科研之友
微信
新浪微博
Facebook
分享链接
摘要
It has been a challenging but significant research topic to extract the centerlines of the coronary arteries of the CTA volume in clinical applications. A new method is proposed to full-automatically extract and recognize the centerlines of the major branches in this paper. This method, which is called the model-mapped directional minimal path (MDMP), originates from an improved minimal path algorithm. In the MDMP method, the cost function is redefined, and the starting point of the coronary is automatically detected. Both the ROI (region of interest) and directional information of the coronary centerline are provided by a prior model. Firstly, the prior model is mapped onto the CTA volume by the image registration. After getting the ROI of the starting point of the coronary centerline based on the registration, the two ostia position are detected automatically by the learning-based algorithm in the CTA volume. The ostia are set as the starting points in the MDMP evolution. And, the directional factor of the coronary artery is imported into the cost function of the minimal path, guiding the coronary centerline tracking process. The robustness of the centerline extraction is improved by the minimal path based on the prior model. Finally, the three major branches are automatically extracted and recognized in the ROI of the coronary centerline. The proposed method is validated by extracting and recognizing the three major coronary branches, and the average overlap percentage was 81.7%.
