SSMD: Semi-Supervised medical image detection with adaptive consistency and heterogeneous perturbation
•We propose a semi-supervised medical image detector with a novel adaptive consistency cost function which takes into account the confidence of proposals at each spatial position.•We develop novel heterogeneous perturbation strategies that consist of two novel components: a noisy residual block for...
Uložené v:
| Vydané v: | Medical image analysis Ročník 72; s. 102117 |
|---|---|
| Hlavní autori: | , , , , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Amsterdam
Elsevier B.V
01.08.2021
Elsevier BV |
| Predmet: | |
| ISSN: | 1361-8415, 1361-8423, 1361-8423 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | •We propose a semi-supervised medical image detector with a novel adaptive consistency cost function which takes into account the confidence of proposals at each spatial position.•We develop novel heterogeneous perturbation strategies that consist of two novel components: a noisy residual block for feature space, and an instance-level adversarial perturbation strategy for image space. The proposed heterogeneous perturbation strategies can improve the detection accuracy by enhancing the robustness of image features.•We experimentally verify the effectiveness of the proposed modules. The experiments suggest that the proposed detector outperforms the existing state-of-the-arts considerably.
[Display omitted]
Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of fully-supervised methods with additional unlabeled data. However, as a fundamental task, semi-supervised object detection has not gained enough attention in the field of medical image analysis. In this paper, we propose a novel Semi-Supervised Medical image Detector (SSMD). The motivation behind SSMD is to provide free yet effective supervision for unlabeled data, by regularizing the predictions at each position to be consistent. To achieve the above idea, we develop a novel adaptive consistency cost function to regularize different components in the predictions. Moreover, we introduce heterogeneous perturbation strategies that work in both feature space and image space, so that the proposed detector is promising to produce powerful image representations and robust predictions. Extensive experimental results show that the proposed SSMD achieves the state-of-the-art performance at a wide range of settings. We also demonstrate the strength of each proposed module with comprehensive ablation studies. |
|---|---|
| AbstractList | •We propose a semi-supervised medical image detector with a novel adaptive consistency cost function which takes into account the confidence of proposals at each spatial position.•We develop novel heterogeneous perturbation strategies that consist of two novel components: a noisy residual block for feature space, and an instance-level adversarial perturbation strategy for image space. The proposed heterogeneous perturbation strategies can improve the detection accuracy by enhancing the robustness of image features.•We experimentally verify the effectiveness of the proposed modules. The experiments suggest that the proposed detector outperforms the existing state-of-the-arts considerably.
[Display omitted]
Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of fully-supervised methods with additional unlabeled data. However, as a fundamental task, semi-supervised object detection has not gained enough attention in the field of medical image analysis. In this paper, we propose a novel Semi-Supervised Medical image Detector (SSMD). The motivation behind SSMD is to provide free yet effective supervision for unlabeled data, by regularizing the predictions at each position to be consistent. To achieve the above idea, we develop a novel adaptive consistency cost function to regularize different components in the predictions. Moreover, we introduce heterogeneous perturbation strategies that work in both feature space and image space, so that the proposed detector is promising to produce powerful image representations and robust predictions. Extensive experimental results show that the proposed SSMD achieves the state-of-the-art performance at a wide range of settings. We also demonstrate the strength of each proposed module with comprehensive ablation studies. Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of fully-supervised methods with additional unlabeled data. However, as a fundamental task, semi-supervised object detection has not gained enough attention in the field of medical image analysis. In this paper, we propose a novel Semi-Supervised Medical image Detector (SSMD). The motivation behind SSMD is to provide free yet effective supervision for unlabeled data, by regularizing the predictions at each position to be consistent. To achieve the above idea, we develop a novel adaptive consistency cost function to regularize different components in the predictions. Moreover, we introduce heterogeneous perturbation strategies that work in both feature space and image space, so that the proposed detector is promising to produce powerful image representations and robust predictions. Extensive experimental results show that the proposed SSMD achieves the state-of-the-art performance at a wide range of settings. We also demonstrate the strength of each proposed module with comprehensive ablation studies.Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of fully-supervised methods with additional unlabeled data. However, as a fundamental task, semi-supervised object detection has not gained enough attention in the field of medical image analysis. In this paper, we propose a novel Semi-Supervised Medical image Detector (SSMD). The motivation behind SSMD is to provide free yet effective supervision for unlabeled data, by regularizing the predictions at each position to be consistent. To achieve the above idea, we develop a novel adaptive consistency cost function to regularize different components in the predictions. Moreover, we introduce heterogeneous perturbation strategies that work in both feature space and image space, so that the proposed detector is promising to produce powerful image representations and robust predictions. Extensive experimental results show that the proposed SSMD achieves the state-of-the-art performance at a wide range of settings. We also demonstrate the strength of each proposed module with comprehensive ablation studies. Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of fully-supervised methods with additional unlabeled data. However, as a fundamental task, semi-supervised object detection has not gained enough attention in the field of medical image analysis. In this paper, we propose a novel Semi-Supervised Medical image Detector (SSMD). The motivation behind SSMD is to provide free yet effective supervision for unlabeled data, by regularizing the predictions at each position to be consistent. To achieve the above idea, we develop a novel adaptive consistency cost function to regularize different components in the predictions. Moreover, we introduce heterogeneous perturbation strategies that work in both feature space and image space, so that the proposed detector is promising to produce powerful image representations and robust predictions. Extensive experimental results show that the proposed SSMD achieves the state-of-the-art performance at a wide range of settings. We also demonstrate the strength of each proposed module with comprehensive ablation studies. |
| ArticleNumber | 102117 |
| Author | Li, Haofeng Zhou, Hong-Yu Li, Weimin Wang, Chengdi Wang, Gang Zhang, Shu Yu, Yizhou |
| Author_xml | – sequence: 1 givenname: Hong-Yu surname: Zhou fullname: Zhou, Hong-Yu organization: Department of Respiratory and Critical Care Medicine, National Clinical Research Center for Geriatrics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu 610041, P.R. China – sequence: 2 givenname: Chengdi surname: Wang fullname: Wang, Chengdi organization: Department of Respiratory and Critical Care Medicine, National Clinical Research Center for Geriatrics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu 610041, P.R. China – sequence: 3 givenname: Haofeng orcidid: 0000-0001-9120-9843 surname: Li fullname: Li, Haofeng organization: Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen 518000, P.R. China – sequence: 4 givenname: Gang surname: Wang fullname: Wang, Gang organization: Department of Respiratory and Critical Care Medicine, National Clinical Research Center for Geriatrics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu 610041, P.R. China – sequence: 5 givenname: Shu orcidid: 0000-0002-2049-0970 surname: Zhang fullname: Zhang, Shu organization: AI Lab, Deepwise Healthcare, Beijing 100080, P.R. China – sequence: 6 givenname: Weimin surname: Li fullname: Li, Weimin email: weimi003@scu.edu.cn organization: Department of Respiratory and Critical Care Medicine, National Clinical Research Center for Geriatrics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu 610041, P.R. China – sequence: 7 givenname: Yizhou orcidid: 0000-0002-0470-5548 surname: Yu fullname: Yu, Yizhou email: yizhouy@acm.org organization: Department of Computer Science, The University of Hong Kong, Pokfulam, Hong Kong |
| BookMark | eNqFkU1PIyEYgIlxE7XrL_BC4sXLVGAoMCYejK4fiZs9dPdMKLxTaaZDBaab_ntpazx40BOEPM-bNw8n6LAPPSB0RsmYEiouF-MlOG_GjDBaXhil8gAd01rQSnFWH37c6eQInaS0IIRIzskxaqfT33dXeApLX02HFcS1T-Dwdpw1HfZLMwfsIIPNPvT4v88v2Dizyn4N2IY--ZShtxtseodfChfDHHoIQ8JlWB7izGzFn-hHa7oEp-_nCP27__X39rF6_vPwdHvzXNla8FzJVhEhgEnXOKVmUjZWWUNAOGiJZKppheBctLNG0hmTjQIqBXcTQ7icWGPqEbrYz13F8DpAynrpk4WuM7udNJtwrqSSNSvo-Sd0EYbYl-0KJZmgnFBeqHpP2RhSitDqVSxR4kZTorft9ULv2utte71vX6zmk2V93oXI0fjuG_d670LptPYQdbK-JC5gLL-gXfBf-m8oBaLX |
