Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images
Cerebrovascular segmentation is important in various clinical applications, such as surgical planning and computer-aided diagnosis. In order to achieve high segmentation performance, three challenging problems should be taken into consideration: (1) large variations in vascular anatomies and voxel i...
Guardado en:
Autores principales: | , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Frontiers Media S.A.
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/98a6599f25f1483da5e145fbba7a0f06 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:98a6599f25f1483da5e145fbba7a0f06 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:98a6599f25f1483da5e145fbba7a0f062021-11-16T07:39:22ZVoxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images1662-453X10.3389/fnins.2021.756536https://doaj.org/article/98a6599f25f1483da5e145fbba7a0f062021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fnins.2021.756536/fullhttps://doaj.org/toc/1662-453XCerebrovascular segmentation is important in various clinical applications, such as surgical planning and computer-aided diagnosis. In order to achieve high segmentation performance, three challenging problems should be taken into consideration: (1) large variations in vascular anatomies and voxel intensities; (2) severe class imbalance between foreground and background voxels; (3) image noise with different magnitudes. Limited accuracy was achieved without considering these challenges in deep learning-based methods for cerebrovascular segmentation. To overcome the limitations, we propose an end-to-end adversarial model called FiboNet-VANGAN. Specifically, our contributions can be summarized as follows: (1) to relieve the first problem mentioned above, a discriminator is proposed to regularize for voxel-wise distribution consistency between the segmentation results and the ground truth; (2) to mitigate the problem of class imbalance, we propose to use the addition of cross-entropy and Dice coefficient as the loss function of the generator. Focal loss is utilized as the loss function of the discriminator; (3) a new feature connection is proposed, based on which a generator called FiboNet is built. By incorporating Dice coefficient in the training of FiboNet, noise robustness can be improved by a large margin. We evaluate our method on a healthy magnetic resonance angiography (MRA) dataset to validate its effectiveness. A brain atrophy MRA dataset is also collected to test the performance of each method on abnormal cases. Results show that the three problems in cerebrovascular segmentation mentioned above can be alleviated and high segmentation accuracy can be achieved on both datasets using our method.Bin GuoFugen ZhouBo LiuXiangzhi BaiFrontiers Media S.A.articleindex terms-class imbalanceimage noiseCNNGANmedical image segmentationNeurosciences. Biological psychiatry. NeuropsychiatryRC321-571ENFrontiers in Neuroscience, Vol 15 (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
index terms-class imbalance image noise CNN GAN medical image segmentation Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571 |
spellingShingle |
index terms-class imbalance image noise CNN GAN medical image segmentation Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571 Bin Guo Fugen Zhou Bo Liu Xiangzhi Bai Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images |
description |
Cerebrovascular segmentation is important in various clinical applications, such as surgical planning and computer-aided diagnosis. In order to achieve high segmentation performance, three challenging problems should be taken into consideration: (1) large variations in vascular anatomies and voxel intensities; (2) severe class imbalance between foreground and background voxels; (3) image noise with different magnitudes. Limited accuracy was achieved without considering these challenges in deep learning-based methods for cerebrovascular segmentation. To overcome the limitations, we propose an end-to-end adversarial model called FiboNet-VANGAN. Specifically, our contributions can be summarized as follows: (1) to relieve the first problem mentioned above, a discriminator is proposed to regularize for voxel-wise distribution consistency between the segmentation results and the ground truth; (2) to mitigate the problem of class imbalance, we propose to use the addition of cross-entropy and Dice coefficient as the loss function of the generator. Focal loss is utilized as the loss function of the discriminator; (3) a new feature connection is proposed, based on which a generator called FiboNet is built. By incorporating Dice coefficient in the training of FiboNet, noise robustness can be improved by a large margin. We evaluate our method on a healthy magnetic resonance angiography (MRA) dataset to validate its effectiveness. A brain atrophy MRA dataset is also collected to test the performance of each method on abnormal cases. Results show that the three problems in cerebrovascular segmentation mentioned above can be alleviated and high segmentation accuracy can be achieved on both datasets using our method. |
format |
article |
author |
Bin Guo Fugen Zhou Bo Liu Xiangzhi Bai |
author_facet |
Bin Guo Fugen Zhou Bo Liu Xiangzhi Bai |
author_sort |
Bin Guo |
title |
Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images |
title_short |
Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images |
title_full |
Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images |
title_fullStr |
Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images |
title_full_unstemmed |
Voxel-Wise Adversarial FiboNet for 3D Cerebrovascular Segmentation on Magnetic Resonance Angiography Images |
title_sort |
voxel-wise adversarial fibonet for 3d cerebrovascular segmentation on magnetic resonance angiography images |
publisher |
Frontiers Media S.A. |
publishDate |
2021 |
url |
https://doaj.org/article/98a6599f25f1483da5e145fbba7a0f06 |
work_keys_str_mv |
AT binguo voxelwiseadversarialfibonetfor3dcerebrovascularsegmentationonmagneticresonanceangiographyimages AT fugenzhou voxelwiseadversarialfibonetfor3dcerebrovascularsegmentationonmagneticresonanceangiographyimages AT boliu voxelwiseadversarialfibonetfor3dcerebrovascularsegmentationonmagneticresonanceangiographyimages AT xiangzhibai voxelwiseadversarialfibonetfor3dcerebrovascularsegmentationonmagneticresonanceangiographyimages |
_version_ |
1718426620287516672 |