AI systems found to be better than doctors at gauging heart attack risk

April 17, 2017 by Bob Yirka report
Credit: CC0 Public Domain

(Medical Xpress)—A team of researchers at the University of Nottingham in the U.K. has found that artificial intelligence systems can be better at gauging a patient's risk of a heart attack than doctors using a standard protocol. In their paper uploaded to the open access site PLOS ONE, the team describes how they tested four AI systems against humans using a protocol, what they found, and why they believe there is still room for improvement.

Most doctors today use a protocol or guideline very much like the one created by the American College of Cardiology/American Heart Association (ACC/AHA). It is essentially a questionnaire assessing and generating a statistical result. The guideline accounts for such factors as a patient's age, weight, blood pressure and cholesterol levels. Prior research has shown that by using such a guideline, doctors can correctly assess a patient's risk of having a heart attack at a rate of 72.8 percent, which is quite impressive. But it still leaves a lot of room for improvement, because that failure rate of 27.2 percent oftentimes represents people who die.

In this new effort, the researchers sought to find out if computers running AI systems might do a better job of it. They set up four such AI systems: logistic regression, random forest, neural networks and gradient boosting. Each was given thousands of patient data records from a British medical database as a means to study health patterns in patients and to identify . They were then each assigned the task of predicting a heart attack in those same patients over the course of the next 10 years—the data covered the period 2005 to 2015.

The researchers report that all four AI systems did better than the human average using the ACC/AHA guidelines, ranging from 74.5 percent to 76.4 percent correct, which, the team noted, possibly amounted to 355 patients whose lives could have been saved had they known they were at risk. They report also that the systems found risk factors that are not currently used by doctors to assess risk, such as mental illness and the use of oral corticosteroids. They suggest that tweaking the system and perhaps adding other known risks manually could improve the rate even more.

Explore further: Study finds most major heart attacks occur in people with normal cholesterol

More information: Stephen F. Weng et al. Can machine-learning improve cardiovascular risk prediction using routine clinical data?, PLOS ONE (2017). DOI: 10.1371/journal.pone.0174944

Abstract

Background
Current approaches to predict cardiovascular risk fail to identify many people who would benefit from preventive treatment, while others receive unnecessary intervention. Machine-learning offers opportunity to improve accuracy by exploiting complex interactions between risk factors. We assessed whether machine-learning can improve cardiovascular risk prediction.

Methods
Prospective cohort study using routine clinical data of 378,256 patients from UK family practices, free from cardiovascular disease at outset. Four machine-learning algorithms (random forest, logistic regression, gradient boosting machines, neural networks) were compared to an established algorithm (American College of Cardiology guidelines) to predict first cardiovascular event over 10-years. Predictive accuracy was assessed by area under the 'receiver operating curve' (AUC); and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) to predict 7.5% cardiovascular risk (threshold for initiating statins).

Findings
24,970 incident cardiovascular events (6.6%) occurred. Compared to the established risk prediction algorithm (AUC 0.728, 95% CI 0.723–0.735), machine-learning algorithms improved prediction: random forest +1.7% (AUC 0.745, 95% CI 0.739–0.750), logistic regression +3.2% (AUC 0.760, 95% CI 0.755–0.766), gradient boosting +3.3% (AUC 0.761, 95% CI 0.755–0.766), neural networks +3.6% (AUC 0.764, 95% CI 0.759–0.769). The highest achieving (neural networks) algorithm predicted 4,998/7,404 cases (sensitivity 67.5%, PPV 18.4%) and 53,458/75,585 non-cases (specificity 70.7%, NPV 95.7%), correctly predicting 355 (+7.6%) more patients who developed cardiovascular disease compared to the established algorithm.

Conclusions
Machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others.

Related Stories

Study finds most major heart attacks occur in people with normal cholesterol

April 12, 2017
For decades, national guidelines on which Americans should take cholesterol-lowering medications relied heavily on an individual's level of bad cholesterol (LDL). In 2013, new guidelines moved to treatment based on a person's ...

Younger heart attack patients more likely to have low 'good' vs. high 'bad' cholesterol

March 8, 2017
Men under 45 years old and women under 50 years old who suffer a heart attack are far more likely to have abnormally low good cholesterol than elevated bad cholesterol, according to research scheduled for presentation at ...

Too many Americans have high blood pressure, doctors warn

February 7, 2017
(HealthDay)—A group of family physicians warns that too many Americans struggle with high blood pressure.

A simple blood test could help identify people at risk of a heart attack, a study has found

December 21, 2016
Research shows the troponin test – currently used to help diagnose a heart attack – could be used to assess future heart disease risk.

Young women less likely to be informed of heart disease risk by providers

October 26, 2015
Even when young women had a similar or greater risk for heart disease than young men, they were 11 percent less likely to report that healthcare providers told them they were at-risk for heart disease prior to a heart attack, ...

Low statin use in people with diabetes despite cardioprotective effects, guidelines

September 12, 2016
Nearly 2 out of 5 people with diabetes who could benefit from statin therapy to lower their risk of future heart attack, stroke and related death were not prescribed one, according to a research letter published today in ...

Recommended for you

Could aggressive blood pressure treatments lead to kidney damage?

July 18, 2017
Aggressive combination treatments for high blood pressure that are intended to protect the kidneys may actually be damaging the organs, new research from the University of Virginia School of Medicine suggests.

Quantifying effectiveness of treatment for irregular heartbeat

July 17, 2017
In a small proof-of-concept study, researchers at Johns Hopkins report a complex mathematical method to measure electrical communications within the heart can successfully predict the effectiveness of catheter ablation, the ...

Concerns over side effects of statins stopping stroke survivors taking medication

July 17, 2017
Negative media coverage of the side effects associated with taking statins, and patients' own experiences of taking the drugs, are among the reasons cited by stroke survivors and their carers for stopping taking potentially ...

Study discovers anticoagulant drugs are being prescribed against safety advice

July 17, 2017
A study by researchers at the University of Birmingham has shown that GPs are prescribing anticoagulants to patients with an irregular heartbeat against official safety advice.

Protein may protect against heart attack

July 14, 2017
DDK3 could be used as a new therapy to stop the build-up of fatty material inside the arteries

Heart study finds faulty link between biomarkers and clinical outcomes

July 14, 2017
Surrogate endpoints (biomarkers), which are routinely used in clinical research to test new drugs, should not be trusted as the ultimate measure to approve new health interventions in cardiovascular medicine, according to ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.