AI Case Study
University of Nottingham uses machine learning to beat doctors at predicting who will have heart attacks over the next ten years that could result in an additional 355 additional lives being saved
Every year millions die from heart attacks. The current guidelines for predicting heart attacks are too simplistic and do not take into account the complex of interaction of lifestyle and disease. The University of Nottingham built prediction models with random forest, logistic regression, gradient boosting and neural networks models trained on nearly 400,000 patient records. All methods improved on current guidelines with neural networks predicting 7.6% more events and raising 1.6% fewer false alarms.
Public And Social Sector
Education And Academia
University of Nottingham researchers "compared use of the ACC/AHA guidelines with four machine-learning algorithms: random forest, logistic regression, gradient boosting, and neural networks. All four techniques analyze lots of data in order to come up with predictive tools without any human instruction. In this case, the data came from the electronic medical records of 378,256 patients in the United Kingdom."
"First, the artificial intelligence (AI) algorithms had to train themselves. They used about 78% of the data—some 295,267 records—to search for patterns and build their own internal “guidelines.” They then tested themselves on the remaining records. Using record data available in 2005, they predicted which patients would have their first cardiovascular event over the next 10 years, and checked the guesses against the 2015 records. Unlike the ACC/AHA guidelines, the machine-learning methods were allowed to take into account 22 more data points, including ethnicity, arthritis, and kidney disease."
R And D
Core Research And Development
"Each year, nearly 20 million people die from the effects of cardiovascular disease, including heart attacks, strokes, blocked arteries, and other circulatory system malfunctions. In an effort to predict these cases, many doctors use guidelines similar to those of the American College of Cardiology/American Heart Association (ACC/AHA). Those are based on eight risk factors—including age, cholesterol level, and blood pressure—that physicians effectively add up. But that’s too simplistic to account for the many medications a patient might be on, or other disease and lifestyle factors." The question is whether machine learning could be used to analyse interactions and make better prediction in complex biological systems.
"All four AI methods performed significantly better than the ACC/AHA guidelines...the best one—neural networks—correctly predicted 7.6% more events than the ACC/AHA method, and it raised 1.6% fewer false alarms. In the test sample of about 83,000 records, that amounts to 355 additional patients whose lives could have been saved. That’s because prediction often leads to prevention, Weng says, through cholesterol-lowering medication or changes in diet.
They used "four machine-learning algorithms: random forest, logistic regression, gradient boosting, and neural networks."
"They used about 78% of the data—some 295,267 records—to search for patterns and build their own internal 'guidelines.' They then tested themselves on the remaining records."
"Unlike the ACC/AHA guidelines, the machine-learning methods were allowed to take into account 22 more data points, including ethnicity, arthritis, and kidney disease."