Cerebral AVM segmentation from 3D rotational angiography images by convolutional neural networks
3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be reconstructed in 3D. However, these reconstructions are limited to only 3D visualization without possible interactive exploration of geometric characteristics of cerebral...
Saved in:
Published in | Neuroscience informatics Vol. 3; no. 3; p. 100138 |
---|---|
Main Authors | , , , , , |
Format | Journal Article |
Language | English |
Published |
Elsevier Masson SAS
01.09.2023
Elsevier |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | 3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be reconstructed in 3D. However, these reconstructions are limited to only 3D visualization without possible interactive exploration of geometric characteristics of cerebral structures. Refined understanding of the AVM angioarchitecture prior to treatment is mandatory and vascular segmentation is an important preliminary step that allow physicians analyze the complex vascular networks and can help guide microcatheters navigation and embolization of AVM.
A deep learning method was developed for the segmentation of 3DRA images of AVM patients. The method uses a fully convolutional neural network with a U-Net-like architecture and a DenseNet backbone. A compound loss function, combining Cross Entropy and Focal Tversky, is employed for robust segmentation. Binary masks automatically generated from region-growing segmentation have been used to train and validate our model.
The developed network was able to achieve the segmentation of the vessels and the malformation and significantly outperformed the region-growing algorithm. Our experiments were performed on 9 AVM patients. The trained network achieved a Dice Similarity Coefficient (DSC) of 80.43%, surpassing other U-Net like architectures and the region-growing algorithm on the manually approved test set by physicians.
This work demonstrates the potential of a learning-based segmentation method for characterizing very complex and tiny vascular structures even when the training phase is performed with the results of an automatic or a semi-automatic method. The proposed method can contribute to the planning and guidance of endovascular procedures. |
---|---|
AbstractList | 3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be reconstructed in 3D. However, these reconstructions are limited to only 3D visualization without possible interactive exploration of geometric characteristics of cerebral structures. Refined understanding of the AVM angioarchitecture prior to treatment is mandatory and vascular segmentation is an important preliminary step that allow physicians analyze the complex vascular networks and can help guide microcatheters navigation and embolization of AVM.
A deep learning method was developed for the segmentation of 3DRA images of AVM patients. The method uses a fully convolutional neural network with a U-Net-like architecture and a DenseNet backbone. A compound loss function, combining Cross Entropy and Focal Tversky, is employed for robust segmentation. Binary masks automatically generated from region-growing segmentation have been used to train and validate our model.
The developed network was able to achieve the segmentation of the vessels and the malformation and significantly outperformed the region-growing algorithm. Our experiments were performed on 9 AVM patients. The trained network achieved a Dice Similarity Coefficient (DSC) of 80.43%, surpassing other U-Net like architectures and the region-growing algorithm on the manually approved test set by physicians.
