Automated Segmentation of the Human Abdominal Vascular System Using a Hybrid Approach Combining Expert System and Supervised Deep Learning.
artificial intelligence
automated segmentation
deep learning
machine learning
vascular system
Journal
Journal of clinical medicine
ISSN: 2077-0383
Titre abrégé: J Clin Med
Pays: Switzerland
ID NLM: 101606588
Informations de publication
Date de publication:
29 Jul 2021
29 Jul 2021
Historique:
received:
03
07
2021
revised:
25
07
2021
accepted:
27
07
2021
entrez:
7
8
2021
pubmed:
8
8
2021
medline:
8
8
2021
Statut:
epublish
Résumé
Computed tomography angiography (CTA) is one of the most commonly used imaging technique for the management of vascular diseases. Here, we aimed to develop a hybrid method combining a feature-based expert system with a supervised deep learning (DL) algorithm to enable a fully automatic segmentation of the abdominal vascular tree. We proposed an algorithm based on the hybridization of a data-driven convolutional neural network and a knowledge-based model dedicated to vascular system segmentation. By using two distinct datasets of CTA from patients to evaluate independence to training dataset, the accuracy of the hybrid method for lumen and thrombus segmentation was evaluated compared to the feature-based expert system alone and to the ground truth provided by a human expert. The hybrid approach demonstrated a better accuracy for lumen segmentation compared to the expert system alone (volume similarity: 0.8128 vs. 0.7912, By enabling a robust and fully automatic segmentation, the method could be used to develop real-time decision support to help in the management of vascular diseases.
Sections du résumé
BACKGROUND
BACKGROUND
Computed tomography angiography (CTA) is one of the most commonly used imaging technique for the management of vascular diseases. Here, we aimed to develop a hybrid method combining a feature-based expert system with a supervised deep learning (DL) algorithm to enable a fully automatic segmentation of the abdominal vascular tree.
METHODS
METHODS
We proposed an algorithm based on the hybridization of a data-driven convolutional neural network and a knowledge-based model dedicated to vascular system segmentation. By using two distinct datasets of CTA from patients to evaluate independence to training dataset, the accuracy of the hybrid method for lumen and thrombus segmentation was evaluated compared to the feature-based expert system alone and to the ground truth provided by a human expert.
RESULTS
RESULTS
The hybrid approach demonstrated a better accuracy for lumen segmentation compared to the expert system alone (volume similarity: 0.8128 vs. 0.7912,
CONCLUSIONS
CONCLUSIONS
By enabling a robust and fully automatic segmentation, the method could be used to develop real-time decision support to help in the management of vascular diseases.
Identifiants
pubmed: 34362129
pii: jcm10153347
doi: 10.3390/jcm10153347
pmc: PMC8347188
pii:
doi:
Types de publication
Journal Article
Langues
eng
Références
Mach Learn Med Eng Cardiovasc Health Intravasc Imaging Comput Assist Stenting (2019). 2019;11794:167-174
pubmed: 34113925
Nat Rev Cardiol. 2021 Aug;18(8):600-609
pubmed: 33712806
Inf Process Med Imaging. 2003 Jul;18:136-47
pubmed: 15344453
Eur J Radiol. 2019 Dec;121:108713
pubmed: 31683252
Front Neurosci. 2020 Dec 08;14:592352
pubmed: 33363452
J Am Coll Cardiol. 2019 Mar 26;73(11):1317-1335
pubmed: 30898208
Comput Methods Programs Biomed. 2018 May;158:71-91
pubmed: 29544791
Conf Proc IEEE Eng Med Biol Soc. 2005;2005:4002-5
pubmed: 17281109
IEEE Trans Med Imaging. 2005 Apr;24(4):477-85
pubmed: 15822806
J Vasc Surg. 2020 Jul;72(1):321-333.e1
pubmed: 32093909
Sci Rep. 2019 Sep 24;9(1):13750
pubmed: 31551507
Front Cardiovasc Med. 2020 Mar 05;7:25
pubmed: 32195270
Sci Rep. 2020 Sep 29;10(1):16057
pubmed: 32994452
J Vasc Surg. 2009 Oct;50(4 Suppl):S2-49
pubmed: 19786250
Stud Health Technol Inform. 2000;77:1195-200
pubmed: 11187511
Med Image Anal. 2018 May;46:202-214
pubmed: 29609054
J Digit Imaging. 2018 Aug;31(4):490-504
pubmed: 29352385
Med Biol Eng Comput. 2019 Feb;57(2):543-564
pubmed: 30255236
Eur J Vasc Endovasc Surg. 2017 Apr;53(4):460-510
pubmed: 28359440
Med Phys. 2006 May;33(5):1440-53
pubmed: 16752579
PLoS One. 2019 Mar 13;14(3):e0213539
pubmed: 30865678
BMC Med Imaging. 2015 Aug 12;15:29
pubmed: 26263899
Eur J Vasc Endovasc Surg. 2019 Jan;57(1):8-93
pubmed: 30528142
Med Image Anal. 2014 Jan;18(1):1-8
pubmed: 24077409
Comput Methods Programs Biomed. 2012 Aug;107(2):202-17
pubmed: 21880391
Ann Vasc Surg. 2020 May;65:254-260
pubmed: 31857229
Cardiovasc Eng Technol. 2019 Sep;10(3):490-499
pubmed: 31218516
IEEE Trans Pattern Anal Mach Intell. 2014 Jan;36(1):2-17
pubmed: 24231862
Int J Comput Assist Radiol Surg. 2017 Sep;12(9):1501-1510
pubmed: 28455765
Radiology. 2018 Jul;288(1):177-185
pubmed: 29584598
Med Image Anal. 2004 Jun;8(2):127-38
pubmed: 15063862
Comput Biol Med. 2010 Mar;40(3):271-8
pubmed: 20074719
J Digit Imaging. 2019 Aug;32(4):582-596
pubmed: 31144149
Eur Heart J. 2019 Jul 1;40(25):2058-2073
pubmed: 30815669
J Vasc Surg. 2021 Jul;74(1):348
pubmed: 34172199
Med Image Anal. 2009 Dec;13(6):819-45
pubmed: 19818675