New AI software a low-cost, efficient option for coronary artery calcium scoring

In this study, the diagnostic performance of AI-driven CACS in a clinical setting was evaluated. The authors enrolled 901 patients who had undergone both chest CT and ECG-gated non-contrast-enhanced cardiac CT with the same equipment. AI-CACS software was based on a deep learning algorithm and trained on multi-vendor, multi-scanner and multi-hospital data from a chest CT database. 

The results showed a 95% agreement between both methods, with a concordance rate of 80.6%. AI-developed risk categories were similar across three different types of machines. Overall, the AI-CACS and manual…