Artificial Intelligence Beats Doctors At Predicting Heart Attacks

Every year an estimated 20 million people all across the globe die due to cardiovascular disease. It becomes the leading cause for both men and women and most of the times these heart attacks are preventable. Doctors are pretty good at identifying the early symptoms of heart attacks and also warning the patients about the upcoming risks. But now it became little surprising to know that computers do the job even better than the doctors.

Scientists at the University of Nottingham have a new way of anticipating heart failure among patients. The researchers developed a set of computer programs which used artificial intelligence that could predict heart attacks even better than the doctors. These algorithms were trained on real patient records and developed criteria which then outperformed the current guidelines set by the American Heart Association (AHA)

The AHA has developed a set of guidelines to estimate a patient’s probable risk of cardiovascular disease, which is based on 8 factors including age, cholesterol level, and blood pressure. On average, this system correctly guesses a person’s risk at a rate of 72.8 percent.

This is how the research went on:

• Stephen Weng, an epidemiologist at the UK University and his team built four computer learning algorithms and fed them data from as many as 4 lakh patients suffering from cardiovascular disease.
• The system first used 295,000 records fro generating their internal predictive models
• Then they used the remaining records to test and refine them.
• These algorithm results outperformed the AHA guidelines by 7.6 percent while raising 1.6 percent fewer false alarms.

It is interesting to learn that the AI systems identified a number of risk factors like severe mental illness and the consumption of oral corticosteroids which are not currently included in the AHA guidelines.

Weng told in an interview: “There is a lot of interactions in biological systems. That is the reality of the human body. What computer science allows us to do is to explore those associations.”