An automated method for coronary calcification detection from ECG-triggered multi-slice CT data is presented. The method first segments the heart region. In the obtained volume candidate objects are extracted by thresholding. They include coronary calcification, calcium located elsewhere in the heart, for example, in the valves or the myocardium, and other high density structures mostly representing noise and bone. A set of 57 features is calculated for each candidate object. In the feature space objects are classified with a k-NN classifier and feature selection in three consecutive stages. The method is tested on 51 scans of the heart. They contain 320 calcification in the coronary arteries, 291 in the aorta and 62 calcifications in the heart. The system correctly detected 177 calcifications in the coronaries at the expense of 56 false positive objects. On average the method makes 3.8 errors per scan.
Automated coronary calcification detection and scoring
I. Išgum, B. van Ginneken, A. Rutten and M. Prokop
4th International Symposium on Image and Signal Processing and Analysis 2005:127-132.