This work demonstrates the potential of a learning-based segmentation method for characterizing very complex and tiny vascular structures even when the training phase is performed with the results of an automatic or a semi-automatic method. The proposed method can contribute to the planning and guidance of endovascular procedures. Background and objective: 3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be reconstructed in 3D. However, these reconstructions are limited to only 3D visualization without possible interactive exploration of geometric characteristics of cerebral structures. Refined understanding of the AVM angioarchitecture prior to treatment is mandatory and vascular segmentation is an important preliminary step that allow physicians analyze the complex vascular networks and can help guide microcatheters navigation and embolization of AVM. Methods: A deep learning method was developed for the segmentation of 3DRA images of AVM patients. The method uses a fully convolutional neural network with a U-Net-like architecture and a DenseNet backbone. A compound loss function, combining Cross Entropy and Focal Tversky, is employed for robust segmentation. Binary masks automatically generated from region-growing segmentation have been used to train and validate our model. Results: The developed network was able to achieve the segmentation of the vessels and the malformation and significantly outperformed the region-growing algorithm. Our experiments were performed on 9 AVM patients. The trained network achieved a Dice Similarity Coefficient (DSC) of 80.43%, surpassing other U-Net like architectures and the region-growing algorithm on the manually approved test set by physicians. Conclusions: This work demonstrates the potential of a learning-based segmentation method for characterizing very complex and tiny vascular structures even when the training phase is performed with the results of an automatic or a semi-automatic method. The proposed method can contribute to the planning and guidance of endovascular procedures. Background and objective: 3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be reconstructed in 3D. However, these reconstructions are limited to only 3D visualization without possible interactive exploration of geometric characteristics of cerebral structures. Refined understanding of the AVM angioarchitecture prior to treatment is mandatory and vascular segmentation is an important preliminary step that allow physicians analyze the complex vascular networks and can help guide microcatheters navigation and embolization of AVM.Methods: A deep learning method was developed for the segmentation of 3DRA images of AVM patients. The method uses a fully convolutional neural network with a U-Net-like architecture and a DenseNet backbone. A compound loss function, combining Cross Entropy and Focal Tversky, is employed for robust segmentation. Binary masks automatically generated from region-growing segmentation have been used to train and validate our model.Results: The developed network was able to achieve the segmentation of the vessels and the malformationand significantly outperformed the region-growing algorithm. Our experiments were performed on 9 AVMpatients. The trained network achieved a Dice Similarity Coefficient (DSC) of 80.43%, surpassing other U-Net like architectures and the region-growing algorithm on the manually approved test set by physicians.Conclusions: This work demonstrates the potential of a learning-based segmentation method for characterizing very complex and tiny vascular structures even when the training phase is performed with the results of an automatic or a semi-automatic method. The proposed method can contribute to the planning and guidance of endovascular procedures. |
ArticleNumber | 100138 |
Author | Lahlouh, Mounir Passat, Nicolas Blanc, Raphaël Piotin, Michel Szewczyk, Jérôme Chenoune, Yasmina |
Author_xml | – sequence: 1 givenname: Mounir orcidid: 0000-0002-6871-3459 surname: Lahlouh fullname: Lahlouh, Mounir email: mounir.lahlouh@esme.fr organization: Université de Reims Champagne Ardenne, CReSTIC EA 3804, Reims 51100, France – sequence: 2 givenname: Raphaël orcidid: 0000-0002-3975-3865 surname: Blanc fullname: Blanc, Raphaël email: rblanc@for.paris organization: Fondation Ophtalmologique de Rothschild, Interventional Neuroradiology Department, Paris, France – sequence: 3 givenname: Michel surname: Piotin fullname: Piotin, Michel email: mpiotin@for.paris organization: Fondation Ophtalmologique de Rothschild, Interventional Neuroradiology Department, Paris, France – sequence: 4 givenname: Jérôme surname: Szewczyk fullname: Szewczyk, Jérôme email: sz@isir.upmc.fr organization: Sorbonne Université, CNRS UMR 7222, Inserm, U1150, ISIR, F-75005, Paris, France – sequence: 5 givenname: Nicolas surname: Passat fullname: Passat, Nicolas email: nicolas.passat@univ-reims.fr organization: Université de Reims Champagne Ardenne, CReSTIC EA 3804, Reims 51100, France – sequence: 6 givenname: Yasmina orcidid: 0000-0002-8143-4796 surname: Chenoune fullname: Chenoune, Yasmina email: yasmina.chenoune@esme.fr organization: ESME Sudria Research Lab, Paris, France |
