Automatic segmentation of abdominal aortic aneurysms from CT angiography using a context-aware cascaded U-Net. (May 2023)

• Record Type: Journal Article
• Authors: Mu, Nan; Lyu, Zonghan; Rezaeitaleshmahalleh, Mostafa; Zhang, Xiaoming; Rasmussen, Todd; McBane, Robert; Jiang, Jingfeng
• Is Part Of: Computers in biology and medicine. Volume 158(2023)
• Publication Date: 2023-05
• Subjects: Abdominal aortic aneurysm -- Context-aware -- Geometrical analysis -- Image segmentation -- Neural network -- Deep-learning; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2023.106569
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: We delineate abdominal aortic aneurysms, including lumen and intraluminal thrombosis (ILT), from contrast-enhanced computed tomography angiography (CTA) data in 70 patients with complete automation. A novel context-aware cascaded U-Net configuration enables automated image segmentation. Notably, auto-context structure, in conjunction with dilated convolutions, anisotropic context module, hierarchical supervision, and a multi-class loss function, are proposed to improve the delineation of ILT in an unbalanced, low-contrast multi-class labeling problem. A quantitative analysis shows that the automated image segmentation produces comparable results with trained human users (e.g., DICE scores of 0.945 and 0.804 for lumen and ILT, respectively). Resultant morphological metrics (e.g., volume, surface area, etc.) are highly correlated to those parameters generated by trained human users. In conclusion, the proposed automated multi-class image segmentation tool has the potential to be further developed as a translational software tool that can be used to improve the clinical management of AAAs. Graphical abstract: Image 1 Highlights: This paper aims to automatically segment abdominal aortic aneurysm from CTA images. Two-stage context-aware cascaded U-Net is designed to capture surrounding context. An auto-context algorithm involving deep supervision at different resolutions. An anisotropic context module is used to capture short and long rang dependency. A rebalancing multi-class loss is explored to handle severely imbalanced classes.