Uncertainty quantification using Bayesian neural networks in classification: Application to biomedical image segmentation
Most recent research of deep neural networks in the field of computer vision has focused on improving performances of point predictions by developing network architectures or learning algorithms. Reliable uncertainty quantification accompanied by point estimation can lead to a more informed decision...
Saved in:
| Published in | Computational statistics & data analysis Vol. 142; p. 106816 |
|---|---|
| Main Authors | , , , |
| Format | Journal Article |
| Language | English |
| Published |
Elsevier B.V
01.02.2020
|
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | Most recent research of deep neural networks in the field of computer vision has focused on improving performances of point predictions by developing network architectures or learning algorithms. Reliable uncertainty quantification accompanied by point estimation can lead to a more informed decision, and the quality of prediction can be improved. In this paper, we invoke a Bayesian neural network and propose a natural way of quantifying uncertainties in classification problems by decomposing the moment-based predictive uncertainty into two parts: aleatoric and epistemic uncertainty. The proposed method takes into account the discrete nature of the outcome, yielding the correct interpretation of each uncertainty. We demonstrate that the proposed uncertainty quantification method provides additional insights into the point prediction using two Ischemic Stroke Lesion Segmentation Challenge datasets and the Digital Retinal Images for Vessel Extraction dataset. |
|---|---|
| AbstractList | Most recent research of deep neural networks in the field of computer vision has focused on improving performances of point predictions by developing network architectures or learning algorithms. Reliable uncertainty quantification accompanied by point estimation can lead to a more informed decision, and the quality of prediction can be improved. In this paper, we invoke a Bayesian neural network and propose a natural way of quantifying uncertainties in classification problems by decomposing the moment-based predictive uncertainty into two parts: aleatoric and epistemic uncertainty. The proposed method takes into account the discrete nature of the outcome, yielding the correct interpretation of each uncertainty. We demonstrate that the proposed uncertainty quantification method provides additional insights into the point prediction using two Ischemic Stroke Lesion Segmentation Challenge datasets and the Digital Retinal Images for Vessel Extraction dataset. |
| ArticleNumber | 106816 |
| Author | Won, Joong-Ho Paik, Myunghee Cho Kim, Beom Joon Kwon, Yongchan |
| Author_xml | – sequence: 1 givenname: Yongchan surname: Kwon fullname: Kwon, Yongchan email: ykwon0407@snu.ac.kr organization: Department of Statistics, Seoul National University, Seoul, 08826, South Korea – sequence: 2 givenname: Joong-Ho surname: Won fullname: Won, Joong-Ho email: wonj@stats.snu.ac.kr organization: Department of Statistics, Seoul National University, Seoul, 08826, South Korea – sequence: 3 givenname: Beom Joon surname: Kim fullname: Kim, Beom Joon email: kim.bj.stroke@gmail.com organization: Department of Neurology and Cerebrovascular Center, Seoul National University Bundang Hospital, Bundang, 13620, South Korea – sequence: 4 givenname: Myunghee Cho orcidid: 0000-0001-6239-4883 surname: Paik fullname: Paik, Myunghee Cho email: myungheechopaik@snu.ac.kr organization: Department of Statistics, Seoul National University, Seoul, 08826, South Korea |
| BookMark | eNp9kT1v2zAQhokgAeJ8_IFMHLvI4YclUkUXN8hHAQNd4pmgyJNBVyYdkmrhf186Sjp08HTA4X0Od89doXMfPCB0R8mcEtrcb-cmWT1nhLal0UjanKEZlYJVgtfsHM1KSFTtQvBLdJXSlhDCFkLO0GHtDcSsnc8H_DZqn13vjM4ueDwm5zf4uz5ActpjD2PUQyn5T4i_EnYem0Gn9A_4ipf7_fBJ54A7F3ZgS2PAbqc3gBNsduDze-AGXfR6SHD7Ua_R-unx9eGlWv18_vGwXFWGc56rztiubi2pOSedlYzIGmgvGiOl5KbtBKlb2lludS0tk7xvRM-hOFg0omFM82v0ZZq7j-FthJTVziUDw6A9hDEpxknNGOeNLFE5RU0MKUXolXHTsjlqNyhK1NG22qqjbXW0rSbbBWX_oftYbo6H09C3CYJy_28HUSXjoDzEuggmKxvcKfwvVWCdgA |