BackLink | https://hal.science/hal-04164909$$DView record in HAL |
BookMark | eNp9UTtPwzAYtBBIPH8BS1aGFr8dDwxVeUpFLMBqXOdzcEljZIei_nvSBiFYmO7T6e50-u4Q7baxBYROCR4TTOT5YtzCRwpjiinrGUxYuYMOqFJ0JGgpd3_d--gk5wXGmJaEMEIP0MsUEsyTbYrJ832RoV5C29kuxLbwKS4LdlmkOBC9xrZ1iHWy76_rIixtDbmYrwsX21VsPr41mzJb6D5jesvHaM_bJsPJNx6hp-urx-ntaPZwczedzEaOclaOQEgiQIP0TFovMQXtuFKEKu2tUIJx4YnSWpfeUuLk3GmlKq-cJ4JrgdkRuhtyq2gX5j319dLaRBvMloipNjZ1wTVguOTeltZxCRWvmJxX4EBQKMERSpXss86GrFfb_Im6nczMhsOcSK6xXpFeywatSzHnBP7HQLDZ7GMWZruP2exjhn1618Xggv4lqwDJZBegdVCFBK7rO4d__V-6pZtv |
Cites_doi | 10.2307/1932409 10.1016/j.compbiomed.2017.12.011 10.1016/j.neucom.2019.10.035 10.1016/j.compbiomed.2019.103489 10.1037/0033-295X.84.4.327 10.1136/neurintsurg-2020-016450 10.3389/fnins.2019.00097 10.1016/j.isprsjprs.2020.01.013 10.1007/s11547-022-01567-5 10.1016/j.compbiomed.2017.01.012 10.1136/neurintsurg-2014-011402 10.1016/j.ijrobp.2022.07.910 10.1002/mp.13560 10.1117/1.JRS.14.044512 10.1161/01.STR.32.6.1458 |
ContentType | Journal Article |
Copyright | 2023 The Authors Attribution - NoDerivatives |
Copyright_xml | – notice: 2023 The Authors – notice: Attribution - NoDerivatives |
DBID | 6I. AAFTH AAYXX CITATION 1XC VOOES DOA |
DOI | 10.1016/j.neuri.2023.100138 |
DatabaseName | ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access) DOAJ Directory of Open Access Journals |
DatabaseTitle | CrossRef |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: DOA name: Open Access: DOAJ - Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Medicine Computer Science |
EISSN | 2772-5286 |
ExternalDocumentID | oai_doaj_org_article_464fa8ac46ed4d36bdece52e8ec12276 oai_HAL_hal_04164909v1 10_1016_j_neuri_2023_100138 S2772528623000237 |
GroupedDBID | 6I. AAFTH AAFWJ AAXUO AFPKN ALMA_UNASSIGNED_HOLDINGS AMRAJ EBS FDB GROUPED_DOAJ M41 M~E OK1 ROL .1- .FO 0R~ 0SF AALRI AAYXX ADVLN AFJKZ AFRHN AITUG AJUYK CITATION Z5R 1XC VOOES |
ID | FETCH-LOGICAL-c2438-e5615e9e6f36af602e9c4771279fa575345f179998fa21c6bc977df7cf1549503 |
IEDL.DBID | DOA |
ISSN | 2772-5286 |
IngestDate | Tue Oct 22 15:14:11 EDT 2024 Tue Oct 15 15:08:21 EDT 2024 Thu Sep 26 19:19:59 EDT 2024 Fri Feb 23 02:36:03 EST 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 3 |
Keywords | Cerebrovascular segmentation Fully automatic segmentation 3D rotational angiography Convolutional neural networks Focal Tversky convolutional neural networks fully automatic segmentation cerebrovascular segmentation focal Tversky |
Language | English |
License | This is an open access article under the CC BY-NC-ND license. Attribution - NoDerivatives: http://creativecommons.org/licenses/by-nd |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c2438-e5615e9e6f36af602e9c4771279fa575345f179998fa21c6bc977df7cf1549503 |
ORCID | 0000-0002-3975-3865 0000-0002-8143-4796 0000-0002-6871-3459 0000-0002-1354-4328 0000-0003-1925-3086 0000-0002-0320-4581 |
OpenAccessLink | https://doaj.org/article/464fa8ac46ed4d36bdece52e8ec12276 |
ParticipantIDs | doaj_primary_oai_doaj_org_article_464fa8ac46ed4d36bdece52e8ec12276 hal_primary_oai_HAL_hal_04164909v1 crossref_primary_10_1016_j_neuri_2023_100138 elsevier_sciencedirect_doi_10_1016_j_neuri_2023_100138 |
PublicationCentury | 2000 |
PublicationDate | September 2023 |
PublicationDateYYYYMMDD | 2023-09-01 |
PublicationDate_xml | – month: 09 year: 2023 text: September 2023 |
PublicationDecade | 2020 |
PublicationTitle | Neuroscience informatics |
PublicationYear | 2023 |
Publisher | Elsevier Masson SAS Elsevier |
Publisher_xml | – name: Elsevier Masson SAS – name: Elsevier |
