An accurate and reproducible scheme for quantification of coronary artery calcification in CT scans

The coronary artery disease is a major cause of deaths in the western world. One indicator for coronary artery disease (CAD) is coronary artery calcification (CAC). An accurate and reproducible scheme is desired to monitor the progression of patient's coronary calcification in follow-up studies. Traditional approaches for CAC estimation lack to provide accurate and reproducible results. In This work, a new adaptive and stochastic 3D method has been proposed by employing a modified expectation-maximisation (MEM) algorithm. It is less sensitive to partial volume effects, motion effects, slice thickness and low dose. Accuracy of the proposed method was measured by a cardiac CT stationary phantom containing 6 calcium inserts of predetermined size and density that were scanned 90 times using 15 different protocols based on slice thickness and radiation. Reproducibility was measured in 35 patients who were each scanned twice with the patient being repositioned before the second scan. Compared with the Agatston based method, it is shown that the proposed algorithm gives better results in terms of accuracy and reproducibility.