Improving Computer-Aided Detection Using Convolutional Neural Networks and Random View Aggregation
Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a can...
Saved in:
| Published in | IEEE transactions on medical imaging Vol. 35; no. 5; pp. 1170 - 1181 |
|---|---|
| Main Authors | , , , , , , , |
| Format | Journal Article |
| Language | English |
| Published |
United States
IEEE
01.05.2016
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities ~ 100% of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively. |
|---|---|
| AbstractList | Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities ∼ 100% of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively. Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities ∼ 100% of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively.Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities ∼ 100% of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively. Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities ~ 100% of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively. Automated computer-aided detection (CADe) in medical imaging has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities but at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities of ~100% but at high FP levels. By leveraging existing CAD systems, coordinates of regions or volumes of interest (ROI or VOI) for lesion candidates are generated in this step and function as input for a second tier, which is our focus in this study. In this second stage, we generate N 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations with respect to each ROI’s centroid coordinates. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the trained ConvNets are employed to assign class (e.g., lesion, pathology) probabilities for a new set of N random views that are then averaged at each ROI to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three different data sets with different numbers of patients: 59 patients for sclerotic metastases detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve CADe performance markedly in all cases. CADe sensitivities improved from 57% to 70%, from 43% to 77% and from 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively. Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities [Formula Omitted] of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively. |
| Author | Kim, Lauren Jiamin Liu Roth, Holger R. Summers, Ronald M. Le Lu Seff, Ari Cherry, Kevin Jianhua Yao |
| Author_xml | – sequence: 1 givenname: Holger R. surname: Roth fullname: Roth, Holger R. email: holger.roth@nih.gov organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 2 surname: Le Lu fullname: Le Lu organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 3 surname: Jiamin Liu fullname: Jiamin Liu organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 4 surname: Jianhua Yao fullname: Jianhua Yao organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 5 givenname: Ari surname: Seff fullname: Seff, Ari organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 6 givenname: Kevin surname: Cherry fullname: Cherry, Kevin organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 7 givenname: Lauren surname: Kim fullname: Kim, Lauren organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA – sequence: 8 givenname: Ronald M. surname: Summers fullname: Summers, Ronald M. organization: Radiol. & Imaging Sci. Dept., Nat. Inst. of Health Clinical Center, Bethesda, MD, USA |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/26441412$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9UU1v1DAQtVAR3RbuSEgoEhcuWcZf-bggrbalXamAhFrEzXLiSXBJ4sVOtuLf4_1gBT1w8Vie98Zv3jsjJ4MbkJCXFOaUQvnu9uNqzoDKORMFKxk8ITMqZZEyKb6dkBmwvEgBMnZKzkK4B6BCQvmMnLJMCCoom5Fq1a-929ihTZauX08j-nRhDZrkAkesR-uG5C7s28PGddP2RXfJJ5z8rowPzv8IiR5M8iUerk--WnxIFm3rsdVb9HPytNFdwBeHek7uPlzeLq_Tm89Xq-XiJq0FsDE1lTF5SUWeZWAyRNHUTCCtaol51vBGa6BaiIZyUwrMgIuyriod7ybuCcjPyfv93PVU9WhqHMYoUa297bX_pZy26t_OYL-r1m1UzgVwLuKAt4cB3v2cMIyqt6HGrtMDuikomheFFJBDGaFvHkHv3eSjMTtUXsioj0fU678VHaX8sT8CYA-ovQvBY3OEUFDbhFVMWG0TVoeEIyV7RKntuPM57mS7_xFf7YkWEY__5Cx6LjP-G8uytEg |
| CODEN | ITMID4 |
