改进型骨骼细化算法提取冠状动脉中心线

目的 运用全自动骨骼细化算法从CT图像中精确提取冠状动脉的中心线。方法 分割CT图像中的冠状动脉区域，经三维重建得到完整的冠状动脉三维数据；利用骨骼细化算法提取该冠状动脉的中心线，引入Dijkstra最短路径算法提升提取精度。结果 相比未移除分支的骨骼细化算法，重叠率提升2%，平均距离减少38.2%，平均运行时间0.48 s。结论 改进型骨骼细化算法可有效提取冠状动脉中心线。

Extracting coronary artery centerline based on improved skeleton thinning algorithm

Objective To propose an automatic skeleton thinning algorithmfor accurate extracting the center line of coronary artery from CT images. Methods The region of coronary artery on CT images were segmented, and complete 3D data of coronary artery were obtained after 3D reconstruction. Then the center line of coronary artery was extracted using skeleton thinning algorithm. Dijkstra algorithm was introduced according to the characteristics of coronary artery to improve the accuracy of extraction. Results Compared with the skeleton thinning algorithm which didn''t remove mini branches, the rate of overlap was increased by 2%, the average distance was reduced by 38.2%, and the average running time was 0.48 s. Conclusion Using improved skeleton thinning algorithm could effectively extract the center line of coronary arteries.