Cargando…

Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes

Segmenting vessels in brain images is a critical step for many medical interventions and diagnoses of illnesses. Recent advances in artificial intelligence provide better models, achieving a human-like level of expertise in many tasks. In this paper, we present a new approach to segment Time-of-Flig...

Descripción completa

Detalles Bibliográficos
Autores principales: Quintana-Quintana, Oliver J., De León-Cuevas, Alejandro, González-Gutiérrez, Arturo, Gorrostieta-Hurtado, Efrén, Tovar-Arriaga, Saúl
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9229670/
https://www.ncbi.nlm.nih.gov/pubmed/35744437
http://dx.doi.org/10.3390/mi13060823
_version_ 1784734808067276800
author Quintana-Quintana, Oliver J.
De León-Cuevas, Alejandro
González-Gutiérrez, Arturo
Gorrostieta-Hurtado, Efrén
Tovar-Arriaga, Saúl
author_facet Quintana-Quintana, Oliver J.
De León-Cuevas, Alejandro
González-Gutiérrez, Arturo
Gorrostieta-Hurtado, Efrén
Tovar-Arriaga, Saúl
author_sort Quintana-Quintana, Oliver J.
collection PubMed
description Segmenting vessels in brain images is a critical step for many medical interventions and diagnoses of illnesses. Recent advances in artificial intelligence provide better models, achieving a human-like level of expertise in many tasks. In this paper, we present a new approach to segment Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) images, relying on fewer training samples than state-of-the-art methods. We propose a conditional generative adversarial network with an adapted generator based on a concatenated U-Net with a residual U-Net architecture (UUr-cGAN) to carry out blood vessel segmentation in TOF-MRA images, relying on data augmentation to diminish the drawback of having few volumes at disposal for training the model, while preventing overfitting by using regularization techniques. The proposed model achieves 89.52% precision and 87.23% in Dice score on average from the cross-validated experiment for brain blood vessel segmentation tasks, which is similar to other state-of-the-art methods while using considerably fewer training samples. UUr-cGAN extracts important features from small datasets while preventing overfitting compared to other CNN-based methods and still achieve a relatively good performance in image segmentation tasks such as brain blood vessels from TOF-MRA.
format Online
Article
Text
id pubmed-9229670
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-92296702022-06-25 Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes Quintana-Quintana, Oliver J. De León-Cuevas, Alejandro González-Gutiérrez, Arturo Gorrostieta-Hurtado, Efrén Tovar-Arriaga, Saúl Micromachines (Basel) Article Segmenting vessels in brain images is a critical step for many medical interventions and diagnoses of illnesses. Recent advances in artificial intelligence provide better models, achieving a human-like level of expertise in many tasks. In this paper, we present a new approach to segment Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) images, relying on fewer training samples than state-of-the-art methods. We propose a conditional generative adversarial network with an adapted generator based on a concatenated U-Net with a residual U-Net architecture (UUr-cGAN) to carry out blood vessel segmentation in TOF-MRA images, relying on data augmentation to diminish the drawback of having few volumes at disposal for training the model, while preventing overfitting by using regularization techniques. The proposed model achieves 89.52% precision and 87.23% in Dice score on average from the cross-validated experiment for brain blood vessel segmentation tasks, which is similar to other state-of-the-art methods while using considerably fewer training samples. UUr-cGAN extracts important features from small datasets while preventing overfitting compared to other CNN-based methods and still achieve a relatively good performance in image segmentation tasks such as brain blood vessels from TOF-MRA. MDPI 2022-05-25 /pmc/articles/PMC9229670/ /pubmed/35744437 http://dx.doi.org/10.3390/mi13060823 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Quintana-Quintana, Oliver J.
De León-Cuevas, Alejandro
González-Gutiérrez, Arturo
Gorrostieta-Hurtado, Efrén
Tovar-Arriaga, Saúl
Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes
title Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes
title_full Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes
title_fullStr Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes
title_full_unstemmed Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes
title_short Dual U-Net-Based Conditional Generative Adversarial Network for Blood Vessel Segmentation with Reduced Cerebral MR Training Volumes
title_sort dual u-net-based conditional generative adversarial network for blood vessel segmentation with reduced cerebral mr training volumes
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9229670/
https://www.ncbi.nlm.nih.gov/pubmed/35744437
http://dx.doi.org/10.3390/mi13060823
work_keys_str_mv AT quintanaquintanaoliverj dualunetbasedconditionalgenerativeadversarialnetworkforbloodvesselsegmentationwithreducedcerebralmrtrainingvolumes
AT deleoncuevasalejandro dualunetbasedconditionalgenerativeadversarialnetworkforbloodvesselsegmentationwithreducedcerebralmrtrainingvolumes
AT gonzalezgutierrezarturo dualunetbasedconditionalgenerativeadversarialnetworkforbloodvesselsegmentationwithreducedcerebralmrtrainingvolumes
AT gorrostietahurtadoefren dualunetbasedconditionalgenerativeadversarialnetworkforbloodvesselsegmentationwithreducedcerebralmrtrainingvolumes
AT tovararriagasaul dualunetbasedconditionalgenerativeadversarialnetworkforbloodvesselsegmentationwithreducedcerebralmrtrainingvolumes