| CitedBy_id | crossref_primary_10_1038_s41392_025_02256_1 crossref_primary_10_3390_bioengineering10020225 crossref_primary_10_1145_3702638 crossref_primary_10_1109_TCSVT_2024_3427645 crossref_primary_10_3390_app13063752 crossref_primary_10_1038_s41598_022_17843_3 crossref_primary_10_1109_JBHI_2025_3527706 crossref_primary_10_1109_TMI_2024_3388328 crossref_primary_10_1109_JBHI_2024_3375889 crossref_primary_10_1016_j_bspc_2022_104502 crossref_primary_10_1109_TMI_2025_3548919 crossref_primary_10_1002_jmri_29212 crossref_primary_10_1007_s10278_025_01650_z crossref_primary_10_1016_j_compbiomed_2023_107320 crossref_primary_10_1016_j_jfranklin_2025_107546 crossref_primary_10_1016_j_bbe_2022_09_005 crossref_primary_10_1016_j_rineng_2025_106174 crossref_primary_10_1038_s41598_024_81842_9 crossref_primary_10_1109_JBHI_2024_3360102 crossref_primary_10_1016_j_media_2022_102506 crossref_primary_10_1148_radiol_210817 crossref_primary_10_1109_TMI_2024_3366940 crossref_primary_10_1039_D3MH01500A crossref_primary_10_1109_TIP_2025_3573516 crossref_primary_10_1109_TMI_2022_3221666 crossref_primary_10_1002_tee_23902 crossref_primary_10_1007_s00521_023_09214_4 crossref_primary_10_1016_j_semcancer_2023_02_006 crossref_primary_10_1007_s11042_023_16383_w crossref_primary_10_1016_j_compbiomed_2023_107751 crossref_primary_10_1016_j_pccm_2023_05_001 crossref_primary_10_1016_j_compbiomed_2022_106120 crossref_primary_10_1109_TNNLS_2023_3282809 crossref_primary_10_1016_j_compbiomed_2023_107471 crossref_primary_10_46632_jeae_1_2_12 |
| Cites_doi | 10.1109/TNNLS.2019.2892409 10.1117/1.JMI.5.3.036501 10.1109/JBHI.2018.2879449 10.1016/j.media.2019.101539 10.1038/s41746-018-0065-x 10.1109/TPAMI.2018.2858821 |
| ContentType | Journal Article |
| Copyright | 2021 Copyright Elsevier BV Aug 2021 Copyright © 2021. Published by Elsevier B.V. |
| Copyright_xml | – notice: 2021 – notice: Copyright Elsevier BV Aug 2021 – notice: Copyright © 2021. Published by Elsevier B.V. |
| DBID | AAYXX CITATION 7QO 8FD FR3 K9. NAPCQ P64 7X8 |
| DOI | 10.1016/j.media.2021.102117 |
| DatabaseName | CrossRef Biotechnology Research Abstracts Technology Research Database Engineering Research Database ProQuest Health & Medical Complete (Alumni) Nursing & Allied Health Premium Biotechnology and BioEngineering Abstracts MEDLINE - Academic |
| DatabaseTitle | CrossRef ProQuest Health & Medical Complete (Alumni) Nursing & Allied Health Premium Engineering Research Database Biotechnology Research Abstracts Technology Research Database Biotechnology and BioEngineering Abstracts MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic ProQuest Health & Medical Complete (Alumni) |
| Database_xml | – sequence: 1 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine Engineering |
| EISSN | 1361-8423 |
| ExternalDocumentID | 10_1016_j_media_2021_102117 S1361841521001638 |
| GroupedDBID | --- --K --M .~1 0R~ 1B1 1~. 1~5 29M 4.4 457 4G. 53G 5GY 5VS 7-5 71M 8P~ AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABBQC ABJNI ABLVK ABMAC ABMZM ABXDB ABYKQ ACDAQ ACGFS ACIUM ACIWK ACNNM ACPRK ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADTZH AEBSH AECPX AEKER AENEX AFKWA AFRAH AFTJW AFXIZ AGHFR AGUBO AGYEJ AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV AJRQY ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ ANZVX AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC BNPGV C45 CAG COF CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HVGLF HX~ HZ~ IHE J1W JJJVA KOM LCYCR M41 MO0 N9A O-L O9- OAUVE OVD OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ SDF SDG SDP SEL SES SEW SPC SPCBC SSH SST SSV SSZ T5K TEORI UHS ~G- 9DU AATTM AAXKI AAYWO AAYXX ABWVN ACIEU ACLOT ACRPL ACVFH ADCNI ADNMO ADVLN AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD 7QO 8FD FR3 K9. NAPCQ P64 7X8 |
| ID | FETCH-LOGICAL-c364t-7f8066e27d9d88b779c8ca0e6def07289f66446fb971b2798e1764d5a0475caa3 |
| ISICitedReferencesCount | 48 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000681131600008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1361-8415 1361-8423 |
| IngestDate | Sun Sep 28 10:05:58 EDT 2025 Tue Oct 07 07:14:24 EDT 2025 Tue Nov 18 22:16:41 EST 2025 Sat Nov 29 06:59:52 EST 2025 Fri Feb 23 02:47:50 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Nuclei detection 41A10 65D05 Semi-Supervised learning 65D17 Lesion detection 41A05 |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c364t-7f8066e27d9d88b779c8ca0e6def07289f66446fb971b2798e1764d5a0475caa3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0002-0470-5548 0000-0002-2049-0970 0000-0001-9120-9843 |
| PQID | 2572614014 |
| PQPubID | 2045428 |
| ParticipantIDs | proquest_miscellaneous_2544878732 proquest_journals_2572614014 crossref_primary_10_1016_j_media_2021_102117 crossref_citationtrail_10_1016_j_media_2021_102117 elsevier_sciencedirect_doi_10_1016_j_media_2021_102117 |
| PublicationCentury | 2000 |
| PublicationDate | August 2021 2021-08-00 20210801 |
| PublicationDateYYYYMMDD | 2021-08-01 |
| PublicationDate_xml | – month: 08 year: 2021 text: August 2021 |
| PublicationDecade | 2020 |
| PublicationPlace | Amsterdam |
| PublicationPlace_xml | – name: Amsterdam |
| PublicationTitle | Medical image analysis |
| PublicationYear | 2021 |
| Publisher | Elsevier B.V Elsevier BV |
| Publisher_xml | – name: Elsevier B.V – name: Elsevier BV |
