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

作者：Sun Qiaoyu; Yang Guanyu^*; Shu Huazhong

来源：Journal of Southeast University(Natural Science Edition), 2016, 46(1): 30-34.

DOI：10.3969/j.issn.1001-0505.2016.01.006

• Computed tomography angiography
• Fuzzy C mean
• Soft plaques
• Stenosis quantification
• Vascular segmentations

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 C-means 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.

• 出版日期2016
• 单位东南大学