Automatic Segmentation of Calcified Plaque in Carotid Arteries

Jiehyun Kim, Kevin Wang, Yu Sakai, Youxiang Zhu, Andrew C. Hu, Huy Q. Phi, Nathan Arnett, Grace J. Wang, Brett L. Cucchiara, Jae W. Song, Daniel Haehn

Abstract: Stroke remains a leading cause of death globally, with calcified plaque in the carotid artery being a significant risk factor. Evaluating the impact of calcified carotid plaque, particularly in patients with embolic stroke of undetermined source (ESUS), is critical yet challenging. Manual segmentation, essential for assessing stroke risk, is time-consuming, and conventional methods like 2D and 3D UNet often struggle with the small size of calcified plaques. Therefore, this paper in-troduces a two-step segmentation process. First, segments the carotid artery to narrow the search space and focus on the region of interest around the artery. Then, it segments the calcified plaque within that targeted region. This approach achieves an intersection over union (IoU) of 0.9412 for the 2D model and 0.8095 for the 3D model, outperforming the baseline methods that directly segment plaques. All developments are open source and publicly accessible on our GitHub: https://github.com/mpsych/CACTAS-AI.

External IDs:dblp:conf/isbi/KimWSZHPAWCSH25