&rft_id=info:doi/10.1109%2FACCESS.2024.3442569&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_ACCESS_2024_3442569 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2169-3536&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2169-3536&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2169-3536&client=summon |