| CitedBy_id | crossref_primary_10_1021_acs_jcim_0c00416 crossref_primary_10_1016_j_compositesa_2025_108822 crossref_primary_10_1002_eqe_3759 crossref_primary_10_1109_ACCESS_2024_3421330 crossref_primary_10_1016_j_neucom_2021_01_090 crossref_primary_10_1016_j_ejrh_2024_102141 crossref_primary_10_1088_1742_6596_2014_1_012007 crossref_primary_10_1016_j_ejrad_2023_110786 crossref_primary_10_1097_RLI_0000000000000755 crossref_primary_10_3390_jimaging11020064 crossref_primary_10_1103_PhysRevD_103_063034 crossref_primary_10_1109_TGRS_2023_3269139 crossref_primary_10_1190_geo2021_0318_1 crossref_primary_10_1016_j_heliyon_2025_e41802 crossref_primary_10_1109_TSG_2020_3019263 crossref_primary_10_1016_j_inffus_2021_05_008 crossref_primary_10_1007_JHEP04_2021_296 crossref_primary_10_1093_bib_bbac211 crossref_primary_10_1016_j_eswa_2021_116494 crossref_primary_10_1109_TMI_2022_3222730 crossref_primary_10_1007_s11548_023_03001_1 crossref_primary_10_1088_1361_6420_ad2f63 crossref_primary_10_1016_j_neunet_2022_02_019 crossref_primary_10_1016_j_media_2020_101790 crossref_primary_10_1002_asjc_2758 crossref_primary_10_1007_s00204_023_03666_2 crossref_primary_10_1080_00207543_2023_2271093 crossref_primary_10_1039_D3CC01988H crossref_primary_10_1016_j_inffus_2022_11_008 crossref_primary_10_1038_s44172_023_00103_1 crossref_primary_10_1016_j_patcog_2021_108515 crossref_primary_10_1016_j_media_2024_103162 crossref_primary_10_1038_s41598_022_18874_6 crossref_primary_10_1080_01431161_2021_1966853 crossref_primary_10_1016_j_inffus_2022_12_013 crossref_primary_10_1016_j_jmatprotec_2021_117485 crossref_primary_10_3390_s22124324 crossref_primary_10_1089_big_2021_0343 crossref_primary_10_3390_pr9060922 crossref_primary_10_1016_j_radonc_2021_09_034 crossref_primary_10_1093_ehjdh_ztab045 crossref_primary_10_3390_s23010062 crossref_primary_10_1109_ACCESS_2024_3372423 crossref_primary_10_1016_j_media_2022_102426 crossref_primary_10_1016_j_compeleceng_2022_107986 crossref_primary_10_1007_s10278_024_01151_5 crossref_primary_10_1111_exsy_13320 crossref_primary_10_1016_j_ijmedinf_2025_105846 crossref_primary_10_1016_j_cirpj_2022_01_002 crossref_primary_10_1103_PhysRevD_102_103509 crossref_primary_10_1088_1361_6587_ad48b7 crossref_primary_10_1016_j_marpolbul_2023_115098 crossref_primary_10_1016_j_media_2024_103392 crossref_primary_10_1002_smo_20230012 crossref_primary_10_1038_s41467_021_25493_8 crossref_primary_10_1088_2632_2153_aba6f1 crossref_primary_10_1111_jon_12997 crossref_primary_10_1109_TAES_2024_3408141 crossref_primary_10_1016_j_cirpj_2021_03_004 crossref_primary_10_3390_electronics11233863 crossref_primary_10_1109_TEM_2023_3270301 crossref_primary_10_1186_s13321_023_00682_3 crossref_primary_10_1016_j_compeleceng_2024_109555 crossref_primary_10_1016_j_metrad_2023_100003 crossref_primary_10_1088_1361_6560_ac60b9 crossref_primary_10_1109_JBHI_2022_3207313 crossref_primary_10_3389_frsen_2022_1100012 crossref_primary_10_3390_rs13081472 crossref_primary_10_1111_coin_12411 crossref_primary_10_7554_eLife_81217 crossref_primary_10_1016_j_compind_2024_104146 crossref_primary_10_1016_j_ins_2022_08_094 crossref_primary_10_1016_j_compag_2023_108414 crossref_primary_10_1093_mnras_stae2500 crossref_primary_10_1109_ACCESS_2024_3513253 crossref_primary_10_1016_j_jclepro_2024_143457 crossref_primary_10_1016_j_compstruct_2024_118087 crossref_primary_10_1007_s10844_023_00828_7 crossref_primary_10_1016_j_neucom_2021_10_119 crossref_primary_10_1016_j_ejso_2023_106996 crossref_primary_10_1111_mice_12808 crossref_primary_10_1088_1742_6596_2909_1_012032 crossref_primary_10_1109_JOE_2023_3252624 crossref_primary_10_1145_3467978 crossref_primary_10_1016_j_tbs_2022_08_011 crossref_primary_10_1177_21582440231182060 crossref_primary_10_1007_s00540_022_03044_9 crossref_primary_10_1007_s11548_022_02578_3 crossref_primary_10_1137_22M1523765 crossref_primary_10_1016_j_jmsy_2022_07_002 crossref_primary_10_1109_TNSRE_2023_3286688 crossref_primary_10_1155_2020_1428968 crossref_primary_10_1016_j_eswa_2022_119121 crossref_primary_10_32604_cmc_2022_023581 crossref_primary_10_1007_s10115_023_01955_x crossref_primary_10_1016_j_pacs_2024_100645 crossref_primary_10_1016_j_neucom_2024_127568 crossref_primary_10_1088_2632_2153_accace crossref_primary_10_3390_rs14133072 crossref_primary_10_1109_TCSVT_2022_3144184 crossref_primary_10_1190_geo2020_0424_1 crossref_primary_10_1016_j_asoc_2020_106597 crossref_primary_10_1063_5_0201613 crossref_primary_10_1186_s12880_023_01121_3 crossref_primary_10_1007_s41060_023_00397_6 crossref_primary_10_1007_s11263_024_02032_8 crossref_primary_10_1016_j_bspc_2024_107383 crossref_primary_10_1103_PhysRevD_105_023531 crossref_primary_10_1002_mp_16795 crossref_primary_10_1007_s10207_023_00675_1 crossref_primary_10_1111_cgf_14333 crossref_primary_10_1109_TMI_2020_3029161 crossref_primary_10_1038_s41598_022_23431_2 crossref_primary_10_1016_j_asoc_2024_111641 crossref_primary_10_3390_biomedicines10061323 