Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO
Summary: Automated tools to detect large vessel occlusion (LVO) in acute ischemic stroke patients using brain computed tomography angiography (CTA) have been shown to reduce the time for treatment, leading to better clinical outcomes. There is a lot of information in a single CTA and deep learning m...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-02-01
|
| Series: | iScience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589004224001020 |
| _version_ | 1797345907004407808 |
|---|---|
| author | Uma M. Lal-Trehan Estrada Arnau Oliver Sunil A. Sheth Xavier Lladó Luca Giancardo |
| author_facet | Uma M. Lal-Trehan Estrada Arnau Oliver Sunil A. Sheth Xavier Lladó Luca Giancardo |
| author_sort | Uma M. Lal-Trehan Estrada |
| collection | DOAJ |
| description | Summary: Automated tools to detect large vessel occlusion (LVO) in acute ischemic stroke patients using brain computed tomography angiography (CTA) have been shown to reduce the time for treatment, leading to better clinical outcomes. There is a lot of information in a single CTA and deep learning models do not have an obvious way of being conditioned on areas most relevant for LVO detection, i.e., the vasculature structure. In this work, we compare and contrast strategies to make convolutional neural networks focus on the vasculature without discarding context information of the brain parenchyma and propose an attention-inspired strategy to encourage this. We use brain CTAs from which we obtain 3D vasculature images. Then, we compare ways of combining the vasculature and the CTA images using a general-purpose network trained to detect LVO. The results show that the proposed strategies allow to improve LVO detection and could potentially help to learn other cerebrovascular-related tasks. |
| first_indexed | 2024-03-08T11:25:35Z |
| format | Article |
| id | doaj.art-3d83791a1a544e39b8d373163e845b09 |
| institution | Directory Open Access Journal |
| issn | 2589-0042 |
| language | English |
| last_indexed | 2024-03-08T11:25:35Z |
| publishDate | 2024-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | iScience |
| spelling | doaj.art-3d83791a1a544e39b8d373163e845b092024-01-26T05:34:23ZengElsevieriScience2589-00422024-02-01272108881Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVOUma M. Lal-Trehan Estrada0Arnau Oliver1Sunil A. Sheth2Xavier Lladó3Luca Giancardo4Research institute of Computer Vision and Robotics, University of Girona, Girona, SpainResearch institute of Computer Vision and Robotics, University of Girona, Girona, SpainMcGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX, USAResearch institute of Computer Vision and Robotics, University of Girona, Girona, SpainCenter for Precision Health, McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, Houston, TX, USA; Corresponding authorSummary: Automated tools to detect large vessel occlusion (LVO) in acute ischemic stroke patients using brain computed tomography angiography (CTA) have been shown to reduce the time for treatment, leading to better clinical outcomes. There is a lot of information in a single CTA and deep learning models do not have an obvious way of being conditioned on areas most relevant for LVO detection, i.e., the vasculature structure. In this work, we compare and contrast strategies to make convolutional neural networks focus on the vasculature without discarding context information of the brain parenchyma and propose an attention-inspired strategy to encourage this. We use brain CTAs from which we obtain 3D vasculature images. Then, we compare ways of combining the vasculature and the CTA images using a general-purpose network trained to detect LVO. The results show that the proposed strategies allow to improve LVO detection and could potentially help to learn other cerebrovascular-related tasks.http://www.sciencedirect.com/science/article/pii/S2589004224001020Medical imagingNeuroanatomyNeural networksMachine learning |
| spellingShingle | Uma M. Lal-Trehan Estrada Arnau Oliver Sunil A. Sheth Xavier Lladó Luca Giancardo Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO iScience Medical imaging Neuroanatomy Neural networks Machine learning |
| title | Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO |
| title_full | Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO |
| title_fullStr | Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO |
| title_full_unstemmed | Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO |
| title_short | Strategies to combine 3D vasculature and brain CTA with deep neural networks: Application to LVO |
| title_sort | strategies to combine 3d vasculature and brain cta with deep neural networks application to lvo |
| topic | Medical imaging Neuroanatomy Neural networks Machine learning |
| url | http://www.sciencedirect.com/science/article/pii/S2589004224001020 |
| work_keys_str_mv | AT umamlaltrehanestrada strategiestocombine3dvasculatureandbrainctawithdeepneuralnetworksapplicationtolvo AT arnauoliver strategiestocombine3dvasculatureandbrainctawithdeepneuralnetworksapplicationtolvo AT sunilasheth strategiestocombine3dvasculatureandbrainctawithdeepneuralnetworksapplicationtolvo AT xavierllado strategiestocombine3dvasculatureandbrainctawithdeepneuralnetworksapplicationtolvo AT lucagiancardo strategiestocombine3dvasculatureandbrainctawithdeepneuralnetworksapplicationtolvo |