Data Science Investment Counter —Funding raised by UK data science companies in 2018.
£ 5.640 Million

Machine learning could predict death or heart attack with over 90 per cent accuracy

Artificial intelligence could predict death or heart attacks with more than 90 per cent accuracy, a new study has revealed.

Presented at the International Conference on Nuclear Cardiology and Cardiac CT 2019 on Sunday, the study showcased the results of an algorithm called LogitBoost.

Powered by machine learning, the algorithm analysed 85 different variables from 950 patients of the centre with chest pain who underwent the usual routine checks to look for coronary artery, and who the researchers had followed for six years. It then predicted with an accuracy of more than 90 per cent which of the participants had died or suffered heart attacks.

“The algorithm progressively learns from the data and after numerous rounds of analyses, it figures out the high dimensional patterns that should be used to efficiently identify patients who have the event,” said Luis Eduardo Juarez-Orozco, an author of the research from the Turku PET Centre in Finland. “The result is a score of individual risk.”

Doctors use risk scores to make treatment decisions. Because of this, accuracy in predictions in paramount. However, these scores are normally based on a few variables and often have limited accuracy in individual patients.

In this light, Juarez-Orozco highlighted the importance of machine learning and artificial intelligence in processing data for medical purposes. Because these technologies can exploit large amounts of data, they are able to identify complicated patterns that may not be evident to humans.

“Doctors already collect a lot of information about patients—for example, those with chest pain,” he said. “We found that machine learning can integrate these data and accurately predict individual risk. This should allow us to personalise treatment and ultimately lead to better outcomes for patients.”

However, Juarez-Orozco added, further efforts are needed to make the most out of these technologies: “these advances are far beyond what has been done in medicine, where we need to be cautious about how we evaluate risk and outcomes.”

“We have the data but we are not using it to its full potential yet.”

More information about the research and the Turku PET Centre is available on their website here.

Image via Pixabay.


Co-working space and blog dedicated to all things data science.

Subscribe to our newsletter