References | Chenoune, Tankyevych, Li, Piotin, Blanc, Petit (br0060) 2019; 115 Huang, Lin, Tong, Hu, Zhang, Iwamoto, Han, Chen, Wu (br0300) 2020 Bash, Villablanca, Jahan, Duckwiler, Tillis, Kidwell, Saver, Sayre (br0010) 2005; 26 Colombo, Fick, Esposito, Germans, Regli, van Doormaal (br0240) 2022; 127 He, Zhang, Ren, Sun (br0130) 2016 Dolz, Ayed, Desrosiers (br0160) 2018 Ogilvy, Stieg, Awad, Brown, Kondziolka, Rosenwasser, Young, Hademenos (br0040) 2001; 32 Tversky (br0210) 1977; 84 Ronneberger, Fischer, Brox (br0090) 2015 Feddal, Escalard, Delvoye, Fahed, Desilles, Zuber, Redjem, Savatovsky, Ciccio, Smajda, Ben Maacha, Mazighi, Piotin, Blanc (br0030) 2020; 41 Hsu, Ghaffari, Alaraj, Flannery, Zhou, Linninger (br0070) 2017; 82 Dice (br0180) 1945; 26 Zhou, Siddiquee, Tajbakhsh, Liang (br0290) 2018 Kingma, Ba Adam (br0280) 2017 Salehi, Erdogmus, Gholipour (br0220) 2017 Oktay, Schlemper, Folgoc, Lee, Heinrich, Misawa, Mori, McDonagh, Hammerla, Kainz (br0320) 2018 Babin, Pižurica, Velicki, Matić, Galić, Leventić, Zlokolica, Philips (br0080) 2018; 93 Livne, Rieger, Aydin, Taha, Akay, Kossen, Sobesky, Kelleher, Hildebrand, Frey (br0110) 2019; 13 Soomro, Afifi, Gao, Hellwich, Paul, Zheng (br0190) 2018 Wang, Lei, Tian, Jiang, Zhou, Liu, Dresser, Curran, Shu, Yang (br0250) 2019; 46 Diakogiannis, Waldner, Caccetta, Wu (br0310) 2020; 162 Huang, Liu, Van Der Maaten, Weinberger (br0140) 2017 Brügger, Baumgartner, Konukoglu (br0200) 2019 García, Fang, Liu, Narata, Orlando, Larrabide (br0270) 2022 Ghafoorian, Teuwen, Manniesing, de Leeuw, van Ginneken, Karssemeijer, Platel (br0340) 2018 Baharvahdat, Blanc, Fahed, Pooyan, Mowla, Escalard, Delvoye, Desilles, Redjem, Ciccio, Smajda, Hamdani, Mazighi, Piotin (br0050) 2021; 13 Jégou, Drozdzal, Vazquez, Romero, Bengio (br0150) 2017 Abraham, Khan (br0230) 2019 Tetteh, Efremov, Forkert, Schneider, Kirschke, Weber, Zimmer, Piraud, Menze (br0100) 2018 Delibasoglu, Cetin (br0330) 2020; 14 Blanc, Seiler, Robert, Baharvahdat, Lafarge, Savatovsky, Hodel, Ciccio, Chauvet, Pistocchi, Bartolini, Redjem, Piotin (br0020) 2015; 7 Sanchesa, Meyer, Vigon, Naegel (br0120) 2019 Meng, Sun, Guan, Wang, Zong, Liu (br0170) 2020; 373 You, W, Hong, Lin, Sun, Lu, Chen, Huang, Lee (br0260) 2022; 114 Sanchesa (10.1016/j.neuri.2023.100138_br0120) 2019 Chenoune (10.1016/j.neuri.2023.100138_br0060) 2019; 115 Oktay (10.1016/j.neuri.2023.100138_br0320) Ghafoorian (10.1016/j.neuri.2023.100138_br0340) 2018 Hsu (10.1016/j.neuri.2023.100138_br0070) 2017; 82 Livne (10.1016/j.neuri.2023.100138_br0110) 2019; 13 Delibasoglu (10.1016/j.neuri.2023.100138_br0330) 2020; 14 Brügger (10.1016/j.neuri.2023.100138_br0200) 2019 Huang (10.1016/j.neuri.2023.100138_br0140) 2017 Babin (10.1016/j.neuri.2023.100138_br0080) 2018; 93 Bash (10.1016/j.neuri.2023.100138_br0010) 2005; 26 Diakogiannis (10.1016/j.neuri.2023.100138_br0310) 2020; 162 Soomro (10.1016/j.neuri.2023.100138_br0190) 2018 Baharvahdat (10.1016/j.neuri.2023.100138_br0050) 2021; 13 Huang (10.1016/j.neuri.2023.100138_br0300) 2020 Feddal (10.1016/j.neuri.2023.100138_br0030) 2020; 41 He (10.1016/j.neuri.2023.100138_br0130) 2016 Zhou (10.1016/j.neuri.2023.100138_br0290) You (10.1016/j.neuri.2023.100138_br0260) 2022; 114 Blanc (10.1016/j.neuri.2023.100138_br0020) 2015; 7 García (10.1016/j.neuri.2023.100138_br0270) Colombo (10.1016/j.neuri.2023.100138_br0240) 2022; 127 Tetteh (10.1016/j.neuri.2023.100138_br0100) Dice (10.1016/j.neuri.2023.100138_br0180) 1945; 26 Tversky (10.1016/j.neuri.2023.100138_br0210) 1977; 84 Salehi (10.1016/j.neuri.2023.100138_br0220) 2017 Jégou (10.1016/j.neuri.2023.100138_br0150) 2017 Kingma (10.1016/j.neuri.2023.100138_br0280) Dolz (10.1016/j.neuri.2023.100138_br0160) 2018 Wang (10.1016/j.neuri.2023.100138_br0250) 2019; 46 Abraham (10.1016/j.neuri.2023.100138_br0230) 2019 Ogilvy (10.1016/j.neuri.2023.100138_br0040) 2001; 32 Meng (10.1016/j.neuri.2023.100138_br0170) 2020; 373 Ronneberger (10.1016/j.neuri.2023.100138_br0090) 2015 |
References_xml | – start-page: 271 year: 2018 end-page: 282 ident: br0160 article-title: Dense multi-path U-Net for ischemic stroke lesion segmentation in multiple image modalities publication-title: Int. MICCAI Brainlesion Workshop contributor: fullname: Desrosiers – start-page: 1 year: 2018 end-page: 8 ident: br0190 article-title: Strided U-Net model: retinal vessels segmentation using Dice loss publication-title: Digit. Image Comput. Tech. Appl. DICTA contributor: fullname: Zheng – volume: 13 start-page: 241 year: 2021 end-page: 246 ident: br0050 article-title: Endovascular treatment as the main approach for Spetzler-Martin grade III brain arteriovenous malformations publication-title: J. Neurointerventional Surg. contributor: fullname: Piotin – start-page: 429 year: 2019 end-page: 437 ident: br0200 article-title: A partially reversible U-Net for memory-efficient volumetric image segmentation publication-title: Int. Conf. Med. Image Comput. Comput.