A new study has shown how the bedrock of AI – machine learning – is now more accurate at predicting death or a heart attack.
Machine learning through artificial intelligence (AI) has reached a tipping point in healthcare when it comes to predictive medicine, according to new research presented at the International Conference on Nuclear Cardiology and Cardiac CT.
Dr Luis Eduardo Juarez-Orozco of the Turku PET Centre in Finland revealed that machine learning is overtaking humans in predicting death or heart attacks, based on data from almost 1,000 patients.
The study involved patients with chest pain who underwent the centre’s usual protocol to look for coronary artery disease, with 58 pieces of data on the presence of coronary plaque, vessel narrowing and calcification discovered during scans.
During a six-year follow-up, the researchers recorded 24 heart attacks and 49 deaths from various causes. 85 variables were then entered into an algorithm called LogitBoost, which analysed them over and over again until it found the best structure to predict who had a heart attack or died.
In a relatively short time, LogitBoost was able to identify patterns correlating the variables to death and heart attack with more than 90pc accuracy.
“Humans have a very hard time thinking further than three dimensions or four dimensions,” said Juarez-Orozco. “The moment we jump into the fifth dimension, we’re lost. Our study shows that very high-dimensional patterns are more useful than single-dimensional patterns to predict outcomes in individuals, and for that we need machine learning.”
He added: “Doctors already collect a lot of information about patients – for example, those with chest pain. 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.”
Elsewhere in the medtech space, AI is also expected to save the lives of thousands of animals used in chemical toxicity testing. By comparing 85,000 chemical compounds with one of the largest databases of chemicals around, it is hoped that a new algorithm could tell researchers which chemicals are toxic without harming any life.
Updated, 3.31pm, 13 May 2019: This article was amended to clarify that the researchers recorded 49 deaths from various causes, not 39.