| CitedBy_id | crossref_primary_10_1007_s00521_022_06973_4 crossref_primary_10_1111_den_13670 crossref_primary_10_1109_ACCESS_2019_2950286 crossref_primary_10_1016_j_cmpb_2020_105361 crossref_primary_10_1109_ACCESS_2019_2937993 crossref_primary_10_1155_2018_2391925 crossref_primary_10_1109_TBME_2018_2881952 crossref_primary_10_1109_ACCESS_2019_2960411 crossref_primary_10_1117_1_JEI_28_4_043023 crossref_primary_10_5114_pjr_2021_108257 crossref_primary_10_1109_ACCESS_2017_2788044 crossref_primary_10_1186_s13244_019_0832_5 crossref_primary_10_1016_j_media_2018_07_008 crossref_primary_10_1016_j_media_2018_03_002 crossref_primary_10_1016_j_media_2019_101541 crossref_primary_10_1097_BRS_0000000000004308 crossref_primary_10_1109_ACCESS_2020_2994762 crossref_primary_10_1109_ACCESS_2019_2920980 crossref_primary_10_1016_j_patrec_2019_08_003 crossref_primary_10_1155_2016_6584725 crossref_primary_10_1002_jmri_26534 crossref_primary_10_1016_j_compmedimag_2019_04_004 crossref_primary_10_1136_postgradmedj_2020_139620 crossref_primary_10_1016_j_jiph_2020_06_033 crossref_primary_10_1109_ACCESS_2021_3075294 crossref_primary_10_1080_03772063_2021_1905562 crossref_primary_10_1186_s13018_023_04509_7 crossref_primary_10_3390_jimaging8040097 crossref_primary_10_1177_1533033818802789 crossref_primary_10_1002_ima_22642 crossref_primary_10_1016_j_compbiomed_2023_107617 crossref_primary_10_1109_TMI_2020_2980839 crossref_primary_10_3390_diagnostics12092102 crossref_primary_10_1016_j_compmedimag_2022_102108 crossref_primary_10_1088_1742_6596_1815_1_012005 crossref_primary_10_1007_s00371_023_03030_6 crossref_primary_10_1016_j_bspc_2022_103501 crossref_primary_10_1088_1361_6560_ad965c crossref_primary_10_7717_peerj_6201 crossref_primary_10_1016_j_cmpb_2022_107095 crossref_primary_10_1016_j_eswa_2019_112957 crossref_primary_10_1109_ACCESS_2018_2846685 crossref_primary_10_1142_S0219843619500154 crossref_primary_10_1371_journal_pone_0206229 crossref_primary_10_1007_s13177_020_00225_2 crossref_primary_10_1016_j_neucom_2019_09_027 crossref_primary_10_1016_j_ejmp_2020_01_011 crossref_primary_10_1016_j_patcog_2018_02_026 crossref_primary_10_32604_cmes_2023_030236 crossref_primary_10_3390_en15249593 crossref_primary_10_1186_s13014_023_02311_7 crossref_primary_10_1007_s00521_021_06182_5 crossref_primary_10_1117_1_JEI_27_6_063006 crossref_primary_10_1007_s11042_022_12908_x crossref_primary_10_7717_peerj_8052 crossref_primary_10_1080_02564602_2019_1576550 crossref_primary_10_1109_MMUL_2018_2875861 crossref_primary_10_35713_aic_v3_i1_11 crossref_primary_10_1002_lio2_1008 crossref_primary_10_1117_1_JMI_5_3_036501 crossref_primary_10_1038_s41598_019_56989_5 crossref_primary_10_1002_ima_22517 crossref_primary_10_1088_1361_6560_ac35c7 crossref_primary_10_1016_j_jflm_2020_102066 crossref_primary_10_1016_j_neunet_2019_03_003 crossref_primary_10_1016_j_patcog_2017_08_004 crossref_primary_10_1038_s41598_017_13448_3 crossref_primary_10_1097_BRS_0000000000004889 crossref_primary_10_1109_ACCESS_2022_3160290 crossref_primary_10_1371_journal_pone_0285608 crossref_primary_10_1016_j_heliyon_2023_e13444 crossref_primary_10_1109_JBHI_2019_2940695 crossref_primary_10_35377_saucis_02_01_538249 crossref_primary_10_1007_s11042_023_17210_y crossref_primary_10_1016_j_media_2022_102550 crossref_primary_10_32604_cmes_2021_016817 crossref_primary_10_1016_j_media_2020_101896 crossref_primary_10_1109_TMI_2021_3112923 crossref_primary_10_32084_tekapr_2021_14_1_18 crossref_primary_10_3389_fonc_2021_805911 crossref_primary_10_1016_j_compbiomed_2024_109302 crossref_primary_10_1016_j_ins_2019_05_043 crossref_primary_10_1007_s10586_024_04532_1 crossref_primary_10_4103_ijpvm_ijpvm_373_22 crossref_primary_10_3390_app9163261 crossref_primary_10_1007_s10916_018_1072_9 crossref_primary_10_1109_TNNLS_2018_2790388 crossref_primary_10_6009_jjrt_2021_JSRT_77_8_787 crossref_primary_10_1109_ACCESS_2020_3042889 crossref_primary_10_1088_1361_6560_ab326a crossref_primary_10_1007_s12559_017_9487_z crossref_primary_10_1109_ACCESS_2022_3183604 crossref_primary_10_1155_2020_6687733 crossref_primary_10_1111_cas_14377 crossref_primary_10_12677_JISP_2018_71001 crossref_primary_10_1109_JBHI_2018_2818620 