基于模糊C均值法的CTA图像冠状动脉狭窄量化

Stenosis quantification of coronary artery CTA images based on fuzzy C-means algorithm

针对冠状动脉造影图像中由软斑块造成的冠状动脉狭窄,提出了一种基于模糊C均值法的狭窄精确量化方法.首先,在沿冠状动脉中心线的三维数据图上,对专家给定起讫位置的冠状动脉血管片段进行切割和阈值化.其次,利用模糊C均值法分割出血管腔区域,计算血管每个横截面的血管腔面积检测值,通过起讫位置的检测面积拟合得到参考值.然后,通过比较血管横截面面积的检测值和参考值来确定分段中最狭窄的位置及其狭窄比率.最后,利用该方法对13个病人的CTA数据进行了测试,并将测试结果与专家给出的狭窄比率进行比较.结果表明,所提方法能够对CTA中给定起讫位置的冠状动脉血管片段的狭窄区域进行精确量化,狭窄比率的平均绝对差和均方根差分别为2.21%和3.11%.

As for coronary artery stenosis caused by soft plaques in coronary artery angiography images,an accurate stenosis quantification method based on the fuzzy C-means algorithm is proposed.First,the segment of the coronary artery with starting and ending positions defined by experts is cropped and thresholded on the three-dimensional data along the centerline. Secondly,the fuzzy Cmeans algorithm is applied to separate the region of vessel lumen from other tissues. The detected value of the vessel lumen area in each slice is computed and the reference value is fitted by the detection area between the starting and ending positions. Then,the location and the ratio of the most stenosis are determined by comparing the detected value with the reference value of the vessel area. Finally,the proposed method is tested on clinical computed tomography angiography（ CTA） datasets of thirteen patients and the detection results are compared with the stenosis ratios defined by experts. The experimental results indicate that the proposed method can accurately quantify the stenosis of the coronary artery segment with given starting and ending positions in CTA. The absolute average difference and the root mean squared difference of the stenosis ratios are 2. 21% and 3. 11%,respectively.