| References | Laine, Aila (bib0017) 2016 Zhang, Yang, Chen, Fredericksen, Hughes, Chen (bib0043) 2017 Zhu, Liu, Fan, Xie (bib0047) 2018 Nie, Gao, Wang, Shen (bib0031) 2018 Wang, Zhang, Zhang, Wang (bib0037) 2020 Verma, Lamb, Kannala, Bengio, Lopez-Paz (bib0036) 2019 Zhou, Yu, Bian, Hu, Ma, Zheng (bib0044) 2020 Chang, Yan, Lou, Axel, Metaxas (bib0004) 2020 Yan, Wang, Lu, Summers (bib0039) 2018; 5 Li, Zhang, Zhang, Huang, Wang, Yu (bib0022) 2019 Pezeshk, Hamidian, Petrick, Sahiner (bib0032) 2018; 23 Lin, Goyal, Girshick, He, Dollár (bib0025) 2017 Berthelot, Carlini, Goodfellow, Papernot, Oliver, Raffel (bib0003) 2019 Ding, Li, Hu, Wang (bib0008) 2017 He, Zhang, Ren, Sun (bib0013) 2016 Zhou, He, Huang, Liu, Zhu, Cui, Shao (bib0045) 2019 Athiwaratkun, Finzi, Izmailov, Wilson (bib0001) 2018 Li, Yu, Chen, Fu, Xing, Heng (bib0021) 2020 Berthelot, Carlini, Cubuk, Kurakin, Sohn, Zhang, Raffel (bib0002) 2019 He, Gkioxari, Dollár, Girshick (bib0012) 2017 Khosravan, Bagci (bib0016) 2018 Sohn, Zhang, Li, Zhang, Lee, Pfister (bib0033) 2020 Yi, Walia, Babyn (bib0040) 2018 Zlocha, Dou, Glocker (bib0048) 2019 Yu, Wang, Li, Fu, Heng (bib0041) 2019 Lin, Dollár, Girshick, He, Hariharan, Belongie (bib0024) 2017 Ghiasi, Lin, Le (bib0009) 2018 Girshick (bib0010) 2015 Liu, Cao, Akin, Tian (bib0026) 2019 Lee (bib0019) 2013 Ke, Wang, Yan, Ren, Lau (bib0015) 2019 Lecouat, Chang, Foo, Unnikrishnan, Brown, Zenati, Beers, Chandrasekhar, Kalpathy-Cramer, Krishnaswamy (bib0018) 2018 Jeong, Lee, Kim, Kwak (bib0014) 2019 DeVries, Taylor (bib0007) 2017 Goodfellow, Shlens, Szegedy (bib0011) 2014 Srivastava, Hinton, Krizhevsky, Sutskever, Salakhutdinov (bib0034) 2014; 15 Liao, Liang, Li, Hu, Song (bib0023) 2019; 30 Wang, Wu (bib0038) 2019 Chen, Bortsova, Juárez, van Tulder, de Bruijne (bib0006) 2019 Li, Yu, Chen, Fu, Heng (bib0020) 2018 Liu, Yu, Luo, Dou, Heng (bib0027) 2020 Tarvainen, Valpola (bib0035) 2017 Madani, Moradi, Karargyris, Syeda-Mahmood (bib0028) 2018 Madani, Ong, Tibrewal, Mofrad (bib0029) 2018; 1 Miyato, Maeda, Koyama, Ishii (bib0030) 2018; 41 Zhang, Cisse, Dauphin, Lopez-Paz (bib0042) 2017 Chen, Bentley, Mori, Misawa, Fujiwara, Rueckert (bib0005) 2019; 58 Zhou, Sodha, Siddiquee, Feng, Tajbakhsh, Gotway, Liang (bib0046) 2019 Zhang (10.1016/j.media.2021.102117_bib0043) 2017 Madani (10.1016/j.media.2021.102117_bib0029) 2018; 1 Zhou (10.1016/j.media.2021.102117_bib0046) 2019 Zlocha (10.1016/j.media.2021.102117_bib0048) 2019 Verma (10.1016/j.media.2021.102117_bib0036) 2019 Berthelot (10.1016/j.media.2021.102117_bib0003) 2019 Yu (10.1016/j.media.2021.102117_bib0041) 2019 Laine (10.1016/j.media.2021.102117_bib0017) 2016 Liao (10.1016/j.media.2021.102117_bib0023) 2019; 30 Athiwaratkun (10.1016/j.media.2021.102117_bib0001) 2018 Goodfellow (10.1016/j.media.2021.102117_bib0011) 2014 Lin (10.1016/j.media.2021.102117_bib0024) 2017 Yan (10.1016/j.media.2021.102117_bib0039) 2018; 5 Chen (10.1016/j.media.2021.102117_bib0005) 2019; 58 He (10.1016/j.media.2021.102117_bib0012) 2017 Chang (10.1016/j.media.2021.102117_bib0004) 2020 Ding (10.1016/j.media.2021.102117_bib0008) 2017 Madani (10.1016/j.media.2021.102117_bib0028) 2018 Chen (10.1016/j.media.2021.102117_bib0006) 2019 Tarvainen (10.1016/j.media.2021.102117_bib0035) 2017 Li (10.1016/j.media.2021.102117_bib0021) 2020 Berthelot (10.1016/j.media.2021.102117_bib0002) 2019 Li (10.1016/j.media.2021.102117_bib0022) 2019 Zhou (10.1016/j.media.2021.102117_bib0044) 2020 DeVries (10.1016/j.media.2021.102117_bib0007) 2017 Srivastava (10.1016/j.media.2021.102117_bib0034) 2014; 15 Nie (10.1016/j.media.2021.102117_bib0031) 2018 Lecouat (10.1016/j.media.2021.102117_bib0018) 2018 Pezeshk (10.1016/j.media.2021.102117_bib0032) 2018; 23 Zhang (10.1016/j.media.2021.102117_bib0042) 2017 Zhou (10.1016/j.media.2021.102117_bib0045) 2019 Wang (10.1016/j.media.2021.102117_bib0038) 2019 Li (10.1016/j.media.2021.102117_bib0020) 2018 Jeong (10.1016/j.media.2021.102117_bib0014) 2019 Lin (10.1016/j.media.2021.102117_bib0025) 2017 He (10.1016/j.media.2021.102117_bib0013) 2016 Liu (10.1016/j.media.2021.102117_bib0026) 2019 Khosravan (10.1016/j.media.2021.102117_bib0016) 2018 Ghiasi (10.1016/j.media.2021.102117_bib0009) 2018 Lee (10.1016/j.media.2021.102117_bib0019) 2013 Sohn (10.1016/j.media.2021.102117_bib0033) 2020 Miyato (10.1016/j.media.2021.102117_bib0030) 2018; 41 Wang (10.1016/j.media.2021.102117_bib0037) 2020 Zhu (10.1016/j.media.2021.102117_bib0047) 2018 Girshick (10.1016/j.media.2021.102117_bib0010) 2015 Liu (10.1016/j.media.2021.102117_bib0027) 2020 Yi (10.1016/j.media.2021.102117_bib0040) 2018 Ke (10.1016/j.media.2021.102117_bib0015) 2019 |
| References_xml | – start-page: 2079 year: 2019 end-page: 2088 ident: bib0045 article-title: Collaborative Learning of Semi-Supervised Segmentation and Classification for Medical Images publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – year: 2019 ident: bib0002 article-title: Remixmatch: Semi-supervised learning with distribution matching and augmentation anchoring publication-title: International Conference on Learning Representations – year: 2016 ident: bib0017 article-title: Temporal ensembling for semi-Supervised learning publication-title: arXiv preprint arXiv:1610.02242 – year: 2013 ident: bib0019 article-title: Pseudo-label: The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks publication-title: ICML Workshop – start-page: 2117 year: 2017 end-page: 2125 ident: bib0024 article-title: Feature Pyramid Networks for Object Detection publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – year: 