crossref_primary_10_1016_j_media_2022_102421 crossref_primary_10_1088_1361_6560_aa82ec crossref_primary_10_1109_ACCESS_2019_2906116 crossref_primary_10_1016_j_aap_2020_105716 crossref_primary_10_1016_j_csbj_2021_09_001 crossref_primary_10_1016_j_neunet_2020_09_004 crossref_primary_10_2174_1573405614666181017122109 crossref_primary_10_1007_s00521_017_3138_x crossref_primary_10_1016_j_drudis_2019_07_006 crossref_primary_10_1007_s12553_024_00879_y crossref_primary_10_1016_j_media_2020_101884 crossref_primary_10_1016_j_ultras_2016_08_004 crossref_primary_10_1109_TMI_2019_2936244 crossref_primary_10_1148_radiol_2019190372 crossref_primary_10_1109_TMI_2020_3031289 crossref_primary_10_1002_mp_13642 crossref_primary_10_1111_exsy_12882 crossref_primary_10_1002_mp_13886 crossref_primary_10_3390_s21186187 crossref_primary_10_1016_j_mex_2024_103034 crossref_primary_10_1021_acs_jpca_8b09376 crossref_primary_10_1016_j_bbe_2021_01_005 crossref_primary_10_1109_JBHI_2020_3017540 crossref_primary_10_1177_0161734621989598 crossref_primary_10_3390_app132111898 crossref_primary_10_1038_s41598_017_08040_8 crossref_primary_10_1259_dmfr_20200375 crossref_primary_10_1109_ACCESS_2019_2896409 crossref_primary_10_1109_TMI_2016_2553401 crossref_primary_10_1117_1_JMI_4_4_041304 crossref_primary_10_1186_s13550_017_0260_9 crossref_primary_10_1016_j_cmpb_2018_01_017 crossref_primary_10_3748_wjg_v27_i21_2681 crossref_primary_10_4015_S1016237222500119 crossref_primary_10_1002_mp_15010 crossref_primary_10_3390_s19194251 crossref_primary_10_1002_ima_22287 crossref_primary_10_1109_ACCESS_2024_3396999 crossref_primary_10_1186_s12968_018_0516_1 crossref_primary_10_1007_s00521_020_04787_w crossref_primary_10_1007_s11548_020_02275_z crossref_primary_10_1109_TMI_2022_3171418 crossref_primary_10_3390_electronics9081237 crossref_primary_10_1109_RBME_2021_3075500 crossref_primary_10_1109_ACCESS_2020_2997946 crossref_primary_10_1097_RLI_0000000000000600 crossref_primary_10_1002_cncr_31630 crossref_primary_10_1109_TCYB_2017_2671898 crossref_primary_10_1016_j_neunet_2020_05_003 crossref_primary_10_3390_cancers14082008 crossref_primary_10_1016_j_ejmp_2021_02_024 crossref_primary_10_1007_s00138_020_01074_5 crossref_primary_10_1038_srep24454 crossref_primary_10_1109_ACCESS_2018_2874803 crossref_primary_10_1111_exsy_12789 crossref_primary_10_1109_JBHI_2018_2879449 crossref_primary_10_31127_tuje_652358 crossref_primary_10_1109_JBHI_2017_2705583 crossref_primary_10_1016_j_cmpb_2022_107201 crossref_primary_10_1002_jemt_23773 crossref_primary_10_1016_j_compeleceng_2019_106449 crossref_primary_10_1016_j_asoc_2024_111544 crossref_primary_10_1016_j_compbiomed_2022_106059 crossref_primary_10_1109_JPROC_2021_3054390 crossref_primary_10_1007_s00256_021_03873_x crossref_primary_10_1016_j_neucom_2018_09_013 crossref_primary_10_1089_sur_2019_154 crossref_primary_10_1016_j_compbiomed_2016_11_003 crossref_primary_10_1109_JBHI_2018_2808199 crossref_primary_10_1002_mp_15033 crossref_primary_10_1109_TMI_2020_3021493 crossref_primary_10_3233_XST_200656 crossref_primary_10_1002_cnm_3601 crossref_primary_10_3390_math12070958 crossref_primary_10_3348_jksr_2020_0158 crossref_primary_10_17341_gazimmfd_416530 crossref_primary_10_1088_1361_6560_ac5297 crossref_primary_10_3389_fnins_2017_00310 crossref_primary_10_1098_rsif_2017_0387 crossref_primary_10_3233_XST_17302 crossref_primary_10_3390_healthcare10030422 crossref_primary_10_1142_S021951942340081X crossref_primary_10_1109_ACCESS_2022_3202560 crossref_primary_10_1088_1742_6596_2089_1_012058 crossref_primary_10_1055_a_2157_6670 crossref_primary_10_3390_app132312725 crossref_primary_10_1016_j_media_2020_101840 crossref_primary_10_1038_s41746_020_00353_9 crossref_primary_10_1093_jamia_ocab046 crossref_primary_10_1002_ima_22371 crossref_primary_10_1109_TCSVT_2024_3427645 crossref_primary_10_3390_bioengineering10080948 crossref_primary_10_1142_S0219691321500545 crossref_primary_10_1109_TASE_2019_2936645 crossref_primary_10_1016_j_bspc_2020_101952 crossref_primary_10_1016_j_neucom_2021_01_135 crossref_primary_10_1164_rccm_201705_0860OC crossref_primary_10_1109_JBHI_2019_2894713 crossref_primary_10_2174_1573405615666190716122040 