Back

Coronary artery calcification in the Moscow population (based on the Moscow Experiment on the use of computer vision in medical imaging)

Vasilev, Y. A.; Goncharova, I. V.; Vladzymyrskyy, A. V.; Shulkin, I. M.; Arzamasov, K. M.

2023-04-19 radiology and imaging
10.1101/2023.04.12.23288344 medRxiv
Show abstract

IntroductionThe challenges related to the prevention, diagnosis, and treatment of cardiovascular diseases remain acute. One of the measures to prevent these conditions from occurring is early detection of risk factors, one of which is coronary calcium. The latest achievements in computer vision made it possible to conduct opportunistic screening for coronary calcium. ObjectiveTo study the prevalence of coronary artery calcification (CAC) as a risk factor for cardiovascular diseases in the Moscow population based on automated analysis of imaging findings. Materials and methodsA retrospective descriptive epidemiology study was carried out. Chest CT studies of 165,234 patients were analyzed. AI services carried out the automated analysis to detect CAC and to calculate the CAC score (CACS). ResultsCoronary calcium was detected in 61.4% of the participants. The proportion of diagnosed men was 68.9% (of all men), women - 55.7% (of all women) (p<0.001). The CAC score ranged between 1 to 60,306; the mean value was 558.2. The mean CACS increase for the entire population was 170.75; the mean growth was 168.13% the mean growth rate was 68.13%. 47.6% of men and 36.5% of women had clinically significant CACS (p<0.001). Most participants with clinically significant CAC belonged to Elderly and Senile age groups (42.0% each). ConclusionsThe prevalence of coronary calcium in the Moscow population was 8.03 per 1000 people. CAC (including clinically significant) was significantly more common in men. The mean CACS was significantly higher compared to the female population across most age groups. There is a continuous increase in the mean CACS with age.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.