-Assist. Interv. contributor: fullname: Konukoglu – year: 2018 ident: br0290 article-title: UNet++: a nested U-Net architecture for medical image segmentation contributor: fullname: Liang – year: 2018 ident: br0100 article-title: Deepvesselnet: vessel segmentation, centerline prediction, and bifurcation detection in 3-d angiographic volumes contributor: fullname: Menze – year: 2017 ident: br0280 article-title: A method for stochastic optimization contributor: fullname: Ba Adam – start-page: 105742U year: 2018 ident: br0340 article-title: Student beats the teacher: deep neural networks for lateral ventricles segmentation in brain MR publication-title: Med. Imaging 2018 Image Process. contributor: fullname: Platel – volume: 46 start-page: 3133 year: 2019 end-page: 3141 ident: br0250 article-title: Learning-based automatic segmentation of arteriovenous malformations on contrast CT images in brain stereotactic radiosurgery publication-title: Med. Phys. contributor: fullname: Yang – volume: 26 start-page: 297 year: 1945 end-page: 302 ident: br0180 article-title: Measures of the amount of ecologic association between species publication-title: Ecology contributor: fullname: Dice – year: 2018 ident: br0320 article-title: Attention U-Net: learning where to look for the pancreas contributor: fullname: Kainz – volume: 127 start-page: 1333 year: 2022 end-page: 1341 ident: br0240 article-title: Segmentation techniques of brain arteriovenous malformations for 3D visualization: a systematic review publication-title: Radiol. Med. (Torino) contributor: fullname: van Doormaal – volume: 82 start-page: 29 year: 2017 end-page: 39 ident: br0070 article-title: Gap-free segmentation of vascular networks with automatic image processing pipeline publication-title: Comput. Biol. Med. contributor: fullname: Linninger – start-page: 768 year: 2019 end-page: 771 ident: br0120 article-title: Cerebrovascular network segmentation of MRA images with deep learning publication-title: IEEE 16th Int. Symp. Biomed. Imaging ISBI contributor: fullname: Naegel – start-page: 1055 year: 2020 end-page: 1059 ident: br0300 article-title: Unet 3+: a full-scale connected unet for medical image segmentation publication-title: IEEE Int. Conf. Acoust. Speech Signal Process. ICASSP contributor: fullname: Wu – year: 2022 ident: br0270 article-title: A deep learning model for brain vessel segmentation in 3DRA with arteriovenous malformations contributor: fullname: Larrabide – volume: 26 start-page: 1012 year: 2005 end-page: 1021 ident: br0010 article-title: Intracranial vascular stenosis and occlusive disease: evaluation with CT angiography, MR angiography, and digital subtraction angiography publication-title: Am. J. Neuroradiol. contributor: fullname: Sayre – volume: 41 start-page: 1663 year: 2020 end-page: 1669 ident: br0030 article-title: Fusion image guidance for supra-aortic vessel catheterization in neurointerventions: a feasibility study publication-title: Am. J. Neuroradiol. contributor: fullname: Blanc – start-page: 379 year: 2017 end-page: 387 ident: br0220 article-title: Tversky loss function for image segmentation using 3D fully convolutional deep networks publication-title: Int. Workshop Mach. Learn. Med. Imaging contributor: fullname: Gholipour – start-page: 11 year: 2017 end-page: 19 ident: br0150 article-title: The one hundred layers tiramisu: fully convolutional densenets for semantic segmentation publication-title: IEEE Conf. Comput. Vis. Pattern Recognit contributor: fullname: Bengio – start-page: 770 year: 2016 end-page: 778 ident: br0130 article-title: Deep residual learning for image recognition publication-title: IEEE Conf. Comput. Vis. Pattern Recognit. contributor: fullname: Sun – start-page: 234 year: 2015 end-page: 241 ident: br0090 article-title: U-net: convolutional networks for biomedical image segmentation publication-title: Int. Conf. Med. Image Comput. Comput.