crossref_primary_10_1007_s00521_022_06928_9 crossref_primary_10_1016_j_adhoc_2021_102438 crossref_primary_10_1016_j_media_2017_07_005 crossref_primary_10_1016_j_media_2017_03_002 crossref_primary_10_3390_jimaging10100239 crossref_primary_10_1016_j_compbiomed_2018_05_018 crossref_primary_10_1016_j_rpor_2020_03_015 crossref_primary_10_1155_2021_1994764 crossref_primary_10_1038_s41598_021_84287_6 crossref_primary_10_1088_1361_6560_ac4667 crossref_primary_10_1016_j_patrec_2022_03_022 crossref_primary_10_1109_TIE_2019_2942548 crossref_primary_10_1002_jemt_24008 crossref_primary_10_1109_TBME_2016_2631245 crossref_primary_10_1117_1_JMI_4_4_044501 crossref_primary_10_1002_mp_14915 crossref_primary_10_1038_s41598_017_12320_8 crossref_primary_10_1016_j_compmedimag_2017_06_006 crossref_primary_10_3389_fcvm_2023_1173769 crossref_primary_10_1186_s12920_018_0416_0 crossref_primary_10_35940_ijrte_C6386_0910321 crossref_primary_10_1109_TMI_2020_2975853 crossref_primary_10_1109_ACCESS_2020_2974617 crossref_primary_10_1002_mp_14397 crossref_primary_10_1109_ACCESS_2019_2916557 crossref_primary_10_32604_cmc_2021_016536 crossref_primary_10_1016_j_compmedimag_2022_102050 crossref_primary_10_1186_s12968_021_00824_2 crossref_primary_10_3174_ajnr_A6883 crossref_primary_10_1155_2020_4864835 crossref_primary_10_1007_s10278_018_0160_1 crossref_primary_10_1088_1361_6560_abcd17 crossref_primary_10_1089_thy_2018_0082 crossref_primary_10_1364_BOE_9_003049 crossref_primary_10_1002_spe_2878 crossref_primary_10_1007_s10462_021_10074_4 crossref_primary_10_1016_j_media_2018_10_010 crossref_primary_10_1007_s00330_022_08741_3 crossref_primary_10_1088_1361_6560_ac944d crossref_primary_10_2174_1573405616666201217112521 crossref_primary_10_1016_j_jestch_2022_101214 crossref_primary_10_1088_1361_6560_aaf241 crossref_primary_10_1002_ima_22581 crossref_primary_10_3233_JIFS_219312 crossref_primary_10_1016_j_sciaf_2023_e01649 crossref_primary_10_1016_j_neucom_2016_06_077 crossref_primary_10_1007_s00464_022_09470_w crossref_primary_10_1109_TIFS_2021_3059340 crossref_primary_10_1016_j_media_2021_102082 crossref_primary_10_1109_ACCESS_2019_2907040 crossref_primary_10_1007_s00138_022_01347_1 crossref_primary_10_1016_j_jbi_2021_103803 crossref_primary_10_1007_s00521_021_06368_x crossref_primary_10_1016_j_artmed_2020_101881 crossref_primary_10_1146_annurev_bioeng_071516_044442 crossref_primary_10_1016_j_media_2018_01_006 crossref_primary_10_1007_s10462_019_09788_3 crossref_primary_10_1148_rg_2017170077 crossref_primary_10_3390_cancers14020277 crossref_primary_10_1016_j_gmod_2019_101031 crossref_primary_10_3233_IDT_190083 crossref_primary_10_1109_TMI_2020_3047598 crossref_primary_10_1155_2022_2472988 crossref_primary_10_1007_s11548_021_02478_y crossref_primary_10_1016_j_cpet_2021_09_010 crossref_primary_10_1109_JBHI_2022_3161466 crossref_primary_10_1186_s12968_020_00678_0 crossref_primary_10_2174_1573405616666210108122048 crossref_primary_10_3390_jimaging7090190 crossref_primary_10_1049_iet_ipr_2016_0526 crossref_primary_10_1016_j_jksuci_2019_11_013 crossref_primary_10_1200_EDBK_350652 crossref_primary_10_1103_PhysRevX_11_011052 crossref_primary_10_1002_mp_12155 crossref_primary_10_1007_s12272_019_01162_9 crossref_primary_10_1088_1742_6596_1831_1_012006 crossref_primary_10_1038_s41746_021_00411_w crossref_primary_10_1109_ACCESS_2021_3137317 crossref_primary_10_1155_2022_8616535 crossref_primary_10_1364_BOE_8_002732 crossref_primary_10_1007_s00521_022_06960_9 crossref_primary_10_1142_S2196888823500057 crossref_primary_10_1016_j_ajo_2018_10_007 crossref_primary_10_1016_j_media_2023_102988 crossref_primary_10_1016_j_bspc_2024_106667 crossref_primary_10_2478_amns_2023_2_00504 crossref_primary_10_1109_ACCESS_2020_3029907 crossref_primary_10_1016_j_ceh_2022_04_001 crossref_primary_10_1016_j_ejmp_2020_02_012 crossref_primary_10_1016_j_bbe_2022_06_003 crossref_primary_10_3174_ajnr_A5667 crossref_primary_10_1038_s41568_020_00327_9 crossref_primary_10_1007_s11042_022_12376_3 crossref_primary_10_1109_JBHI_2019_2912659 crossref_primary_10_3389_fonc_2021_588010 crossref_primary_10_1016_j_media_2022_102444 crossref_primary_10_1186_s12880_021_00703_3 