Three-dimensional vessel segmentation in magnetic resonance angiography using mask modeling
10.3969/j.issn.1005-202X.2025.10.014
- VernacularTitle:基于掩码建模的磁共振血管造影的三维血管分割
- Author:
Dexuan LI
1
;
Chenglong WANG
;
Qi ZHANG
;
Xuefeng ZHANG
;
Guang YANG
Author Information
1. 华东师范大学医学磁共振与分子影像技术研究院/上海市磁共振重点实验室,上海 200062
- Publication Type:Journal Article
- Keywords:
deep learning;
vessel segmentation;
magnetic resonance angiography;
topological connectivity
- From:
Chinese Journal of Medical Physics
2025;42(10):1361-1368
- CountryChina
- Language:Chinese
-
Abstract:
Magnetic resonance angiography(MRA)is a non-invasive imaging technique used to observe blood vessels.Quantitative analysis of MRA images enables visualization of vascular pathways,condition,and blood flow dynamics,which is essential for diagnosing vascular diseases such as vascular lesions,stenosis,and occlusions.Vessel segmentation serves as the fundamental basis for quantitative vascular analysis.However,the complex morphology of vessels,difficulties in labeling,and scarcity of accurate 3D vascular annotations pose significant challenges for MRA-based vessel segmentation.A strategy of selectively occluding vessels during model training is proposed to enhance the algorithm's capacity to capture the topological structure of blood vessels,thereby improving the continuity of vessel segmentation results.Additionally,a Refine network is incorporated to refine the binary segmentation results of the segmentation network,thereby further improving segmentation accuracy.Model training and testing are carried out using 42 cases of 3D MRA data from the public MIDAS dataset.For the test set,the 3D U-Net baseline model with vessel occlusion strategy shows a β0 Error of 1.2742±0.2103 and a β1 Error of 0.3393±0.0818,respectively,which are 0.1136 and 0.0280 lower than the baseline.The model integrating vessel occlusion strategy and Refine network achieves an average Dice score of 0.7105±0.0125,which is 0.0028 higher than the baseline.These results demonstrate that the proposed method effectively improves both vascular connectivity and segmentation accuracy.
