Segmentation of Abdominal Aortic Aneurysm (AAA) Based on Topology Prior Model.
MIUA(2017)
摘要
In this paper, we propose a statistical based method using a topology prior model, integrating both intensity and shape information, to segment abdominal aortic aneurysm (AAA) from computed tomography angiography (CTA) scans. The method was tested on a total of 48 slices taken from 6 different patients and has shown competitive performance compared with the best reported results in the literature. Our method has achieved a mean Dice coefficient of 0.9303±0.0499, and mean Hausdorff distance of 3.5703±3.1941 mm. This method overcomes the major problem faced by currently existing solutions of similar Hounsfield values of neighboring tissues to that of the AAA thrombus. This is a promising medical tool which can be used to analyze the AAA in order to generate an accurate rupture risk indicator.
更多查看译文
关键词
Abdominal aortic aneurysm, Segmentation, Lumen, Thrombus, Topology, Probability
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要