crossref_primary_10_1109_TPAMI_2020_3033990 crossref_primary_10_1016_j_media_2022_102562 crossref_primary_10_3390_electronics11010055 crossref_primary_10_1002_int_22452 crossref_primary_10_1055_a_1718_4128 crossref_primary_10_1109_TMI_2017_2774044 crossref_primary_10_1017_S143192761700558X crossref_primary_10_1002_ima_22430 crossref_primary_10_1002_mp_13264 crossref_primary_10_1088_1742_6596_1706_1_012189 crossref_primary_10_1109_TMI_2019_2937271 crossref_primary_10_1016_j_media_2020_101909 crossref_primary_10_1016_j_jncc_2022_09_003 crossref_primary_10_1159_000540728 crossref_primary_10_1049_iet_ipr_2020_1048 crossref_primary_10_1109_JBHI_2016_2636665 crossref_primary_10_2214_AJR_17_18754 crossref_primary_10_26634_jcom_11_2_20132 crossref_primary_10_1007_s00330_020_06663_6 crossref_primary_10_1007_s11760_020_01820_2 crossref_primary_10_1016_j_jacr_2017_12_028 crossref_primary_10_3390_cancers15071958 crossref_primary_10_1111_coin_12476 crossref_primary_10_1109_ACCESS_2019_2925863 crossref_primary_10_4018_IJCAC_2022010109 crossref_primary_10_1016_j_cmpb_2020_105611 crossref_primary_10_1063_5_0025462 crossref_primary_10_1007_s11042_017_5449_4 crossref_primary_10_1088_1555_6611_abb596 crossref_primary_10_1016_j_patcog_2018_05_026 crossref_primary_10_1016_j_dsp_2024_104411 crossref_primary_10_1016_j_jstrokecerebrovasdis_2022_106382 crossref_primary_10_1016_j_media_2018_09_004 crossref_primary_10_1142_S0219519422400322 crossref_primary_10_2147_TCRM_S483907 crossref_primary_10_3390_s24030928 crossref_primary_10_1088_1742_6596_1505_1_012018 crossref_primary_10_1016_j_isci_2022_104277 crossref_primary_10_1016_j_isprsjprs_2019_10_011 crossref_primary_10_1111_nan_12770 crossref_primary_10_1177_02841851241263066 crossref_primary_10_1109_TMI_2018_2837002 crossref_primary_10_1007_s11548_021_02335_y crossref_primary_10_1109_TEM_2021_3073018 crossref_primary_10_1007_s11548_018_1835_2 crossref_primary_10_1016_j_media_2022_102463 |
| Cites_doi | 10.1162/neco.1989.1.4.541 10.1117/12.2008282 10.1109/TMI.2007.892510 10.1109/CVPR.2014.81 10.1007/978-3-319-10404-1_65 10.1109/CVPR.2011.5995359 10.1007/978-3-319-19992-4_46 10.1053/j.gastro.2005.08.054 10.1145/1390156.1390258 10.1186/1475-925X-13-41 10.1148/radiol.2252011619 10.1056/NEJMoa0800996 10.1007/978-3-319-14104-6_16 10.1148/radiol.13121351 10.1109/CVPR.2014.223 10.1109/TMI.2011.2168234 10.1016/j.patcog.2008.09.034 10.1162/neco.2009.10-08-881 10.1109/ISBI.2015.7163869 10.1016/S0031-3203(03)00192-4 10.1117/12.652288 10.3390/a3010021 10.1117/12.811101 10.1109/ICARCV.2014.7064414 10.1117/12.911700 10.1007/978-3-540-88693-8_34 10.1007/978-3-642-04271-3_122 10.1016/j.media.2012.11.001 10.1007/s00330-013-2774-5 10.1109/TMI.2009.2028576 10.1145/2063576.2064004 10.1016/j.beem.2008.01.011 10.1109/42.974920 10.1038/505146a 10.1109/CVPR.2010.5540008 10.1007/11866763_57 10.1145/1961189.1961199 10.1109/ISBI.2011.5872376 10.1007/978-3-319-14148-0_1 10.1117/12.594547 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2016 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2016 |
| DBID | 97E RIA RIE AAYXX CITATION CGR CUY CVF ECM EIF NPM 7QF 7QO 7QQ 7SC 7SE 7SP 7SR 7TA 7TB 7U5 8BQ 8FD F28 FR3 H8D JG9 JQ2 KR7 L7M L~C L~D NAPCQ P64 7X8 5PM |
| DOI | 10.1109/TMI.2015.2482920 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed Aluminium Industry Abstracts Biotechnology Research Abstracts Ceramic Abstracts Computer and Information Systems Abstracts Corrosion Abstracts Electronics & Communications Abstracts Engineered Materials Abstracts Materials Business File Mechanical & Transportation Engineering Abstracts Solid State and Superconductivity Abstracts METADEX Technology Research Database ANTE: Abstracts in New Technology & Engineering Engineering Research Database Aerospace Database Materials Research Database ProQuest Computer Science Collection Civil Engineering Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional Nursing & Allied Health Premium Biotechnology and BioEngineering Abstracts MEDLINE - Academic PubMed Central (Full Participant titles) |