crossref_primary_10_3390_electronics10040520 crossref_primary_10_1016_j_inffus_2023_101813 crossref_primary_10_1109_TAFFC_2022_3189974 crossref_primary_10_1016_j_neucom_2020_12_063 crossref_primary_10_1109_TIM_2023_3284933 crossref_primary_10_1038_s41467_022_33699_7 crossref_primary_10_1109_TNNLS_2023_3342138 crossref_primary_10_1016_j_compbiomed_2023_107441 crossref_primary_10_1109_ACCESS_2022_3163384 crossref_primary_10_1016_j_media_2024_103223 crossref_primary_10_1109_TGRS_2022_3152516 crossref_primary_10_1016_j_ndteint_2022_102753 crossref_primary_10_1109_ACCESS_2024_3451373 crossref_primary_10_1016_j_eswa_2023_122093 crossref_primary_10_1109_TGRS_2021_3076012 crossref_primary_10_3389_frai_2024_1410841 crossref_primary_10_14232_actacyb_24_3_2020_3 crossref_primary_10_1016_j_oceaneng_2023_115193 crossref_primary_10_1109_TITS_2023_3304452 crossref_primary_10_1016_j_isci_2022_104814 crossref_primary_10_1016_j_isprsjprs_2024_03_004 crossref_primary_10_1051_0004_6361_202451348 crossref_primary_10_1109_TNNLS_2023_3319468 crossref_primary_10_1371_journal_pone_0253057 crossref_primary_10_1016_j_neucom_2024_128434 crossref_primary_10_1016_j_mri_2022_06_001 crossref_primary_10_1016_j_patrec_2023_08_018 crossref_primary_10_1016_j_cose_2024_103909 crossref_primary_10_1016_j_eswa_2023_120473 crossref_primary_10_1016_j_neunet_2021_09_007 crossref_primary_10_1148_ryai_2021210031 crossref_primary_10_1109_ACCESS_2022_3168734 crossref_primary_10_1002_mp_16695 crossref_primary_10_1145_3464423 crossref_primary_10_1093_bioinformatics_btab346 crossref_primary_10_1109_TMI_2021_3123547 crossref_primary_10_1109_TGRS_2022_3233871 crossref_primary_10_1002_mrm_29214 crossref_primary_10_1145_3716317 crossref_primary_10_1007_s11069_024_06641_x crossref_primary_10_1007_s41976_024_00155_7 crossref_primary_10_1088_1402_4896_adbd94 crossref_primary_10_1016_j_jechem_2023_02_028 crossref_primary_10_1016_j_eswa_2022_118196 crossref_primary_10_1016_j_phro_2024_100578 crossref_primary_10_1016_j_ijar_2024_109301 crossref_primary_10_1007_s11063_024_11593_1 crossref_primary_10_1016_j_ejrad_2024_111884 crossref_primary_10_1190_geo2022_0632_1 crossref_primary_10_1016_j_mex_2024_102659 crossref_primary_10_1016_j_ress_2024_110513 crossref_primary_10_3389_fonc_2025_1474590 crossref_primary_10_1088_2632_2153_ad88d5 crossref_primary_10_1186_s13321_022_00619_2 crossref_primary_10_1016_j_artmed_2024_102830 crossref_primary_10_1016_j_csda_2024_108097 crossref_primary_10_1016_j_knosys_2022_108418 crossref_primary_10_1016_j_chaos_2023_114341 crossref_primary_10_34133_2022_9837076 crossref_primary_10_1016_j_jsv_2020_115908 crossref_primary_10_1016_j_radonc_2024_110545 crossref_primary_10_1016_j_ress_2025_110962 crossref_primary_10_1007_s00066_024_02262_2 crossref_primary_10_3847_1538_4365_ac14b7 crossref_primary_10_1109_ACCESS_2024_3479913 crossref_primary_10_1002_srin_202100566 crossref_primary_10_1016_j_compmedimag_2023_102321 crossref_primary_10_1007_s10462_024_10814_2 crossref_primary_10_1190_geo2020_0609_1 crossref_primary_10_1016_j_media_2023_102863 crossref_primary_10_1016_j_apgeochem_2022_105450 crossref_primary_10_1109_TGRS_2023_3314130 crossref_primary_10_1088_1361_6560_ac678a crossref_primary_10_1007_s41060_024_00562_5 crossref_primary_10_1016_j_neucom_2022_05_055 crossref_primary_10_1109_TITS_2024_3361676 crossref_primary_10_1115_1_4067001 crossref_primary_10_3390_brainsci12060765 crossref_primary_10_1007_s42979_024_02842_0 crossref_primary_10_1016_j_compbiomed_2022_105246 crossref_primary_10_3390_s22062406 crossref_primary_10_5194_amt_14_1941_2021 crossref_primary_10_1016_j_tws_2024_112466 crossref_primary_10_1088_1361_6420_ad6a34 crossref_primary_10_1109_JSTARS_2022_3212680 crossref_primary_10_1016_j_ailsci_2021_100004 crossref_primary_10_1016_j_compbiomed_2023_107142 crossref_primary_10_1016_j_neunet_2024_106284 crossref_primary_10_1145_3688573 crossref_primary_10_1007_s00521_021_05789_y crossref_primary_10_1038_s41467_023_36960_9 crossref_primary_10_1016_j_ifacol_2023_10_1671 crossref_primary_10_1038_s42256_020_0209_y crossref_primary_10_1016_j_asoc_2020_107046 crossref_primary_10_2139_ssrn_4133438 crossref_primary_10_1002_VIW_20230050 crossref_primary_10_1002_qute_202300221 crossref_primary_10_1007_s12046_022_01980_6 |
| Cites_doi | 10.1109/TMI.2004.825627 10.1016/j.media.2017.07.005 10.1162/neco.1992.4.3.448 10.3389/fneur.2018.00679 10.1038/s41598-017-17876-z 10.1016/j.media.2016.07.009 10.1016/j.media.2016.10.004 10.1016/S0893-6080(00)00045-9 10.1038/nature21056 10.1080/01621459.2017.1303386 10.1016/j.strusafe.2008.06.020 |
