No abstract

The Agatston score (AS) is a widely used method for evaluating the risk of coronary artery disease [1][2][3] . The AS is a summation of each calcification volume (CV) weighted by lesion scores, which are defi ned by the maximum CT number of each calcifi ed lesion 4) . To calculate the AS, an electrocardiogram (ECG) gating system is essential for data acquisition, and a workstation (WS) is also necessary. Therefore, a computed tomography (CT) scanner without an ECG-gating system cannot be used to determine the AS. In addition, preparation of ECG-gated CT is time-consuming, as electrodes have to be appropriately positioned.Calcifi cation of the coronary artery is one of the most common incidental fi ndings of daily CT examinations. The 2016 Society of Cardiovascular Computed Tomography and the Society of Thoracic Radiology guidelines state that the evaluation of coronary artery calcification on all chest CT examinations is mandatory 5) . In addition, the British Society of Cardiovascular Imaging/British Society of Cardiac Computed Tomography and British Society of Thoracic Imaging recommended reporting coronary artery calcifi cation at all non-gated chest CT examinations using a simple grading system (non, mild, moderate, and severe) 6) . However, the estimation of CV of the coronary artery using only visual criteria is diffi cult, as the inter-observer and intra-observer errors are not acceptable.Recently, several types of artifi cial intelligence (AI) have become commercially available. The CV of the coronary artery can be estimated by AI. Our institute has used a deep learning AI