| DatabaseTitle | CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) Materials Research Database Civil Engineering Abstracts Aluminium Industry Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Mechanical & Transportation Engineering Abstracts Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Ceramic Abstracts Materials Business File METADEX Biotechnology and BioEngineering Abstracts Computer and Information Systems Abstracts Professional Aerospace Database Nursing & Allied Health Premium Engineered Materials Abstracts Biotechnology Research Abstracts Solid State and Superconductivity Abstracts Engineering Research Database Corrosion Abstracts Advanced Technologies Database with Aerospace ANTE: Abstracts in New Technology & Engineering MEDLINE - Academic |
| DatabaseTitleList | MEDLINE MEDLINE - Academic Materials Research Database |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: EIF name: MEDLINE url: https://proxy.k.utb.cz/login?url=https://www.webofscience.com/wos/medline/basic-search sourceTypes: Index Database – sequence: 3 dbid: RIE name: IEEE Electronic Library (IEL) url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine Engineering |
| EISSN | 1558-254X |
| EndPage | 1181 |
| ExternalDocumentID | PMC7340334 4050811741 26441412 10_1109_TMI_2015_2482920 7279156 |
| Genre | orig-research Journal Article |
| GrantInformation_xml | – fundername: Intramural NIH HHS grantid: Z01 CL040004 – fundername: Intramural NIH HHS grantid: Z01 CL040003 |
| GroupedDBID | --- -DZ -~X .GJ 0R~ 29I 4.4 53G 5GY 5RE 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT ACPRK AENEX AETIX AFRAH AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD F5P HZ~ H~9 IBMZZ ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS RXW TAE TN5 VH1 AAYOK AAYXX CITATION RIG CGR CUY CVF ECM EIF NPM 7QF 7QO 7QQ 7SC 7SE 7SP 7SR 7TA 7TB 7U5 8BQ 8FD F28 FR3 H8D JG9 JQ2 KR7 L7M L~C L~D NAPCQ P64 7X8 5PM |
| ID | FETCH-LOGICAL-c402t-dbdd79147660d6ee4fc24e1bc5e76f3faa01a44f13d94e60349cbba94ed0270e3 |
| IEDL.DBID | RIE |
| ISSN | 0278-0062 1558-254X |
| IngestDate | Thu Aug 21 14:02:12 EDT 2025 Fri Jul 11 04:17:48 EDT 2025 Mon Jun 30 06:20:02 EDT 2025 Mon Jul 21 05:58:55 EDT 2025 Tue Jul 01 03:15:56 EDT 2025 Thu Apr 24 22:52:55 EDT 2025 Tue Aug 26 16:42:53 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Issue | 5 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/USG.html Personal use of this material is permitted. However, permission to use this material for any other purposes must be obtained from the IEEE by sending a request to pubs-permissions@ieee.org. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c402t-dbdd79147660d6ee4fc24e1bc5e76f3faa01a44f13d94e60349cbba94ed0270e3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0002-3662-8743 |
| PMID | 26441412 |
| PQID | 1787856033 |
| PQPubID | 85460 |
| PageCount | 12 |
| ParticipantIDs | pubmedcentral_primary_oai_pubmedcentral_nih_gov_7340334 crossref_citationtrail_10_1109_TMI_2015_2482920 crossref_primary_10_1109_TMI_2015_2482920 ieee_primary_7279156 proquest_miscellaneous_1788540709 proquest_journals_1787856033 pubmed_primary_26441412 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 2000 |
| PublicationDate | 2016-05-01 |
| PublicationDateYYYYMMDD | 2016-05-01 |
| PublicationDate_xml | – month: 05 year: 2016 text: 2016-05-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States – name: New York |
| PublicationTitle | IEEE transactions on medical imaging |
| PublicationTitleAbbrev | TMI |
| PublicationTitleAlternate | IEEE Trans Med Imaging |
| PublicationYear | 2016 |
| 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 | ref13 ref56 ref12 ref15 ref58 ref53 ref11 ref10 krizhevsky (ref14) 2012 ref16 tajbakhsh (ref52) 2015 yosinski (ref57) 2014 organization (ref1) 2014 tajbakhsh (ref55) 2015 szegedy (ref60) 2014 ref51 krizhevsky (ref31) 2014 ref46 ref45 ref48 ref47 ref42 ref41 ref44 ref43 ref49 ref8 ref7 ref9 ref4 ref3 ref5 ref40 simonyan (ref50) 2014 seff (ref6) 2014 ref35 ref37 ref36 ref33 ref32 prasoon (ref19) 2013 ciresan (ref18) 2012 ref2 ref39 ref38 park (ref54) 2015 cherry (ref25) 2014 liu (ref26) 2014 simonyan (ref59) 2014 ref24 ref23 ciresan (ref17) 2013 ref20 ref22 ref21 yao (ref34) 2006 ref27 hinton (ref29) 2012 srivastava (ref30) 2014; 15 wan (ref28) 2013 |