2020 ident: bib0033 article-title: A simple semi-supervised learning framework for object detection publication-title: arXiv preprint arXiv:2005.04757 – start-page: 408 year: 2017 end-page: 416 ident: bib0043 article-title: Deep Adversarial Networks for Biomedical Image Segmentation Utilizing Unannotated Images publication-title: International Conference on Medical Image Computing and Computer Assisted Intervention – year: 2018 ident: bib0001 article-title: There are many consistent explanations of unlabeled data: why you should average publication-title: arXiv preprint arXiv:1806.05594 – start-page: 13 year: 2019 end-page: 21 ident: bib0022 article-title: MVP-Net: Multi-view FPN with Position-aware Attention for Deep Universal Lesion Detection publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – year: 2014 ident: bib0011 article-title: Explaining and harnessing adversarial examples publication-title: arXiv preprint arXiv:1412.6572 – volume: 41 start-page: 1979 year: 2018 end-page: 1993 ident: bib0030 article-title: Virtual adversarial training: a regularization method for supervised and semi-Supervised learning publication-title: IEEE Trans Pattern Anal Mach Intell – start-page: 559 year: 2017 end-page: 567 ident: bib0008 article-title: Accurate pulmonary nodule detection in computed tomography images using deep convolutional neural networks publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – start-page: 1752 year: 2020 end-page: 1755 ident: bib0004 article-title: Soft-Label Guided Semi-Supervised Learning for Bi-Ventricle Segmentation in Cardiac Cine MRI publication-title: 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI) – start-page: 673 year: 2018 end-page: 681 ident: bib0047 article-title: Deeplung: Deep 3d dual path nets for automated pulmonary nodule detection and classification publication-title: 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) – volume: 15 start-page: 1929 year: 2014 end-page: 1958 ident: bib0034 article-title: Dropout: A Simple way to prevent neural networks from overfitting publication-title: The Journal of Machine Learning Research – year: 2019 ident: bib0036 article-title: Interpolation consistency training for semi-Supervised learning publication-title: arXiv preprint arXiv:1903.03825 – start-page: 770 year: 2016 end-page: 778 ident: bib0013 article-title: Deep Residual Learning for Image Recognition publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – year: 2020 ident: bib0027 article-title: Semi-Supervised medical image classification with relation-driven self-ensembling model publication-title: IEEE Trans Med Imaging – start-page: 457 year: 2019 end-page: 465 ident: bib0006 article-title: Multi-Task Attention-Based Semi-Supervised Learning for Medical Image Segmentation publication-title: International Conference on Medical Image Computing and Computer Assisted Intervention – volume: 1 start-page: 1 year: 2018 end-page: 11 ident: bib0029 article-title: Deep echocardiography: data-efficient supervised and semi-Supervised deep learning towards automated diagnosis of cardiac disease publication-title: NPJ digital medicine – start-page: 3951 year: 2020 end-page: 3960 ident: bib0037 article-title: FocalMix: Semi-Supervised Learning for 3D Medical Image Detection publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – start-page: 5050 year: 2019 end-page: 5060 ident: bib0003 article-title: MixMatch: A Holistic Approach to Semi-Supervised Learning publication-title: Advances in Neural Information Processing Systems – start-page: 384 year: 2019 end-page: 393 ident: bib0046 article-title: Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – start-page: 10727 year: 2018 end-page: 10737 ident: bib0009 article-title: Dropblock: A Regularization Method for Convolutional Networks publication-title: Advances in Neural Information Processing Systems – year: 2018 ident: bib0018 article-title: Semi-Supervised deep learning for abnormality classification in retinal images publication-title: arXiv preprint arXiv:1812.07832 – year: 2017 ident: bib0042 article-title: Mixup: beyond empirical risk minimization publication-title: arXiv preprint arXiv:1710.09412 – start-page: 794 year: 2018 end-page: 802 ident: bib0016 article-title: S4nd: Single-shot single-scale lung nodule detection publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – volume: 23 start-page: 2080 year: 2018 end-page: 2090 ident: bib0032 article-title: 3-D convolutional neural networks for automatic detection of pulmonary nodules in chest ct publication-title: IEEE J Biomed Health Inform – start-page: 513 year: 2019 end-page: 521 ident: bib0026 article-title: 3dfpn-hs2: 3d feature pyramid network based high sensitivity and specificity pulmonary nodule detection publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – year: 2020 ident: bib0044 article-title: Comparing to Learn: Surpassing ImageNet Pretraining on Radiographs By Comparing Image Representations publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – start-page: 2980 year: 2017 end-page: 2988 ident: bib0025 article-title: Focal Loss for Dense Object Detection publication-title: Proceedings of the IEEE International Conference on Computer Vision – year: 2018 ident: bib0040 article-title: Unsupervised and semi-Supervised learning with categorical generative adversarial networks assisted by wasserstein distance for dermoscopy image classification publication-title: arXiv preprint arXiv:1804.03700 – year: 2020 ident: bib0021 article-title: Transformation-Consistent self-Ensembling model for semisupervised medical image segmentation publication-title: IEEE Trans Neural Netw Learn Syst – start-page: 370 year: 2018 end-page: 378 ident: bib0031 article-title: ASDNet: Attention based Semi-Supervised Deep Networks for Medical Image Segmentation publication-title: International Conference on Medical Image Computing and Computer Assisted Intervention – year: 2018 ident: bib0020 article-title: Semi-Supervised skin lesion segmentation via transformation consistent self-ensembling model publication-title: British Machine Vision Conference – volume: 58 start-page: 101539 year: 2019 ident: bib0005 article-title: Self-supervised learning for medical image analysis using image context restoration publication-title: Med Image Anal – year: 2019 ident: bib0038 article-title: Repetitive reprediction deep decipher for semi-Supervised learning publication-title: arXiv preprint arXiv:1908.04345 – start-page: 402 year: 2019 end-page: 410 ident: bib0048 article-title: Improving Retinanet for CT Lesion Detection with Dense Masks from Weak Recist Labels publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – year: 2017 ident: bib0007 article-title: Improved regularization of convolutional neural networks with cutout publication-title: arXiv preprint arXiv:1708.04552 – start-page: 10758 year: 2019 end-page: 10767 ident: bib0014 article-title: Consistency-based Semi-Supervised Learning for Object detection publication-title: Advances in Neural Information Processing Systems – start-page: 1038 year: 2018 end-page: 1042 ident: bib0028 article-title: Semi-Supervised Learning with Generative Adversarial Networks for Chest X-Ray Classification with Ability of Data Domain Adaptation publication-title: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018) – start-page: 2961 year: 2017 end-page: 2969 ident: bib0012 article-title: Mask r-cnn publication-title: Proceedings of the IEEE International Conference on Computer Vision – volume: 5 start-page: 036501 year: 2018 ident: bib0039 article-title: Deeplesion: automated mining of large-scale lesion annotations and universal lesion detection with deep learning publication-title: J. Med. Imaging – start-page: 6728 year: 2019 end-page: 6736 ident: bib0015 article-title: Dual Student: Breaking the Limits of the Teacher in Semi-Supervised Learning publication-title: Proceedings of the IEEE International Conference on Computer Vision – volume: 30 start-page: 3484 year: 2019 end-page: 3495 ident: bib0023 article-title: Evaluate the malignancy of pulmonary nodules using the 3-d deep leaky noisy-or network publication-title: IEEE Trans Neural Netw Learn Syst – start-page: 605 year: 2019 end-page: 613 ident: bib0041 article-title: Uncertainty-aware Self-ensembling Model for Semi-Supervised 3D Left Atrium Segmentation publication-title: International Conference on Medical Image Computing and Computer Assisted Intervention – start-page: 1195 year: 2017 end-page: 1204 ident: bib0035 article-title: Mean Teachers are Better Role Models: Weight-averaged Consistency Targets Improve Semi-Supervised Deep Learning Results publication-title: Advances in Neural Information Processing Systems – start-page: 1440 year: 2015 end-page: 1448 ident: bib0010 article-title: Fast r-cnn publication-title: Proceedings of the IEEE international conference on computer vision – year: 2020 ident: 10.1016/j.media.2021.102117_bib0044 article-title: Comparing to Learn: Surpassing ImageNet Pretraining on Radiographs By Comparing Image Representations – year: 2016 ident: 10.1016/j.media.2021.102117_bib0017 article-title: Temporal ensembling for semi-Supervised learning publication-title: arXiv preprint arXiv:1610.02242 – start-page: 2079 year: 2019 ident: 10.1016/j.media.2021.102117_bib0045 article-title: Collaborative Learning of Semi-Supervised Segmentation and Classification for Medical Images – start-page: 1195 year: 2017 ident: 10.1016/j.media.2021.102117_bib0035 article-title: Mean Teachers are Better Role Models: Weight-averaged Consistency Targets Improve Semi-Supervised Deep Learning Results – year: 2018 ident: 10.1016/j.media.2021.102117_bib0001 article-title: There are many consistent explanations of unlabeled data: why you should average publication-title: arXiv preprint arXiv:1806.05594 – start-page: 370 year: 2018 ident: 10.1016/j.media.2021.102117_bib0031 article-title: ASDNet: Attention based Semi-Supervised Deep Networks for Medical Image Segmentation – start-page: 5050 year: 2019 ident: 10.1016/j.media.2021.102117_bib0003 article-title: MixMatch: A Holistic Approach to Semi-Supervised Learning – year: 2020 ident: 10.1016/j.media.2021.102117_bib0033 article-title: A simple semi-supervised learning framework for object detection publication-title: arXiv preprint