According to the data from the Center for Disease Control and Prevention, Cardiovascular disease (CVD) in people is the leading cause of death around the world. Every year, nearly 610,000 people die of heart attacks and strokes in the U.S. and the number stood at around 17.9 million globally. It isn’t impossible to predict CVD, but there is a strong risk factor in Coronary Artery Calcium (CAC) deposits that limit blood flow. Thus, for measuring CAC requires experts who can closely examine CT scans (Computerized Tomography) for worsening signs and symptoms.
Keeping this prediction in mind, a latest released paper on the “Direct Automatic Coronary Calcium Scoring in Cardiac and Chest CT” from Arxiv.org proposed an Artificial Intelligence-powered system that can assess and score CAC with human interference. Albeit, this isn’t a new system, automated CAC tests have been around for a while. Moreover, the researchers asserted that their AI-optimized system is up to hundreds of times faster than state-of-the-art methods. This AI system involves two Convolutional Neural Networks (CNN), a set of deep neural networks commonly applied to evaluating visual imagery, where the first take as input CT scans and aligns the fields of view while second performs direct regression, i.e. linear modeling of the relationship between variables of the calcium scores. The networks were trained on two datasets; one from the University Medical Center Utrecht in the Netherlands comprising 903 cardiac CT scans, of which 237 scans were utilized for training, and the 1,687 chest CT scans from the National Lung Screening Trial, in which 1,012 scans were used.
The newly paper issued months after researchers at Florida State University and the University of Florida, Gainesville featured an AI system that could foresee one-year mortality in ICU patients who’d experienced a heart attack, and following Corti, an AI system which finds heart attacks during emergency phone calls, started rolling out to London, Paris, Milan, and Munich.