| References_xml | – ident: ref15 doi: 10.1162/neco.1989.1.4.541 – ident: ref46 doi: 10.1117/12.2008282 – year: 2015 ident: ref55 article-title: A comprehensive computer-aided polyp detection system for colonoscopy videos publication-title: Inf Process Med Imag – ident: ref7 doi: 10.1109/TMI.2007.892510 – start-page: 2843 year: 2012 ident: ref18 article-title: Deep neural networks segment neuronal membranes in electron microscopy images publication-title: Adv Neural Inf Process Syst – ident: ref58 doi: 10.1109/CVPR.2014.81 – ident: ref20 doi: 10.1007/978-3-319-10404-1_65 – volume: 15 start-page: 1929 year: 2014 ident: ref30 article-title: Dropout: A simple way to prevent neural networks from overfitting publication-title: J Mach Learn Res – ident: ref37 doi: 10.1109/CVPR.2011.5995359 – ident: ref53 doi: 10.1007/978-3-319-19992-4_46 – ident: ref27 doi: 10.1053/j.gastro.2005.08.054 – year: 2014 ident: ref31 article-title: One weird trick for parallelizing convolutional neural networks publication-title: arXiv preprint arXiv 1404 5997 – year: 2015 ident: ref54 article-title: Polyp detection in colonoscopy videos using deeply-learned hierarchical features publication-title: Seoul nat'l univ – ident: ref48 doi: 10.1145/1390156.1390258 – year: 2012 ident: ref29 article-title: Improving neural networks by preventing co-adaptation of feature detectors publication-title: arXiv preprint arXiv 1207 0580 – ident: ref12 doi: 10.1186/1475-925X-13-41 – ident: ref9 doi: 10.1148/radiol.2252011619 – year: 2014 ident: ref60 article-title: Going deeper with convolutions publication-title: CoRR abs/1409 4842 – ident: ref56 doi: 10.1056/NEJMoa0800996 – ident: ref40 doi: 10.1007/978-3-319-14104-6_16 – ident: ref4 doi: 10.1148/radiol.13121351 – ident: ref51 doi: 10.1109/CVPR.2014.223 – ident: ref44 doi: 10.1109/TMI.2011.2168234 – ident: ref39 doi: 10.1016/j.patcog.2008.09.034 – ident: ref23 doi: 10.1162/neco.2009.10-08-881 – ident: ref11 doi: 10.1109/ISBI.2015.7163869 – year: 2012 ident: ref14 article-title: ImageNet classification with deep convolutional neural networks publication-title: NIPS – ident: ref13 doi: 10.1016/S0031-3203(03)00192-4 – ident: ref35 doi: 10.1117/12.652288 – year: 2013 ident: ref19 article-title: Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network publication-title: MICCAI – start-page: 568 year: 2014 ident: ref50 article-title: Two-stream convolutional networks for action recognition in videos publication-title: Adv Neural Inform Process Syst – ident: ref38 doi: 10.3390/a3010021 – start-page: 390 year: 2006 ident: ref34 article-title: Automated spinal column extraction and partitioning publication-title: Proc IEEE Int Symp IEEE Biomed Imag Nano to Macro – ident: ref49 doi: 10.1117/12.811101 – ident: ref22 doi: 10.1109/ICARCV.2014.7064414 – start-page: 3320 year: 2014 ident: ref57 article-title: How transferable are features in deep neural networks? publication-title: Adv Neural Inform Process Syst – ident: ref3 doi: 10.1117/12.911700 – ident: ref42 doi: 10.1007/978-3-540-88693-8_34 – year: 2014 ident: ref59 article-title: Very deep convolutional networks for large-scale image recognition publication-title: arXiv preprint arXiv 1409 1556 – year: 2014 ident: ref1 publication-title: Cancer Fact Sheet – ident: ref43 doi: 10.1007/978-3-642-04271-3_122 – year: 2014 ident: ref25 article-title: Abdominal lymphadenopathy detection using random forest publication-title: SPIE Med Imag – year: 2015 ident: ref52 article-title: Computer-aided pulmonary embolism detection using a novel vessel-aligned multi-planar image representation and convolutional neural networks publication-title: Proc MICCAI – start-page: 544 year: 2014 ident: ref6 publication-title: MICCAI – year: 2013 ident: ref28 article-title: Regularization of neural networks using DropConnect publication-title: Proc Int Conf Mach Learn – ident: ref45 doi: 10.1016/j.media.2012.11.001 – year: 2014 ident: ref26 article-title: Mediastinal lymph node detection on thoracic CT scans using spatial prior from multi-atlas label fusion publication-title: SPIE Med Imag – ident: ref5 doi: 10.1007/s00330-013-2774-5 – ident: ref10 doi: 10.1109/TMI.2009.2028576 – year: 2013 ident: ref17 article-title: Mitosis detection in breast cancer histology images with deep neural networks publication-title: MICCAI – ident: ref24 doi: 10.1145/2063576.2064004 – ident: ref2 doi: 10.1016/j.beem.2008.01.011 – ident: ref32 doi: 10.1109/42.974920 – ident: ref16 doi: 10.1038/505146a – ident: ref8 doi: 10.1109/CVPR.2010.5540008 – ident: ref41 doi: 10.1007/11866763_57 – ident: ref47 doi: 10.1145/1961189.1961199 – ident: ref33 doi: 10.1109/ISBI.2011.5872376 – ident: ref21 doi: 10.1007/978-3-319-14148-0_1 – ident: ref36 doi: 10.1117/12.594547 |