| ContentType | Journal Article |
| Copyright | 2019 Elsevier B.V. |
| Copyright_xml | – notice: 2019 Elsevier B.V. |
| DBID | AAYXX CITATION 7S9 L.6 |
| DOI | 10.1016/j.csda.2019.106816 |
| DatabaseName | CrossRef AGRICOLA AGRICOLA - Academic |
| DatabaseTitle | CrossRef AGRICOLA AGRICOLA - Academic |
| DatabaseTitleList | AGRICOLA |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Mathematics |
| EISSN | 1872-7352 |
| ExternalDocumentID | 10_1016_j_csda_2019_106816 S016794731930163X |
| GroupedDBID | --K --M -~X .~1 0R~ 1B1 1OL 1RT 1~. 1~5 29F 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN 9JO AAAKF AAAKG AABNK AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AARIN AAXUO AAYFN ABAOU ABBOA ABFNM ABMAC ABTAH ABUCO ABXDB ABYKQ ACAZW ACDAQ ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADGUI ADJOM ADMUD ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHHHB AHJVU AHZHX AI. AIALX AIEXJ AIGVJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD APLSM ARUGR ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-2 G-Q GBLVA GBOLZ HAMUX HLZ HMJ HVGLF HZ~ H~9 IHE J1W JJJVA KOM LG9 LY1 M26 M41 MHUIS MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG RNS ROL RPZ SBC SDF SDG SDS SES SEW SME SPC SPCBC SSB SSD SST SSV SSW SSZ T5K VH1 VOH WUQ XPP ZMT ZY4 ~02 ~G- AAHBH AATTM AAXKI AAYWO AAYXX ABJNI ABWVN ACRPL ACVFH ADCNI ADNMO ADXHL AEIPS AEUPX AFJKZ AFPUW AFXIZ AGCQF AGQPQ AGRNS AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP BNPGV CITATION SSH 7S9 L.6 |
| ID | FETCH-LOGICAL-c333t-bcdb59d05330bd82085e1f76c8883c9b70591bd3da58d283f67f3e106467622a3 |
| IEDL.DBID | .~1 |
| ISSN | 0167-9473 |
| IngestDate | Fri Jul 11 12:27:00 EDT 2025 Thu Apr 24 23:03:21 EDT 2025 Tue Jul 01 02:24:33 EDT 2025 Fri Feb 23 02:23:50 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Uncertainty quantification Retinal blood vessel segmentation Ischemic stroke lesion segmentation Bayesian neural network Aleatoric and epistemic uncertainty |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c333t-bcdb59d05330bd82085e1f76c8883c9b70591bd3da58d283f67f3e106467622a3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ORCID | 0000-0001-6239-4883 |
| PQID | 2305223368 |
| PQPubID | 24069 |
| ParticipantIDs | proquest_miscellaneous_2305223368 crossref_citationtrail_10_1016_j_csda_2019_106816 crossref_primary_10_1016_j_csda_2019_106816 elsevier_sciencedirect_doi_10_1016_j_csda_2019_106816 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 2000 |
| PublicationDate | 2020-02-01 |
| PublicationDateYYYYMMDD | 2020-02-01 |
| PublicationDate_xml | – month: 02 year: 2020 text: 2020-02-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | Computational statistics & data analysis |
| PublicationYear | 2020 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Kingma, D.P., Ba, J., 2014. Adam: A method for stochastic optimization, arXiv preprint Neal (b30) 1993 Fleiss, Levin, Paik (b10) 2003 Choi, Kwon, Lee, Kim, Paik, Won (b4) 2016 Der Kiureghian, Ditlevsen (b7) 2009; 31 Leibig, Allken, Ayhan, Berens, Wahl (b23) 2017; 7 Litjens, Kooi, Bejnordi, Setio, Ciompi, Ghafoorian, van der Laak, van Ginneken, Sánchez (b25) 2017; 42 Chen, Fox, Guestrin (b3) 2014 Li, Turner (b24) 2016 Miller, Foti, Adams (b29) 2017 Graves (b12) 2011 Kingma, D.P., Welling, M., 2013. Auto-encoding variational bayes, arXiv preprint Wang, Blei (b38) 2018 Winzeck, Hakim, McKinley, Pinto, Alves, Silva, Pisov, Krivov, Belyaev, Monteiro (b40) 2018; 9 Gal (b11) 2016 MacKay (b27) 1992; 4 He, Zhang, Ren, Sun (b14) 2015 2015. Kaggle, Kaggle diabetic retinopathy detection competition. Clevert, D.-A., Unterthiner, T., Hochreiter, S., 2015. Fast and accurate deep network learning by exponential linear units (elus), arXiv preprint Niemeijer, Staal, van Ginneken, Loog, Abramoff (b31) 2004 Jungo, Meier, Ermis, Herrmann, Reyes (b16) 2018 Louizos, Welling (b26) 2017 Esteva, Kuprel, Novoa, Ko, Swetter, Blau, Thrun (b9) 2017; 542 Guo, Pleiss, Sun, Weinberger (b13) 2017 Tanno, Worrall, Ghosh, Kaden, Sotiropoulos, Criminisi, Alexander (b36) 2017 [Online; accessed 31.07.18]. Dieng, Tran, Ranganath, Paisley, Blei (b8) 2017 Pereyra, G., Tucker, G., Chorowski, J., Kaiser, Ł., Hinton, G., 2017. Regularizing neural networks by penalizing confident output distributions, arXiv preprint Blundell, Cornebise, Kavukcuoglu, Wierstra (b2) 2015 Chollet, F., 2015. keras Welling, Teh (b39) 2011 . Kamnitsas, Ledig, Newcombe, Simpson, Kane, Menon, Rueckert, Glocker (b18) 2017; 