Cited 4 times in
Deep Reinforcement Learning with Explicit Spatio-Sequential Encoding Network for Coronary Ostia Identification in CT Images
DC Field | Value | Language |
---|---|---|
dc.date.accessioned | 2022-11-24T00:33:12Z | - |
dc.date.available | 2022-11-24T00:33:12Z | - |
dc.date.issued | 2021-09 | - |
dc.identifier.uri | https://ir.ymlib.yonsei.ac.kr/handle/22282913/190769 | - |
dc.description.abstract | Accurate identification of the coronary ostia from 3D coronary computed tomography angiography (CCTA) is a essential prerequisite step for automatically tracking and segmenting three main coronary arteries. In this paper, we propose a novel deep reinforcement learning (DRL) framework to localize the two coronary ostia from 3D CCTA. An optimal action policy is determined using a fully explicit spatial-sequential encoding policy network applying 2.5D Markovian states with three past histories. The proposed network is trained using a dueling DRL framework on the CAT08 dataset. The experiment results show that our method is more efficient and accurate than the other methods. blueFloating-point operations (FLOPs) are calculated to measure computational efficiency. The result shows that there are 2.5M FLOPs on the proposed method, which is about 10 times smaller value than 3D box-based methods. In terms of accuracy, the proposed method shows that 2.22 ± 1.12 mm and 1.94 ± 0.83 errors on the left and right coronary ostia, respectively. The proposed method can be applied to the tasks to identify other target objects by changing the target locations in the ground truth data. Further, the proposed method can be utilized as a pre-processing method for coronary artery tracking methods. | - |
dc.description.statementOfResponsibility | open | - |
dc.format | application/pdf | - |
dc.language | English | - |
dc.publisher | MDPI | - |
dc.relation.isPartOf | SENSORS | - |
dc.rights | CC BY-NC-ND 2.0 KR | - |
dc.subject.MESH | Computed Tomography Angiography* | - |
dc.subject.MESH | Coronary Vessels* / diagnostic imaging | - |
dc.subject.MESH | Heart | - |
dc.subject.MESH | Tomography, X-Ray Computed | - |
dc.title | Deep Reinforcement Learning with Explicit Spatio-Sequential Encoding Network for Coronary Ostia Identification in CT Images | - |
dc.type | Article | - |
dc.contributor.college | College of Medicine (의과대학) | - |
dc.contributor.department | Yonsei Biomedical Research Center (연세의생명연구원) | - |
dc.contributor.googleauthor | Yeonggul Jang | - |
dc.contributor.googleauthor | Byunghwan Jeon | - |
dc.identifier.doi | 10.3390/s21186187 | - |
dc.relation.journalcode | J03219 | - |
dc.identifier.eissn | 1424-8220 | - |
dc.identifier.pmid | 34577391 | - |
dc.subject.keyword | coronary computed tomography angiography | - |
dc.subject.keyword | coronary ostia | - |
dc.subject.keyword | localization | - |
dc.subject.keyword | reinforcement learning | - |
dc.citation.volume | 21 | - |
dc.citation.number | 18 | - |
dc.citation.startPage | 6187 | - |
dc.identifier.bibliographicCitation | SENSORS, Vol.21(18) : 6187, 2021-09 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.