| SSID | ssj0014509 |
| Score | 2.6443954 |
| Snippet | Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities... Automated computer-aided detection (CADe) in medical imaging has been an important tool in clinical practice and research. State-of-the-art methods often show... |
| SourceID | pubmedcentral proquest pubmed crossref ieee |
| SourceType | Open Access Repository Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 1170 |
| SubjectTerms | Adolescent Adult Aged artificial neural networks Child Colonic polyps Colonic Polyps - diagnostic imaging Computed tomography Computer aided diagnosis Databases, Factual deep learning Feature extraction Female Humans Lymph nodes Lymph Nodes - diagnostic imaging Machine Learning Male medical diagnostic imaging Middle Aged multi-layer neural network Neural networks Neural Networks, Computer object detection Patients Radiographic Image Interpretation, Computer-Assisted - methods Spinal Neoplasms - diagnostic imaging Three-dimensional displays Tomography, X-Ray Computed Training Young Adult |
| Title | Improving Computer-Aided Detection Using Convolutional Neural Networks and Random View Aggregation |
| URI | https://ieeexplore.ieee.org/document/7279156 https://www.ncbi.nlm.nih.gov/pubmed/26441412 https://www.proquest.com/docview/1787856033 https://www.proquest.com/docview/1788540709 https://pubmed.ncbi.nlm.nih.gov/PMC7340334 |
| Volume | 35 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB61PVRwANryCJTKSL0gkd048SM5roCqRVoOqK16i2J70q6ALKJZkPj1jJ2HulVV9ZJEsuPYmnE84_n8DcAhR64kz-uYS2t9mFHGlXYyNppcFlp_ExvomuZf1fGZ-HIhLzbgw3gWBhED-Awn_jHE8t3SrvxW2ZTW2oL8jU3YJMetO6s1RgyE7OAcqWeMTVQ6hCSTYno6P_EYLjlJRe6TM3kCYG8GCJ6urUYhvcpdluZtwOSNFejoKcyHvnfAk--TVWsm9t8tWseHDu4ZPOlNUTbrdGcHNrDZhcc3CAp3YXveh973wIzbD2zIBBHPFg4d-4RtwHM1LOAPqLj50ys0Ne_ZP8ItwM2vWdU49o0uy5_sfIF_2eySPP7LoB_P4ezo8-nH47hP0BBbcjvb2BnnqNNCK5U4hShqmwrkxkrUqs7qqkp4JUTNM1cIVJ4KxxpT0bMjOSWYvYCtZtngK2C10GgzaTTW5KxjboqsMGQMiVQ7xyVGMB0EVdqevdwn0fhRBi8mKUqScumlXPZSjuD9-Mavjrnjnrp7XiBjvV4WEewPulD2U_u65PSLy8lOzLII3o3FNCl9pKVqcLkKdXLPbJgUEbzsVGdse1C9CPSaUo0VPOH3ekmzuArE3zoT9Fnx-u7evoFHNCbVoTH3Yav9vcK3ZDG15iBMlf9I4BFp |
| linkProvider | IEEE |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB6VIvE48Gh5BAoYiQsS2Y0TO06OK6DaQtMD2qLeotielFVLFtEsSPx6xs5D3apCXJJIdhxbM87MZL58A_CGI08lz-qQS2NcmlGGlbIy1IpCFrK_kfF0TcVROj8Wn07kyRa8G_-FQUQPPsOJu_S5fLsya_epbEq2Nqd44wbcJLsvefe31pgzELIDdMSOMzZK4yEpGeXTRXHgUFxyEovMlWdyFMDOERA83rBHvsDKdb7mVcjkJRu0fx-KYfYd9ORssm71xPy5Quz4v8t7APd6Z5TNOu15CFvY7MDdSxSFO3Cr6JPvu6DHDxBsqAURzpYWLfuArUd0NcwjEKi5-dWrNA3v-D_8yQPOL1jVWPaFDqvv7OsSf7PZKcX8p15DHsHx_sfF-3nYl2gIDQWebWi1tTRpodI0simiqE0skGsjUaV1UldVxCshap7YXGDqyHCM1hVdW5JThMlj2G5WDT4FVguFJpFaYU3hOmY6T3JN7pCIlbVcYgDTQVCl6fnLXRmN89LHMVFekpRLJ-Wyl3IAb8c7fnTcHf_ou-sEMvbrZRHA3qALZb-5L0pOL7mMPMUkCeD12Ezb0uVaqgZXa98nc9yGUR7Ak051xrEH1QtAbSjV2MFRfm-2NMtvnvpbJYIeK55dP9tXcHu-KA7Lw4Ojz8_hDq0v7bCZe7Dd_lzjC_KfWv3Sb5u_psEUsg |
| 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=Improving+Computer-Aided+Detection+Using+Convolutional+Neural+Networks+and+Random+View+Aggregation&rft.jtitle=IEEE+transactions+on+medical+imaging&rft.au=Roth%2C+Holger+R&rft.au=Lu%2C+Le&rft.au=Liu%2C+Jiamin&rft.au=Yao%2C+Jianhua&rft.date=2016-05-01&rft.eissn=1558-254X&rft.volume=35&rft.issue=5&rft.spage=1170&rft_id=info:doi/10.1109%2FTMI.2015.2482920&rft_id=info%3Apmid%2F26441412&rft.externalDocID=26441412 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0278-0062&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0278-0062&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0278-0062&client=summon |