36 Ayhan, Berens (b1) 2018 Teye, M., Azizpour, H., Smith, K., 2018. Bayesian Uncertainty Estimation for Batch Normalized Deep Networks, arXiv preprint Staal, Abràmoff, Niemeijer, Viergever, Van Ginneken (b35) 2004; 23 Lee (b22) 2000; 13 Ronneberger, Fischer, Brox (b33) 2015 Kendall, Gal (b19) 2017 Maier, Menze, von der Gablentz, Häni, Heinrich, Liebrand, Winzeck, Basit, Bentley, Chen (b28) 2017; 35 Siva Sundhara Raja, Vasuki (b34) 2015; 2015 Ioffe, Szegedy (b15) 2015 Litjens (10.1016/j.csda.2019.106816_b25) 2017; 42 Welling (10.1016/j.csda.2019.106816_b39) 2011 10.1016/j.csda.2019.106816_b6 Tanno (10.1016/j.csda.2019.106816_b36) 2017 10.1016/j.csda.2019.106816_b5 Der Kiureghian (10.1016/j.csda.2019.106816_b7) 2009; 31 Esteva (10.1016/j.csda.2019.106816_b9) 2017; 542 Guo (10.1016/j.csda.2019.106816_b13) 2017 Maier (10.1016/j.csda.2019.106816_b28) 2017; 35 Graves (10.1016/j.csda.2019.106816_b12) 2011 10.1016/j.csda.2019.106816_b21 Louizos (10.1016/j.csda.2019.106816_b26) 2017 MacKay (10.1016/j.csda.2019.106816_b27) 1992; 4 Dieng (10.1016/j.csda.2019.106816_b8) 2017 10.1016/j.csda.2019.106816_b20 Siva Sundhara Raja (10.1016/j.csda.2019.106816_b34) 2015; 2015 Staal (10.1016/j.csda.2019.106816_b35) 2004; 23 Kendall (10.1016/j.csda.2019.106816_b19) 2017 Miller (10.1016/j.csda.2019.106816_b29) 2017 Niemeijer (10.1016/j.csda.2019.106816_b31) 2004 Li (10.1016/j.csda.2019.106816_b24) 2016 He (10.1016/j.csda.2019.106816_b14) 2015 Ronneberger (10.1016/j.csda.2019.106816_b33) 2015 Ioffe (10.1016/j.csda.2019.106816_b15) 2015 Gal (10.1016/j.csda.2019.106816_b11) 2016 Fleiss (10.1016/j.csda.2019.106816_b10) 2003 Lee (10.1016/j.csda.2019.106816_b22) 2000; 13 10.1016/j.csda.2019.106816_b32 Wang (10.1016/j.csda.2019.106816_b38) 2018 Winzeck (10.1016/j.csda.2019.106816_b40) 2018; 9 10.1016/j.csda.2019.106816_b37 10.1016/j.csda.2019.106816_b17 Jungo (10.1016/j.csda.2019.106816_b16) 2018 Kamnitsas (10.1016/j.csda.2019.106816_b18) 2017; 36 Ayhan (10.1016/j.csda.2019.106816_b1) 2018 Blundell (10.1016/j.csda.2019.106816_b2) 2015 Choi (10.1016/j.csda.2019.106816_b4) 2016 Leibig (10.1016/j.csda.2019.106816_b23) 2017; 7 Chen (10.1016/j.csda.2019.106816_b3) 2014 Neal (10.1016/j.csda.2019.106816_b30) 1993 |
| References_xml | – start-page: 681 year: 2011 end-page: 688 ident: b39 article-title: Bayesian Learning via stochastic gradient langevin dynamics publication-title: International Conference on Machine Learning – reference: Chollet, F., 2015. keras, – year: 2003 ident: b10 publication-title: Statistical Methods for Rates and Proportions – volume: 36 start-page: 61 year: 2017 end-page: 78 ident: b18 article-title: Efficient multi-scale 3d cnn with fully connected crf for accurate brain lesion segmentation publication-title: Med. Image Anal. – volume: 7 start-page: 17816 year: 2017 ident: b23 article-title: Leveraging uncertainty information from deep neural networks for disease detection publication-title: Sci. Rep. – volume: 35 start-page: 250 year: 2017 end-page: 269 ident: b28 article-title: Isles 2015-a public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI publication-title: Med. Image Anal. – start-page: 475 year: 1993 end-page: 482 ident: b30 article-title: Bayesian Learning via stochastic dynamics publication-title: Advances in Neural Information Processing Systems – start-page: 448 year: 2015 end-page: 456 ident: b15 article-title: Batch normalization: Accelerating deep network training by reducing internal covariate shift publication-title: International Conference on Machine Learning – reference: Pereyra, G., Tucker, G., Chorowski, J., Kaiser, Ł., Hinton, G., 2017. Regularizing neural networks by penalizing confident output distributions, arXiv preprint – reference: Clevert, D.-A., Unterthiner, T., Hochreiter, S., 2015. Fast and accurate deep network learning by exponential linear units (elus), arXiv preprint – start-page: 1683 year: 2014 end-page: 1691 ident: b3 article-title: Stochastic gradient hamiltonian monte carlo publication-title: International Conference on Machine Learning – start-page: 648 year: 2004 end-page: 656 ident: b31 article-title: Comparative study of retinal vessel segmentation methods on a new publicly available database publication-title: SPIE Medical Imaging, Vol. 5370 – start-page: 1613 year: 2015 end-page: 1622 ident: b2 article-title: Weight uncertainty in neural networks publication-title: International Conference on Machine Learning – start-page: 1026 year: 2015 end-page: 1034 ident: b14 article-title: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification publication-title: Proceedings of the IEEE International Conference on Computer Vision – year: 2018 ident: b1 article-title: Test-time data augmentation for estimation of heteroscedastic aleatoric uncertainty in deep neural networks publication-title: International Conference on Medical Imaging with Deep Learning – year: 2018 ident: b16 article-title: Uncertainty-driven sanity check: Application to postoperative brain tumor Cavity segmentation publication-title: International Conference on Medical Imaging with Deep Learning – start-page: 231 year: 2016 end-page: 243 ident: b4 article-title: Ensemble of deep convolutional neural networks for prognosis of ischemic stroke publication-title: International Workshop on Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries – start-page: 2348 year: 2011 end-page: 2356 ident: b12 article-title: Practical variational inference for neural networks publication-title: Advances in Neural Information Processing Systems – start-page: 1321 year: 2017 end-page: 1330 ident: b13 article-title: On Calibration of modern neural networks publication-title: International Conference on Machine Learning – volume: 23 start-page: 501 year: 2004 end-page: 509 ident: b35 article-title: Ridge-based vessel segmentation in color images of the retina publication-title: IEEE Trans. Med. Imaging – start-page: 2420 year: 2017 end-page: 2429 ident: b29 article-title: Variational boosting: Iteratively refining posterior approximations publication-title: International Conference on Machine Learning – start-page: 234 year: 2015 end-page: 241 ident: b33 article-title: U-net: Convolutional networks for biomedical image segmentation publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – volume: 31 start-page: 105 year: 2009 end-page: 112 ident: b7 article-title: Aleatory or epistemic? does it matter? publication-title: Struct. Saf. – volume: 42 start-page: 60 year: 2017 end-page: 88 ident: b25 article-title: A survey on deep learning in medical image analysis publication-title: Med. Image Anal. – reference: . – start-page: 1 year: 2018 end-page: 85 ident: b38 article-title: Frequentist consistency of variational Bayes publication-title: J. Amer. Statist. Assoc. – volume: 4 start-page: 448 year: 1992 end-page: 472 ident: b27 article-title: A practical Bayesian framework for backpropagation networks publication-title: Neural Comput. – reference: Kingma, D.P., Ba, J., 2014. Adam: A method for stochastic optimization, arXiv preprint – start-page: 5580 year: 2017 end-page: 5590 ident: b19 article-title: What uncertainties do we need in bayesian deep learning for computer vision? publication-title: Advances in Neural Information Processing Systems – year: 2016 ident: b11 article-title: Uncertainty in Deep Learning – start-page: 611 year: 2017 end-page: 619 ident: b36 article-title: Bayesian image quality transfer with cnns: Exploring uncertainty in dmri super-resolution publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention – start-page: 1073 year: 2016 end-page: 1081 ident: b24 article-title: Rényi Divergence variational inference publication-title: Advances in Neural Information Processing Systems – reference: Teye, M., Azizpour, H., Smith, K., 2018. Bayesian Uncertainty Estimation for Batch Normalized Deep Networks, arXiv preprint – start-page: 2729 year: 2017 end-page: 2738 ident: b8 article-title: Variational inference via publication-title: Advances in Neural Information Processing Systems – volume: 9 year: 2018 ident: b40 article-title: ISLES 2016 and 2017-benchmarking ischemic stroke lesion outcome prediction based on multispectral MRI publication-title: Front. Neurol. – reference: [Online; accessed 31.07.18]. – volume: 2015 year: 2015 ident: b34 article-title: Automatic detection of blood vessels in retinal images for diabetic retinopathy diagnosis publication-title: Comput. Math. Methods Med. – reference: 2015. Kaggle, Kaggle diabetic retinopathy detection competition. – start-page: 2218 year: 2017 end-page: 2227 ident: b26 article-title: Multiplicative normalizing flows for variational Bayesian neural networks publication-title: International Conference on Machine Learning – reference: Kingma, D.P., Welling, M., 2013. Auto-encoding variational bayes, arXiv preprint – volume: 13 start-page: 629 year: 2000 end-page: 642 ident: b22 article-title: Consistency of posterior distributions for neural networks publication-title: Neural Netw. – volume: 542 start-page: 115 year: 2017 ident: b9 