-Assist. Interv. contributor: fullname: Brox – volume: 373 start-page: 123 year: 2020 end-page: 134 ident: br0170 article-title: Multiscale dense convolutional neural network for DSA cerebrovascular segmentation publication-title: Neurocomputing contributor: fullname: Liu – volume: 162 start-page: 94 year: 2020 end-page: 114 ident: br0310 article-title: ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data publication-title: ISPRS J. Photogramm. Remote Sens. contributor: fullname: Wu – volume: 115 year: 2019 ident: br0060 article-title: Three-dimensional segmentation and symbolic representation of cerebral vessels on 3DRA images of arteriovenous malformations publication-title: Comput. Biol. Med. contributor: fullname: Petit – volume: 84 start-page: 327 year: 1977 ident: br0210 article-title: Features of similarity publication-title: Psychol. Rev. contributor: fullname: Tversky – volume: 93 start-page: 93 year: 2018 end-page: 105 ident: br0080 article-title: Skeletonization method for vessel delineation of arteriovenous malformation publication-title: Comput. Biol. Med. contributor: fullname: Philips – volume: 14 year: 2020 ident: br0330 article-title: Improved U-Nets with inception blocks for building detection publication-title: J. Appl. Remote Sens. contributor: fullname: Cetin – volume: 32 start-page: 1458 year: 2001 end-page: 1471 ident: br0040 article-title: Recommendations for the management of intracranial arteriovenous malformations: a statement for healthcare professionals from a special writing group of the Stroke Council, American Stroke Association publication-title: Stroke contributor: fullname: Hademenos – volume: 7 start-page: 841 year: 2015 end-page: 847 ident: br0020 article-title: Multimodal angiographic assessment of cerebral arteriovenous malformations: a pilot study publication-title: J. NeuroInterventional Surg. contributor: fullname: Piotin – volume: 114 start-page: e108 year: 2022 ident: br0260 article-title: Detection and segmentation of arteriovenous malformation lesions using a two-stage deep learning strategy publication-title: Int. J. Radiat. Oncol. Biol. Phys. contributor: fullname: Lee – start-page: 4700 year: 2017 end-page: 4708 ident: br0140 article-title: Densely connected convolutional networks publication-title: IEEE Conf. Comput. Vis. Pattern Recognit. contributor: fullname: Weinberger – volume: 13 start-page: 97 year: 2019 ident: br0110 article-title: A U-Net deep learning framework for high performance vessel segmentation in patients with cerebrovascular disease publication-title: Front. Neurosci. contributor: fullname: Frey – start-page: 683 year: 2019 end-page: 687 ident: br0230 article-title: A novel focal Tversky loss function with improved attention U-Net for lesion segmentation publication-title: IEEE 16th Int. Symp. Biomed. Imaging ISBI contributor: fullname: Khan – start-page: 4700 year: 2017 ident: 10.1016/j.neuri.2023.100138_br0140 article-title: Densely connected convolutional networks contributor: fullname: Huang – start-page: 379 year: 2017 ident: 10.1016/j.neuri.2023.100138_br0220 article-title: Tversky loss function for image segmentation using 3D fully convolutional deep networks contributor: fullname: Salehi – volume: 26 start-page: 1012 year: 2005 ident: 10.1016/j.neuri.2023.100138_br0010 article-title: Intracranial vascular stenosis and occlusive disease: evaluation with CT angiography, MR angiography, and digital subtraction angiography publication-title: Am. J. Neuroradiol. contributor: fullname: Bash – start-page: 105742U year: 2018 ident: 10.1016/j.neuri.2023.100138_br0340 article-title: Student beats the teacher: deep neural networks for lateral ventricles segmentation in brain MR contributor: fullname: Ghafoorian – volume: 41 start-page: 1663 year: 2020 ident: 10.1016/j.neuri.2023.100138_br0030 article-title: Fusion image guidance for supra-aortic vessel catheterization in neurointerventions: a feasibility study publication-title: Am. J. Neuroradiol. contributor: fullname: Feddal – volume: 26 start-page: 297 year: 1945 ident: 10.1016/j.neuri.2023.100138_br0180 article-title: Measures of the amount of ecologic association between species publication-title: Ecology doi: 10.2307/1932409 contributor: fullname: Dice – volume: 93 start-page: 93 year: 2018 ident: 10.1016/j.neuri.2023.100138_br0080 article-title: