Our paper ‘Deep Learning–Quantified Calcium Scores for Automatic Cardiovascular Mortality Prediction at Lung Screening Low-Dose CT’ by De Vos et al. has been published in Radiology: Cardiothoracic Imaging. We present a deep learning-based (DL) algorithm that accurately predicts the risk of death from cardiovascular disease using information from low-dose CT images performed for lung cancer screening.

Low-dose chest CT images are used to screen for lung cancer in high-risk people such as heavy smokers. These CT images also provide an opportunity to screen for cardiovascular disease by extracting information about calcification in the heart and aorta. The presence of calcium in these areas is linked with the buildup of plaque and is a strong predictor for cardiovascular disease mortality, heart attacks and strokes.

The method we developed uses only image information, it is fully automatic, and it is fast. The method obtains calcium scores in a complete chest CT in less than half a second. This means that the method should be easy to implement in routine patient work ups and screening. Most importantly, the method could help identify people in a population of heavy smokers who might be at increased risk of death from cardiovascular disease-related causes.

RSNA news covered the paper in an interview with Bob de Vos.