Machine learning to predict risk of MI and CV death
Machine learning has been used to predict the long term risk of myocardial infarction and cardiac death in a study published in Cardiovascular Research from the European Society of Cardiology. It was a prospective study using clinical parameters, coronary artery calcium scoring and automated epicardial adipose tissue quantification.
The study included 1912 asymptomatic persons from EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) trial with long-term follow-up after coronary artery calcium scoring. Epicardial adipose tissue is a metabolically active fat deposit which has been shown to relate to early atherosclerosis. Fully automated multitask convolutional neural network deep learning technique was used to quantify epicardial fat from the CT images of calcium scoring.
Extreme gradient boosting (XGBoost) machine learning was trained with clinical covariates, plasma lipid panel measurements, risk factors, coronary artery calcium scores, aortic calcium, and automated epicardial adipose tissue measures.
Machine learning technique could predict events better than atherosclerotic cardiovascular disease (ASCVD) risk score and coronary artery calcium score in this study.