Skeletonization method for vessel delineation of arteriovenous malformation publication-title: Comput. Biol. Med. doi: 10.1016/j.compbiomed.2017.12.011 contributor: fullname: Babin – ident: 10.1016/j.neuri.2023.100138_br0100 contributor: fullname: Tetteh – ident: 10.1016/j.neuri.2023.100138_br0270 contributor: fullname: García – start-page: 770 year: 2016 ident: 10.1016/j.neuri.2023.100138_br0130 article-title: Deep residual learning for image recognition contributor: fullname: He – ident: 10.1016/j.neuri.2023.100138_br0320 contributor: fullname: Oktay – start-page: 11 year: 2017 ident: 10.1016/j.neuri.2023.100138_br0150 article-title: The one hundred layers tiramisu: fully convolutional densenets for semantic segmentation contributor: fullname: Jégou – start-page: 234 year: 2015 ident: 10.1016/j.neuri.2023.100138_br0090 article-title: U-net: convolutional networks for biomedical image segmentation contributor: fullname: Ronneberger – start-page: 271 year: 2018 ident: 10.1016/j.neuri.2023.100138_br0160 article-title: Dense multi-path U-Net for ischemic stroke lesion segmentation in multiple image modalities contributor: fullname: Dolz – volume: 373 start-page: 123 year: 2020 ident: 10.1016/j.neuri.2023.100138_br0170 article-title: Multiscale dense convolutional neural network for DSA cerebrovascular segmentation publication-title: Neurocomputing doi: 10.1016/j.neucom.2019.10.035 contributor: fullname: Meng – volume: 115 year: 2019 ident: 10.1016/j.neuri.2023.100138_br0060 article-title: Three-dimensional segmentation and symbolic representation of cerebral vessels on 3DRA images of arteriovenous malformations publication-title: Comput. Biol. Med. doi: 10.1016/j.compbiomed.2019.103489 contributor: fullname: Chenoune – start-page: 429 year: 2019 ident: 10.1016/j.neuri.2023.100138_br0200 article-title: A partially reversible U-Net for memory-efficient volumetric image segmentation contributor: fullname: Brügger – volume: 84 start-page: 327 year: 1977 ident: 10.1016/j.neuri.2023.100138_br0210 article-title: Features of similarity publication-title: Psychol. Rev. doi: 10.1037/0033-295X.84.4.327 contributor: fullname: Tversky – ident: 10.1016/j.neuri.2023.100138_br0290 contributor: fullname: Zhou – volume: 13 start-page: 241 year: 2021 ident: 10.1016/j.neuri.2023.100138_br0050 article-title: Endovascular treatment as the main approach for Spetzler-Martin grade III brain arteriovenous malformations publication-title: J. Neurointerventional Surg. doi: 10.1136/neurintsurg-2020-016450 contributor: fullname: Baharvahdat – volume: 13 start-page: 97 year: 2019 ident: 10.1016/j.neuri.2023.100138_br0110 article-title: A U-Net deep learning framework for high performance vessel segmentation in patients with cerebrovascular disease publication-title: Front. Neurosci. doi: 10.3389/fnins.2019.00097 contributor: fullname: Livne – start-page: 683 year: 2019 ident: 10.1016/j.neuri.2023.100138_br0230 article-title: A novel focal Tversky loss function with improved attention U-Net for lesion segmentation contributor: fullname: Abraham – volume: 162 start-page: 94 year: 2020 ident: 10.1016/j.neuri.2023.100138_br0310 article-title: ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data publication-title: ISPRS J. Photogramm. Remote Sens. doi: 10.1016/j.isprsjprs.2020.01.013 contributor: fullname: Diakogiannis – volume: 127 start-page: 1333 year: 2022 ident: 10.1016/j.neuri.2023.100138_br0240 article-title: Segmentation techniques of brain arteriovenous malformations for 3D visualization: a systematic review publication-title: Radiol. Med. (Torino) doi: 10.1007/s11547-022-01567-5 contributor: fullname: Colombo – ident: 10.1016/j.neuri.2023.100138_br0280 contributor: fullname: Kingma – volume: 82 start-page: 29 year: 2017 ident: 10.1016/j.neuri.2023.100138_br0070 article-title: Gap-free segmentation of vascular networks with automatic image processing pipeline publication-title: Comput. Biol. Med. doi: 10.1016/j.compbiomed.2017.01.012 contributor: fullname: Hsu – volume: 7 start-page: 841 year: 2015 ident: 10.1016/j.neuri.2023.100138_br0020 article-title: Multimodal angiographic assessment of cerebral arteriovenous malformations: a pilot study publication-title: J. NeuroInterventional Surg. doi: 