TONG Hexin, YANG Panfeng, PAN Lei, PENG Yuhang, YANG Zhaoyi, LI Zhu
This paper aims to report an algorithm capable of automatically segmenting and evaluating calcified areas and to perform an automatic analysis of rib cartilage calcification based on existing datasets. In this paper, we proposed an automatic segmentation algorithm for calcified rib cartilage regions in computed tomographic(CT) images, a watershed algorithm based on automatic marking of skeletal lines, and an automatic extraction algorithm for images of the 6th, 7th, and 8th costal cartilages to automatically analyze the calcification of rib cartilage at different ages, and the performance of this method is verified through experiments. It is found that the calcification rate decreased with increasing rib cartilage order. There are significant differences in the degree of calcification between costal cartilages of different ordinal numbers. The automatic segmentation of calcified rib cartilage and the automatic extraction of the 6th, 7th and 8th costal cartilages based on the segmentation results proposed in this paper are generally effective, and any number of datasets can be analyzed with this automated approach. The algorithm currently produces some errors in the segmentation stage, mainly due to the small size of the dataset used for training the model.