Multi-stage deep learning architecture for carotid artery segmentation and stenosis evaluation: comparative study with DSA.
Authors
Affiliations (6)
Affiliations (6)
- Academy for Engineering and Technology, Fudan University, 220 Handan Road, Yangpu District, Shanghai, 200433, PR China.
- Department of Radiology, Huashan Hospital, Fudan University, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China.
- Department of Radiology, Huashan Hospital, Fudan University, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China; Department of Radiology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, No. 600, Yishan Road, Shanghai, 200233, PR China.
- Department of Radiology, the First Affiliated Hospital of Ningbo University, No. 59 Liuting Street, Haishu District, Ningbo, 315000, PR China.
- Department of Radiology, Huashan Hospital, Fudan University, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China; Shanghai Engineering Research Center of Intelligent Imaging for Critical Brain Diseases, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China. Electronic address: [email protected].
- Academy for Engineering and Technology, Fudan University, 220 Handan Road, Yangpu District, Shanghai, 200433, PR China; Department of Radiology, Huashan Hospital, Fudan University, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China; Shanghai Engineering Research Center of Intelligent Imaging for Critical Brain Diseases, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China; Institute of Functional and Molecular Medical Imaging, Fudan University, No. 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, PR China. Electronic address: [email protected].
Abstract
High-resolution magnetic resonance imaging (HR-MRI) provides a non-invasive, radiation-free approach for evaluating stenosis caused by carotid atherosclerosis. However, manual recognition is time-consuming and inter-observer variability. We propose a novel architecture for automated segmentation and stenosis evaluation of extracranial carotid arteries by HR-MRI in comparison with digital subtraction angiography (DSA). The 641 stenotic arteries from 422 patients retrospectively collected from three tertiary hospitals were divided into a training-validation set (372 patients, 545 lesions) and an independent test set (50 patients, 96 lesions). An external validation set (89 patients, 168 lesions) was collected from the fourth tertiary hospital. The architecture demonstrated high consistency with manual segmentation and DSA diagnostic criteria, with mean Dice similarity coefficients of 0.97 ± 0.01, 0.96 ± 0.01, and stenosis evaluation accuracies of 0.88, 0.86 on the independent test and external validation set, respectively. Thus, the proposed architecture achieved accuracy comparable to manual segmentation by physicians and demonstrated high consistency with DSA diagnostic criteria. By shortening diagnostic time and minimizing inter-observer variability, the proposed architecture is promising to offer a reliable, efficient, and intelligent tool for diagnosing head and neck atherosclerotic disease and assessing stroke risk.