An attention residual u-net with differential preprocessing and geometric postprocessing: Learning how to segment vasculature including intracranial aneurysms

• Published in: Medical image analysis 2023-02, Vol.84, p.102697-102697, Article 102697
• Main Authors: Mu, Nan, Lyu, Zonghan, Rezaeitaleshmahalleh, Mostafa, Tang, Jinshan, Jiang, Jingfeng
• Format: Article
• Language: English
• Subjects: 3D segmentation; Angiography; Attention; Conditional random field; Depth-aware attention gate; Humans; Image Processing, Computer-Assisted - methods; Imaging, Three-Dimensional - methods; Intracranial aneurysm; Intracranial Aneurysm - diagnostic imaging; Multiscale supervision
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2022.102697

•This paper proposes automatically segmenting vasculature, including intracranial aneurysms, from 3DRA images with the need for “patient-specific” computational hemodynamics.•An attention residual U-Net with preprocessing and postprocessing is constructed.•We design multi-scale supervision to improve the segmentation of small vessels.•We adopt a fully connected 3D conditional random field to remove unwanted vessel-to-vessel or vessel-to-aneurysm connections.•This is the first study to systematically investigate the potential of deep-learning image segmentation for “patient-specific” computational hemodynamics. Intracranial aneurysms (IA) are lethal, with high morbidity and mortality rates. Reliable, rapid, and accurate segmentation of IAs and their adjacent vasculature from medical imaging data is important to improve the clinical management of patients with IAs. However, due to the blurred boundaries and complex structure of IAs and overlapping with brain tissue or other cerebral arteries, image segmentation of IAs remains challenging. This study aimed to develop an attention residual U-Net (ARU-Net) architecture with differential preprocessing and geometric postprocessing for automatic segmentation of IAs and their adjacent arteries in conjunction with 3D rotational angiography (3DRA) images. The proposed ARU-Net followed the classic U-Net framework with the following key enhancements. First, we preprocessed the 3DRA images based on boundary enhancement to capture more contour information and enhance the presence of small vessels. Second, we introduced the long skip connections of the attention gate at each layer of the fully convolutional decoder-encoder structure to emphasize the field of view (FOV) for IAs. Third, residual-based short skip connections were also embedded in each layer to implement in-depth supervision to help the network converge. Fourth, we devised a multiscale supervision strategy for independent prediction at different levels of the decoding path, integrating multiscale semantic information to facilitate the segmentation of small vessels. Fifth, the 3D conditional random field (3DCRF) and 3D connected component optimization (3DCCO) were exploited as postprocessing to optimize the segmentation results. Comprehensive experimental assessments validated the effectiveness of our ARU-Net. The proposed ARU-Net model achieved comparable or superior performance to the state-of-the-art methods through quantitative and qualitative evaluations.