AI for the diagnosis and treatment of coronary artery disease.
The ISCAD learning model has developed thanks to artificial intelligence, designed to improve the diagnosis and management of the disease.
Researchers at the Icahn School of Medicine at Mount Sinai in New York have built an in silico marker (the term, which has recently appeared in scientific literature, is used to indicate chemical-biological phenomena reproduced in a computer mathematical simulation, instead of in test tube ) of coronary artery disease (coronary artery disease or coronary artery disease means any anatomical or functional alteration of the coronary arteries, i.e. the blood vessels that carry blood to the heart), to better measure the clinically important characteristics of the disease. The findings, published in The Lancet, could lead to a more targeted diagnosis and improved management of coronary artery disease, the most common type of heart disease and a leading cause of death worldwide.
Coronary heart disease spans a spectrum of risk factors and disease processes, and each individual determines their position on the spectrum. However, most of these studies divide this spectrum of disease into rigid classes of cases (the patient has the disease) or controls (the patient does not have the disease). According to the researchers, this can lead to missed diagnoses, inappropriate management and worse clinical outcomes. “It is critical to have the ability to reveal distinct gradations of disease risk, disease consequences such as atherosclerosis, and survival, for example, that might otherwise be ignored with a conventional binary picture. Our model delineates coronary artery disease patient populations on a spectrum and therefore with multiple factors and combinations; this could provide more insight into disease progression and how those affected will respond to treatment.’
In the study, the researchers trained the machine learning model, called the in silico score for coronary artery disease or ISCAD , to accurately measure coronary artery disease using more than 80,000 electronic health records from two large health system-based biobanks, the BioMe Biobank of the Mount Sinai Health System and the UK Biobank. The model, which the researchers termed a “digital marker,” incorporated hundreds of different medical characteristics from the electronic health record, including vital signs, lab test results, medications, symptoms, and diagnoses, and compared it to both a clinical existing for coronary heart disease, which uses only a small number of predetermined characteristics, either with a genetic score . The 95,935 participants included people of African, Hispanic/Latino, Asian and European ethnicity, as well as a large proportion of women. Most clinical and machine learning studies of CAD have focused on white European ethnicity. The researchers found that the probabilities derived from the model accurately tracked the degree of narrowing of the coronary arteries (coronary artery stenosis), mortality and complications such as heart attack.
“Machine learning models like this could also be useful to the healthcare sector in general, designing clinical trials based on appropriate patient stratification. They could also lead to more efficient, data-driven individualized treatment strategies,” says lead author Iain S. Forrest. “Despite these advances, it is important to remember that clinician and procedural diagnosis and management of coronary heart disease are not replaced by artificial intelligence, but rather potentially supported by ISCAD as another powerful tool in the clinician’s toolbox” . The investigators plan to conduct a large-scale prospective study to further validate the clinical utility and actionability of ISCAD, including in other populations.