arXiv:2005.04757 – start-page: 6728 year: 2019 ident: 10.1016/j.media.2021.102117_bib0015 article-title: Dual Student: Breaking the Limits of the Teacher in Semi-Supervised Learning – year: 2019 ident: 10.1016/j.media.2021.102117_bib0002 article-title: Remixmatch: Semi-supervised learning with distribution matching and augmentation anchoring – year: 2018 ident: 10.1016/j.media.2021.102117_bib0040 article-title: Unsupervised and semi-Supervised learning with categorical generative adversarial networks assisted by wasserstein distance for dermoscopy image classification publication-title: arXiv preprint arXiv:1804.03700 – start-page: 13 year: 2019 ident: 10.1016/j.media.2021.102117_bib0022 article-title: MVP-Net: Multi-view FPN with Position-aware Attention for Deep Universal Lesion Detection – start-page: 770 year: 2016 ident: 10.1016/j.media.2021.102117_bib0013 article-title: Deep Residual Learning for Image Recognition – year: 2019 ident: 10.1016/j.media.2021.102117_bib0036 article-title: Interpolation consistency training for semi-Supervised learning publication-title: arXiv preprint arXiv:1903.03825 – year: 2014 ident: 10.1016/j.media.2021.102117_bib0011 article-title: Explaining and harnessing adversarial examples publication-title: arXiv preprint arXiv:1412.6572 – start-page: 10758 year: 2019 ident: 10.1016/j.media.2021.102117_bib0014 article-title: Consistency-based Semi-Supervised Learning for Object detection – start-page: 605 year: 2019 ident: 10.1016/j.media.2021.102117_bib0041 article-title: Uncertainty-aware Self-ensembling Model for Semi-Supervised 3D Left Atrium Segmentation – start-page: 673 year: 2018 ident: 10.1016/j.media.2021.102117_bib0047 article-title: Deeplung: Deep 3d dual path nets for automated pulmonary nodule detection and classification – start-page: 3951 year: 2020 ident: 10.1016/j.media.2021.102117_bib0037 article-title: FocalMix: Semi-Supervised Learning for 3D Medical Image Detection – volume: 30 start-page: 3484 issue: 11 year: 2019 ident: 10.1016/j.media.2021.102117_bib0023 article-title: Evaluate the malignancy of pulmonary nodules using the 3-d deep leaky noisy-or network publication-title: IEEE Trans Neural Netw Learn Syst doi: 10.1109/TNNLS.2019.2892409 – start-page: 1440 year: 2015 ident: 10.1016/j.media.2021.102117_bib0010 article-title: Fast r-cnn – start-page: 1038 year: 2018 ident: 10.1016/j.media.2021.102117_bib0028 article-title: Semi-Supervised Learning with Generative Adversarial Networks for Chest X-Ray Classification with Ability of Data Domain Adaptation – volume: 5 start-page: 036501 issue: 3 year: 2018 ident: 10.1016/j.media.2021.102117_bib0039 article-title: Deeplesion: automated mining of large-scale lesion annotations and universal lesion detection with deep learning publication-title: J. Med. Imaging doi: 10.1117/1.JMI.5.3.036501 – year: 2018 ident: 10.1016/j.media.2021.102117_bib0020 article-title: Semi-Supervised skin lesion segmentation via transformation consistent self-ensembling model publication-title: British Machine Vision Conference – year: 2020 ident: 10.1016/j.media.2021.102117_bib0021 article-title: Transformation-Consistent self-Ensembling model for semisupervised medical image segmentation publication-title: IEEE Trans Neural Netw Learn Syst – start-page: 2117 year: 2017 ident: 10.1016/j.media.2021.102117_bib0024 article-title: Feature Pyramid Networks for Object Detection – start-page: 408 year: 2017 ident: 10.1016/j.media.2021.102117_bib0043 article-title: Deep Adversarial Networks for Biomedical Image Segmentation Utilizing Unannotated Images – start-page: 2980 year: 2017 ident: 10.1016/j.media.2021.102117_bib0025 article-title: Focal Loss for Dense Object Detection – volume: 23 start-page: 2080 issue: 5 year: 2018 ident: 10.1016/j.media.2021.102117_bib0032 article-title: 3-D convolutional neural networks for automatic detection of pulmonary nodules in chest ct publication-title: IEEE J Biomed Health Inform doi: 10.1109/JBHI.2018.2879449 – year: 2020 ident: 10.1016/j.media.2021.102117_bib0027 article-title: Semi-Supervised medical image classification with relation-driven self-ensembling model publication-title: IEEE Trans Med Imaging – year: 2018 ident: 10.1016/j.media.2021.102117_bib0018 article-title: Semi-Supervised deep learning for abnormality classification in retinal images publication-title: arXiv preprint arXiv:1812.07832 – start-page: 457 year: 2019 ident: 10.1016/j.media.2021.102117_bib0006 article-title: Multi-Task Attention-Based Semi-Supervised Learning for Medical Image Segmentation – volume: 58 start-page: 101539 year: 2019 ident: 10.1016/j.media.2021.102117_bib0005 article-title: Self-supervised learning for medical image analysis using image context restoration publication-title: Med Image Anal doi: 10.1016/j.media.2019.101539 – start-page: 559 year: 2017 ident: 10.1016/j.media.2021.102117_bib0008 article-title: Accurate pulmonary nodule detection in computed tomography images using deep convolutional neural networks – start-page: 10727 year: 2018 ident: 10.1016/j.media.2021.102117_bib0009 article-title: Dropblock: A Regularization Method for Convolutional Networks – year: 2017 