Each year, around 20 million people around the world die from cardiovascular disease, according to the American Heart Association. However, the rise of artificial intelligence can drastically reduce that number, potentially saving thousands to millions of lives. Ironically, it seems that a man-made construction without a heart can predict the behavior of that very organ.
Doctors around the globe use set guidelines made by the American College of Cardiology/American Heart Association (ACC/AHA) in predicting patients’ risks of any related diseases. These guidelines weigh in factors including cholesterol level, age, and blood pressure, but has often been viewed as insufficient regarding the complexity and unpredictability of the human body. That’s when artificial intelligence comes into play in maximizing the effectiveness of such guidelines.
Stephen Weng, an epidemiologist at the University of Nottingham, used computer science to compare the ACC/AHA guidelines to four machine-learning algorithms, according to Digital Trends. The algorithms (random forest, logistic regression, gradient boosting, and neural networks) were designed to learn by itself, build its own set of guidelines, and test themselves for accuracy.
After analyzing 378,256 patient medical records from the United Kingdom, the artificial intelligence algorithms were used to find patterns that would point to possible cardiovascular diseases, according to worldhealth.net. The self-learning computers were then tested to analyze and predict possible diseases 10 years into the future, using real patient records from 2005.
All four A.I. algorithms outperformed their human counterparts based on their conclusions made on patient records, scoring at a rate of about 75% compared to the doctors’ 74%. Although the difference may not seem much, 355 patients who died from the 83,000 total cases were identified by the algorithms as possible candidates for cardiovascular problems. That means 355 humans ? with families, dreams, and purposes ? could have been saved.
As the medical machine learning is still in its novice state, we have much to hope for in the near future. Scientists and researchers are optimistic for the promising benefits of A.I. incorporated not only in cardiovascular disease prevention but others as well. However, acceptance of machine learning in the hospital must first be established. Artificial intelligence will cater to the health needs of this growing world, as Andy Schuetz of Sutter Health says, “I have no doubt that sophisticated learning and AI algorithms will find a place in healthcare over the coming years. I don’t know if it’s two years or ten ? but it’s coming.”
<Justin Lee, Cerritos HS 11th