10.1136/neurintsurg-2014-011402 contributor: fullname: Blanc – volume: 114 start-page: e108 year: 2022 ident: 10.1016/j.neuri.2023.100138_br0260 article-title: Detection and segmentation of arteriovenous malformation lesions using a two-stage deep learning strategy publication-title: Int. J. Radiat. Oncol. Biol. Phys. doi: 10.1016/j.ijrobp.2022.07.910 contributor: fullname: You – volume: 46 start-page: 3133 year: 2019 ident: 10.1016/j.neuri.2023.100138_br0250 article-title: Learning-based automatic segmentation of arteriovenous malformations on contrast CT images in brain stereotactic radiosurgery publication-title: Med. Phys. doi: 10.1002/mp.13560 contributor: fullname: Wang – volume: 14 year: 2020 ident: 10.1016/j.neuri.2023.100138_br0330 article-title: Improved U-Nets with inception blocks for building detection publication-title: J. Appl. Remote Sens. doi: 10.1117/1.JRS.14.044512 contributor: fullname: Delibasoglu – start-page: 1 year: 2018 ident: 10.1016/j.neuri.2023.100138_br0190 article-title: Strided U-Net model: retinal vessels segmentation using Dice loss contributor: fullname: Soomro – start-page: 768 year: 2019 ident: 10.1016/j.neuri.2023.100138_br0120 article-title: Cerebrovascular network segmentation of MRA images with deep learning contributor: fullname: Sanchesa – volume: 32 start-page: 1458 year: 2001 ident: 10.1016/j.neuri.2023.100138_br0040 article-title: Recommendations for the management of intracranial arteriovenous malformations: a statement for healthcare professionals from a special writing group of the Stroke Council, American Stroke Association publication-title: Stroke doi: 10.1161/01.STR.32.6.1458 contributor: fullname: Ogilvy – start-page: 1055 year: 2020 ident: 10.1016/j.neuri.2023.100138_br0300 article-title: Unet 3+: a full-scale connected unet for medical image segmentation contributor: fullname: Huang |
SSID | ssj0002811312 |
Score | 2.284968 |
Snippet | 3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be reconstructed in 3D. However,... Background and objective: 3D rotational angiography (3DRA) provides high quality images of the cerebral arteriovenous malformation (AVM) nidus that can be... |
SourceID | doaj hal crossref elsevier |
SourceType | Open Website Open Access Repository Aggregation Database Publisher |
StartPage | 100138 |
SubjectTerms | 3D rotational angiography Artificial Intelligence Cerebrovascular segmentation Computer Science Convolutional neural networks Focal Tversky Fully automatic segmentation Image Processing Medical Imaging |
Title | Cerebral AVM segmentation from 3D rotational angiography images by convolutional neural networks |
URI | https://dx.doi.org/10.1016/j.neuri.2023.100138 https://hal.science/hal-04164909 https://doaj.org/article/464fa8ac46ed4d36bdece52e8ec12276 |
Volume | 3 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3PT8MgFCbqwXgx_ozzV4jxaLWlFMpx6pbFOE9qdkNKQbdkndmmiRf_dh-UmnnRi5c2IQTIe4T3Pfj4QOiUcsV1zFUEsd5GlGgaKUJEZJXJiMlzbQpPkL1jvQd6M8gGC099OU5YLQ9cG-6CMmpVrjRlpqRlyorSaN-M0QkhvBbbTrKFZGrkt4ySJPVHnQTgI2RbOWskhzy5y4lFDs_d0-FehMjdTlkIS169_0d0Wn5p9ll93OluoPUAGHG7HugmWjLVFlrthyPxbfR0Zabu8BeqPPbxzDyPw22iCrurIzi9xtPJPGz5YVU9D4NINR6OYS2Z4eIDO-p5mIJQx43a_zxBfLaDHrqd-6teFJ5NiDShsHwZgESZEYbZlCnLYmKEppwnhAurAJ2lNLNOB07kVpFEs0IDBiwt19bJtWVxuotWqkll9hDWGUvB7rGGrAfyuFQkZaas5QAKIdEScQudNVaTr7U6hmxoYyPpjSydkWVt5Ba6dJb9ruqkrX0BOFwGh8u_HN5CrPGLDCihjv7Q1PD33k_Aiz8677VvpSuLAZZSEYv3ZP8_hniA1ly_NRXtEK3Mp2_mCLDLvDj20xS-_c_OF8hN63Q |
link.rule.ids | 230,315,786,790,870,891,2115,27955,27956 |
linkProvider | Directory of Open Access Journals |
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=Cerebral+AVM+segmentation+from+3D+rotational+angiography+images+by+convolutional+neural+networks&rft.jtitle=Neuroscience+informatics&rft.au=Lahlouh%2C+Mounir&rft.au=Blanc%2C+Rapha%C3%ABl&rft.au=Piotin%2C+Michel&rft.au=Szewczyk%2C+J%C3%A9r%C3%B4me&rft.date=2023-09-01&rft.pub=Elsevier+Masson+SAS&rft.issn=2772-5286&rft.eissn=2772-5286&rft.volume=3&rft.issue=3&rft_id=info:doi/10.1016%2Fj.neuri.2023.100138&rft.externalDocID=S2772528623000237 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2772-5286&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2772-5286&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2772-5286&client=summon |