article-title: Dermatologist-level classification of skin cancer with deep neural networks publication-title: Nature – year: 2018 ident: 10.1016/j.csda.2019.106816_b1 article-title: Test-time data augmentation for estimation of heteroscedastic aleatoric uncertainty in deep neural networks – start-page: 611 year: 2017 ident: 10.1016/j.csda.2019.106816_b36 article-title: Bayesian image quality transfer with cnns: Exploring uncertainty in dmri super-resolution – year: 2016 ident: 10.1016/j.csda.2019.106816_b11 – start-page: 2218 year: 2017 ident: 10.1016/j.csda.2019.106816_b26 article-title: Multiplicative normalizing flows for variational Bayesian neural networks – start-page: 2348 year: 2011 ident: 10.1016/j.csda.2019.106816_b12 article-title: Practical variational inference for neural networks – volume: 23 start-page: 501 issue: 4 year: 2004 ident: 10.1016/j.csda.2019.106816_b35 article-title: Ridge-based vessel segmentation in color images of the retina publication-title: IEEE Trans. Med. Imaging doi: 10.1109/TMI.2004.825627 – ident: 10.1016/j.csda.2019.106816_b6 – ident: 10.1016/j.csda.2019.106816_b21 – volume: 42 start-page: 60 year: 2017 ident: 10.1016/j.csda.2019.106816_b25 article-title: A survey on deep learning in medical image analysis publication-title: Med. Image Anal. doi: 10.1016/j.media.2017.07.005 – ident: 10.1016/j.csda.2019.106816_b17 – volume: 4 start-page: 448 issue: 3 year: 1992 ident: 10.1016/j.csda.2019.106816_b27 article-title: A practical Bayesian framework for backpropagation networks publication-title: Neural Comput. doi: 10.1162/neco.1992.4.3.448 – volume: 9 year: 2018 ident: 10.1016/j.csda.2019.106816_b40 article-title: ISLES 2016 and 2017-benchmarking ischemic stroke lesion outcome prediction based on multispectral MRI publication-title: Front. Neurol. doi: 10.3389/fneur.2018.00679 – start-page: 448 year: 2015 ident: 10.1016/j.csda.2019.106816_b15 article-title: Batch normalization: Accelerating deep network training by reducing internal covariate shift – start-page: 234 year: 2015 ident: 10.1016/j.csda.2019.106816_b33 article-title: U-net: Convolutional networks for biomedical image segmentation – start-page: 2420 year: 2017 ident: 10.1016/j.csda.2019.106816_b29 article-title: Variational boosting: Iteratively refining posterior approximations – start-page: 475 year: 1993 ident: 10.1016/j.csda.2019.106816_b30 article-title: Bayesian Learning via stochastic dynamics – start-page: 681 year: 2011 ident: 10.1016/j.csda.2019.106816_b39 article-title: Bayesian Learning via stochastic gradient langevin dynamics – start-page: 1683 year: 2014 ident: 10.1016/j.csda.2019.106816_b3 article-title: Stochastic gradient hamiltonian monte carlo – volume: 7 start-page: 17816 issue: 1 year: 2017 ident: 10.1016/j.csda.2019.106816_b23 article-title: Leveraging uncertainty information from deep neural networks for disease detection publication-title: Sci. Rep. doi: 10.1038/s41598-017-17876-z – volume: 35 start-page: 250 year: 2017 ident: 10.1016/j.csda.2019.106816_b28 article-title: Isles 2015-a public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI publication-title: Med. Image Anal. doi: 10.1016/j.media.2016.07.009 – volume: 2015 year: 2015 ident: 10.1016/j.csda.2019.106816_b34 article-title: Automatic detection of blood vessels in retinal images for diabetic retinopathy diagnosis publication-title: Comput. Math. Methods Med. – volume: 36 start-page: 61 year: 2017 ident: 10.1016/j.csda.2019.106816_b18 article-title: Efficient multi-scale 3d cnn with fully connected crf for accurate brain lesion segmentation publication-title: Med. Image Anal. doi: 10.1016/j.media.2016.10.004 – volume: 13 start-page: 629 issue: 6 year: 2000 ident: 10.1016/j.csda.2019.106816_b22 article-title: Consistency of posterior distributions for neural networks publication-title: Neural Netw. doi: 10.1016/S0893-6080(00)00045-9 – ident: 10.1016/j.csda.2019.106816_b32 – start-page: 648 year: 2004 ident: 10.1016/j.csda.2019.106816_b31 article-title: Comparative study of retinal vessel segmentation methods on a new publicly available database – start-page: 231 year: 2016 ident: 10.1016/j.csda.2019.106816_b4 article-title: Ensemble of deep convolutional neural networks for prognosis of ischemic stroke – year: 2003 ident: 10.1016/j.csda.2019.106816_b10 – start-page: 1321 year: 2017 ident: 10.1016/j.csda.2019.106816_b13 article-title: On Calibration of modern neural networks – start-page: 2729 year: 2017 ident: 10.1016/j.csda.2019.106816_b8 article-title: Variational inference