ident: 10.1016/j.media.2021.102117_bib0007 article-title: Improved regularization of convolutional neural networks with cutout publication-title: arXiv preprint arXiv:1708.04552 – volume: 1 start-page: 1 issue: 1 year: 2018 ident: 10.1016/j.media.2021.102117_bib0029 article-title: Deep echocardiography: data-efficient supervised and semi-Supervised deep learning towards automated diagnosis of cardiac disease publication-title: NPJ digital medicine doi: 10.1038/s41746-018-0065-x – start-page: 1752 year: 2020 ident: 10.1016/j.media.2021.102117_bib0004 article-title: Soft-Label Guided Semi-Supervised Learning for Bi-Ventricle Segmentation in Cardiac Cine MRI – volume: 41 start-page: 1979 issue: 8 year: 2018 ident: 10.1016/j.media.2021.102117_bib0030 article-title: Virtual adversarial training: a regularization method for supervised and semi-Supervised learning publication-title: IEEE Trans Pattern Anal Mach Intell doi: 10.1109/TPAMI.2018.2858821 – start-page: 384 year: 2019 ident: 10.1016/j.media.2021.102117_bib0046 article-title: Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis – start-page: 513 year: 2019 ident: 10.1016/j.media.2021.102117_bib0026 article-title: 3dfpn-hs2: 3d feature pyramid network based high sensitivity and specificity pulmonary nodule detection – year: 2013 ident: 10.1016/j.media.2021.102117_bib0019 article-title: Pseudo-label: The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks – year: 2019 ident: 10.1016/j.media.2021.102117_bib0038 article-title: Repetitive reprediction deep decipher for semi-Supervised learning publication-title: arXiv preprint arXiv:1908.04345 – year: 2017 ident: 10.1016/j.media.2021.102117_bib0042 article-title: Mixup: beyond empirical risk minimization publication-title: arXiv preprint arXiv:1710.09412 – start-page: 2961 year: 2017 ident: 10.1016/j.media.2021.102117_bib0012 article-title: Mask r-cnn – start-page: 402 year: 2019 ident: 10.1016/j.media.2021.102117_bib0048 article-title: Improving Retinanet for CT Lesion Detection with Dense Masks from Weak Recist Labels – start-page: 794 year: 2018 ident: 10.1016/j.media.2021.102117_bib0016 article-title: S4nd: Single-shot single-scale lung nodule detection – volume: 15 start-page: 1929 issue: 1 year: 2014 ident: 10.1016/j.media.2021.102117_bib0034 article-title: Dropout: A Simple way to prevent neural networks from overfitting publication-title: The Journal of Machine Learning Research |
| SSID | ssj0007440 |
| Score | 2.5561845 |
| Snippet | •We propose a semi-supervised medical image detector with a novel adaptive consistency cost function which takes into account the confidence of proposals at... Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of... |
| SourceID | proquest crossref elsevier |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 102117 |
| SubjectTerms | Ablation Consistency Cost function Image analysis Image classification Image detection Image processing Image segmentation Lesion detection Medical imaging Motivation Nuclei detection Object recognition Performance enhancement Perturbation Predictions Semi-Supervised learning |
| Title | SSMD: Semi-Supervised medical image detection with adaptive consistency and heterogeneous perturbation |
| URI | https://dx.doi.org/10.1016/j.media.2021.102117 https://www.proquest.com/docview/2572614014 https://www.proquest.com/docview/2544878732 |
| Volume | 72 |
| WOSCitedRecordID | wos000681131600008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1361-8423 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0007440 issn: 1361-8415 databaseCode: AIEXJ dateStart: 20161201 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3db9MwELe6DSH2gGCA6BiTkRBPpGqT1HZ4m0bHh7qC1E4qT5ZjO1unNS1rO-2Jv5272EnbDabxwEtUuY4d-X6-O5_vg5C3KgYpw6IIdlqog9hwHQiDgboZVjzSqU7TIlC4y3s9MRwm32u1X2UszNUFz3NxfZ1M_yupoQ2IjaGz_0DualBogN9AdHgC2eF5L8L3-8cf8Zzft-NR0F9MkRnMQK0c-yuZ0RjddIydW1cl3EW3GTUtnIg0OszO5kVAJtrUz9BdZgKTWfSVhcFARKVLapa1oNbGVj7RydIoPVkUIm6SnwY_FksbvuMzh2c2PzWjyjXI1dFWk8x6qbrS95Pybd5QEbYqN7mSt0asFYjYhRc37B_aPEN2tXw8R8XS4y668xazd3aH80YRY9PAKRvL3uuptXvf5NFJtysHneHg3fRngFXH8Hbel2DZIFshbyfA2LcOvnSGXytZjukTXeSe-8wyb1XhIXhr3r_pNjekfKG6DJ6Qx_7MQQ8cVp6Sms13yPZKJsod8vDY-1g8IxkC6AO9AR_q4UMLEtMKPhThQ0v40BX4UIAPXYMPXYXPc3Jy1Bkcfg58NY5ARyyeBzwToJ7akJvECJFynmihVdMyY7Mmh3N7xkC3Zlma8FYa8kTYFmexaatmzNtaqegF2cwnuX1JqMgSODlrWHBr4qyVCpOyOGUmVla1I5PVSViuo9Q-VT1WTLmQpU_iuSwWX-LiS7f4dfK-emnqMrXc3Z2VBJJe2XRKpAR43f3iXklO6bf9TILgCxnaKuI6eVP9DZwar99UscYSkwEKkI9RuHuPPq_Io-Um2iOb88uFfU0e6Kv5aHa5Tzb4UOx7tP4GH-23wQ |
| linkProvider | Elsevier |
| 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=SSMD%3A+Semi-Supervised+medical+image+detection+with+adaptive+consistency+and+heterogeneous+perturbation&rft.jtitle=Medical+image+analysis&rft.au=Zhou%2C+Hong-Yu&rft.au=Wang%2C+Chengdi&rft.au=Li%2C+Haofeng&rft.au=Wang%2C+Gang&rft.date=2021-08-01&rft.issn=1361-8423&rft.eissn=1361-8423&rft.volume=72&rft.spage=102117&rft_id=info:doi/10.1016%2Fj.media.2021.102117&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1361-8415&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1361-8415&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1361-8415&client=summon |