An algorithm for coronary calcification diagnosis based on helical CT images

The authors describe a computer assisted diagnosis algorithm for coronary calcifications based on helical X-ray CT images. The images are obtained from the mass screening process for lung cancer diagnosis. The authors´ diagnostic algorithm consists of two process: Firstly, they choose the CT slices including the heart region based on the heart shape and the diaphragm, and they extract the heart region from the CT images based on the shape of the lung area, and location of the vertebra. Secondly, the candidate regions for coronary calcification are decided after a smoothing and thresholding process applied to the extracted heart region. Finally, to increase the effectiveness of the diagnosis processing, the authors cancel the artifacts included in the candidate regions using some rules. In addition, the authors present here the results of their algorithm applied to the chest helical CT images of 467 patients involved in lung cancer screening