Population-Based Study of Coronary Artery Calcification Using the Automated Analysis of Radiology Reports in Moscow

Abstract

Introduction: Problems of prevention, diagnosis and treatment of diseases of the circulatory system remain highly relevant. One of the effective preventive measures is early identification of risk factors, including coronary calcium. Recent achievements in the field of computer vision have made it possible to conduct opportunistic coronary calcium screening. Objective: To study the prevalence of coronary artery calcification as a risk factor for cardiovascular diseases in the population of Moscow based on the results of an automated analysis of radiology findings. Materials and methods: In July 2021 – December 2022, we conducted a retrospective descriptive epidemiological study, within which we analyzed chest CT images of 165,234 patients (71,635 males and 93,599 females) for coronary artery calcification and calcium scoring using AI services in an automated mode. Results: Coronary calcium was detected in 61.4 % of the examined. The proportion of men was 68.9 %, women – 55.7 % (р < 0.001). The calcium score ranged from 1 to 60,306 units (mean = 558.2). The average growth rate of the calcium score for the whole population was 170.75, the average growth rate was 168.13, and the average increase rate was 68.13 units during study period. In 47.6 % of men and 36.5 % of women with coronary calcium, the calcium score was clinically significant, i.e. ≥ 300 (p < 0.001). Most people with coronary calcium at a clinically significant level belonged to elderly and senile age groups (42.0 % each). Conclusions: The prevalence of coronary calcium in the population of Moscow was 8.03 per 1,000 people. In men, coronary calcium (including that at a clinically significant level) was statistically more frequent while the average calcium score in them was significantly higher than in women of most age groups. The mean calcium score demonstrated a constant increase with age.