Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study
Summary: Non-contrast CT (NCCT) is widely used in clinical practice and holds potential for large-scale atherosclerosis screening, yet its application in detecting and grading aortic atherosclerosis remains limited. To address this, we propose Aortic-AAE, an automated segmentation system based on a...
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | iScience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589004225013616 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849389380674781184 |
|---|---|
| author | Mingliang Yang Jinhao Lyu Yongqin Xiong Aoxue Mei Jianxing Hu Yue Zhang Xiaoyu Wang Xiangbing Bian Jiayu Huang Runze Li Xinbo Xing Sulian Su Junhang Gao Xin Lou |
| author_facet | Mingliang Yang Jinhao Lyu Yongqin Xiong Aoxue Mei Jianxing Hu Yue Zhang Xiaoyu Wang Xiangbing Bian Jiayu Huang Runze Li Xinbo Xing Sulian Su Junhang Gao Xin Lou |
| author_sort | Mingliang Yang |
| collection | DOAJ |
| description | Summary: Non-contrast CT (NCCT) is widely used in clinical practice and holds potential for large-scale atherosclerosis screening, yet its application in detecting and grading aortic atherosclerosis remains limited. To address this, we propose Aortic-AAE, an automated segmentation system based on a cascaded attention mechanism within the nnU-Net framework. The cascaded attention module enhances feature learning across complex anatomical structures, outperforming existing attention modules. Integrated preprocessing and post-processing ensure anatomical consistency and robustness across multi-center data. Trained on 435 labeled NCCT scans from three centers and validated on 388 independent cases, Aortic-AAE achieved 81.12% accuracy in aortic stenosis classification and 92.37% in Agatston scoring of calcified plaques, surpassing five state-of-the-art models. This study demonstrates the feasibility of using deep learning for accurate detection and grading of aortic atherosclerosis from NCCT, supporting improved diagnostic decisions and enhanced clinical workflows. |
| format | Article |
| id | doaj-art-5db042b67f254eef82c5a22b49f6ba91 |
| institution | Kabale University |
| issn | 2589-0042 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | iScience |
| spelling | doaj-art-5db042b67f254eef82c5a22b49f6ba912025-08-20T03:41:58ZengElsevieriScience2589-00422025-08-0128811310010.1016/j.isci.2025.113100Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center studyMingliang Yang0Jinhao Lyu1Yongqin Xiong2Aoxue Mei3Jianxing Hu4Yue Zhang5Xiaoyu Wang6Xiangbing Bian7Jiayu Huang8Runze Li9Xinbo Xing10Sulian Su11Junhang Gao12Xin Lou13School of Medical Technology, Beijing Institute of Technology, No.5 Zhongguancun South Street, Haidian District, Beijing 100081, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Xiangyang NO.1 People’s Hospital, Hubei University of Medicine, No. 15 Jiefang Road, Fancheng District, Xiangyang 441000, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, ChinaDepartment of Radiology, Xiamen Humanity Hospital Fujian Medical University, No. 3777 Xianyue Road, Huli District, Xiamen City, Fujian Province 361000, ChinaDepartment of Radiology, Xiamen Humanity Hospital Fujian Medical University, No. 3777 Xianyue Road, Huli District, Xiamen City, Fujian Province 361000, ChinaSchool of Medical Technology, Beijing Institute of Technology, No.5 Zhongguancun South Street, Haidian District, Beijing 100081, China; Department of Radiology, Chinese PLA General Hospital, No. 28 Fuxing Road, Haidian District, Beijing, Beijing 100853, China; Corresponding authorSummary: Non-contrast CT (NCCT) is widely used in clinical practice and holds potential for large-scale atherosclerosis screening, yet its application in detecting and grading aortic atherosclerosis remains limited. To address this, we propose Aortic-AAE, an automated segmentation system based on a cascaded attention mechanism within the nnU-Net framework. The cascaded attention module enhances feature learning across complex anatomical structures, outperforming existing attention modules. Integrated preprocessing and post-processing ensure anatomical consistency and robustness across multi-center data. Trained on 435 labeled NCCT scans from three centers and validated on 388 independent cases, Aortic-AAE achieved 81.12% accuracy in aortic stenosis classification and 92.37% in Agatston scoring of calcified plaques, surpassing five state-of-the-art models. This study demonstrates the feasibility of using deep learning for accurate detection and grading of aortic atherosclerosis from NCCT, supporting improved diagnostic decisions and enhanced clinical workflows.http://www.sciencedirect.com/science/article/pii/S2589004225013616Cardiovascular medicineArtificial intelligence |
| spellingShingle | Mingliang Yang Jinhao Lyu Yongqin Xiong Aoxue Mei Jianxing Hu Yue Zhang Xiaoyu Wang Xiangbing Bian Jiayu Huang Runze Li Xinbo Xing Sulian Su Junhang Gao Xin Lou Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study iScience Cardiovascular medicine Artificial intelligence |
| title | Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study |
| title_full | Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study |
| title_fullStr | Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study |
| title_full_unstemmed | Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study |
| title_short | Aortic atherosclerosis evaluation using deep learning based on non-contrast CT: A retrospective multi-center study |
| title_sort | aortic atherosclerosis evaluation using deep learning based on non contrast ct a retrospective multi center study |
| topic | Cardiovascular medicine Artificial intelligence |
| url | http://www.sciencedirect.com/science/article/pii/S2589004225013616 |
| work_keys_str_mv | AT mingliangyang aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT jinhaolyu aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT yongqinxiong aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT aoxuemei aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT jianxinghu aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT yuezhang aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT xiaoyuwang aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT xiangbingbian aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT jiayuhuang aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT runzeli aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT xinboxing aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT suliansu aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT junhanggao aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy AT xinlou aorticatherosclerosisevaluationusingdeeplearningbasedonnoncontrastctaretrospectivemulticenterstudy |