via χ upper bound minimization – start-page: 1026 year: 2015 ident: 10.1016/j.csda.2019.106816_b14 article-title: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification – ident: 10.1016/j.csda.2019.106816_b20 – ident: 10.1016/j.csda.2019.106816_b5 – volume: 542 start-page: 115 issue: 7639 year: 2017 ident: 10.1016/j.csda.2019.106816_b9 article-title: Dermatologist-level classification of skin cancer with deep neural networks publication-title: Nature doi: 10.1038/nature21056 – start-page: 5580 year: 2017 ident: 10.1016/j.csda.2019.106816_b19 article-title: What uncertainties do we need in bayesian deep learning for computer vision? – start-page: 1073 year: 2016 ident: 10.1016/j.csda.2019.106816_b24 article-title: Rényi Divergence variational inference – ident: 10.1016/j.csda.2019.106816_b37 – start-page: 1613 year: 2015 ident: 10.1016/j.csda.2019.106816_b2 article-title: Weight uncertainty in neural networks – year: 2018 ident: 10.1016/j.csda.2019.106816_b16 article-title: Uncertainty-driven sanity check: Application to postoperative brain tumor Cavity segmentation – start-page: 1 issue: just-accepted year: 2018 ident: 10.1016/j.csda.2019.106816_b38 article-title: Frequentist consistency of variational Bayes publication-title: J. Amer. Statist. Assoc. doi: 10.1080/01621459.2017.1303386 – volume: 31 start-page: 105 issue: 2 year: 2009 ident: 10.1016/j.csda.2019.106816_b7 article-title: Aleatory or epistemic? does it matter? publication-title: Struct. Saf. doi: 10.1016/j.strusafe.2008.06.020 |
| SSID | ssj0002478 |
| Score | 2.6465883 |
| Snippet | Most recent research of deep neural networks in the field of computer vision has focused on improving performances of point predictions by developing network... |
| SourceID | proquest crossref elsevier |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 106816 |
| SubjectTerms | Aleatoric and epistemic uncertainty algorithms Bayesian neural network Bayesian theory computer vision data collection Ischemic stroke lesion segmentation neural networks prediction Retinal blood vessel segmentation stroke uncertainty Uncertainty quantification |
| Title | Uncertainty quantification using Bayesian neural networks in classification: Application to biomedical image segmentation |
| URI | https://dx.doi.org/10.1016/j.csda.2019.106816 https://www.proquest.com/docview/2305223368 |
| Volume | 142 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV07T8MwELYQLDAgnuItI7Gh0CZOnIStIFChggGo6Gb5FVREUyDp0IXfzl2cFIEQA1MkK7asu8v5u_i7O0KOrPb9DP_aWD-TXphEmSeZASBnTcRlYADxY3LyzS3v9sPrQTSYI-dNLgzSKmvf73x65a3rkVYtzdbrcNi6RwJ9GsZgQ2CknA0wgz2M0cpPPr5oHkHovDHW98a368QZx_HShcHaQ34KAzzBnue_H04_3HR19lyukOUaNNKO29cqmbP5Glm6mVVcLdbJtA87r273yyl9m0jHAarETpHb_kTP5NRixiTFEpawWu4I4AUd5lQjhp5NOKWdr2ttWo6py9FHddLhCBwQLezTqE5ayjdI__Li4bzr1W0VPM0YKz2ljYpSg0m4bWUSbNIJioq5hmCY6VTFgLh8ZZiRUWIAfWQ8zpgFAYFP5UEg2SaZz8e53SLU-lIqztophJGhCVPFE6V1LFksE9vW4TbxG3kKXdccx9YXL6Ihlz0L1IFAHQing21yPJvz6ipu_Pl21KhJfLMbAUfCn_MOG50K-KDwlkTmdjwpBMRkgEkZ48nOP9feJYsBRuUVt3uPzJfvE7sP0KVUB5VtHpCFzlWve4vP3t1j7xPfTvGc |
| linkProvider | Elsevier |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwEB6VcgAOiKfaUsBIcEJhN3HiJJU4lEe1pd1e6Ep7M36lWtRmC8mq2gt_ij_YmdjZCoR6QOrVsi1rZjwP-5sZgNfOxHFFrzYurlSUFlkVKW7RkXM2Eyqx6PFTcvL4SIwm6ZdpNl2D330uDMEqg-73Or3T1mFkEKg5OJ_NBl8JQF-mOcoQCqng04CsPHDLC4zbmvf7n5DJb5Jk7_Pxx1EUWgtEhnPeRtpYnZWWElGH2hbUqBIPmwuDASE3pc7R64i15VZlhUULXIm84g7DJ9QrIkkUx31vwe0U1QW1TXj36wpXkqRe_VNBcTpeyNTxoDLTWCp2FJc4IApqsv5va_iXXeiM3d4DuB-8VLbrCfEQ1lz9CO6NVyVem8ewnCCpOjhBu2Q_FsqDjjo-MwLTn7APaukoRZNRzUzcrfaI84bNambIaV8t2GG7V__orJ0zXxSA5IfNzlDjscadnIUsqfoJTG6E2E9hvZ7XbgOYi5XSgg9LjFtTm5ZaFNqYXPFcFW5o0k2Ie3pKE4qcU6-NU9mj2b5L4oEkHkjPg014u1pz7kt8XDs769kk_xBUiTbo2nWvep5KvMH0LaNqN180EoNAdII5F8XWf-79Eu6MjseH8nD_6OAZ3E3oSaADlm_Devtz4Z6j39TqF52cMvh20xfjEg4HKl4 |
| 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=Uncertainty+quantification+using+Bayesian+neural+networks+in+classification%3A+Application+to+biomedical+image+segmentation&rft.jtitle=Computational+statistics+%26+data+analysis&rft.au=Kwon%2C+Yongchan&rft.au=Won%2C+Joong-Ho&rft.au=Kim%2C+Beom+Joon&rft.au=Paik%2C+Myunghee+Cho&rft.date=2020-02-01&rft.issn=0167-9473&rft.volume=142&rft.spage=106816&rft_id=info:doi/10.1016%2Fj.csda.2019.106816&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_csda_2019_106816 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0167-9473&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0167-9473&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0167-9473&client=summon |