In a new study, researchers found that machine learning is overtaking humans in predicting death or heart attack.
Machine learning is the modern bedrock of artificial intelligence (AI).
It is used every day, such as in Google’s search engine, face recognition on smartphones, self-driving cars, Netflix and Spotify recommendation systems.
In the study, the team trained an algorithm to learn to predict heart attack symptoms by repeatedly analyzing 85 variables in 950 patients.
It then identified patterns correlating the variables to death and heart attack with more than 90% accuracy.
In the analysis, the patients had chest pain and underwent the usual protocol to look for coronary artery disease.
A coronary computed tomography angiography (CCTA) scan yielded 58 pieces of data on the presence of coronary plaque, vessel narrowing, and calcification.
Those with scans suggestive of disease underwent a positron emission tomography (PET) scan which produced 17 variables on blood flow.
Ten clinical variables were obtained from medical records including sex, age, smoking, and diabetes.
The team suggests that these advances are far beyond what has been done in medicine, where doctors need to be cautious about evaluating risk and outcomes.
Compared with human doctors who only have modest accuracy in predicting heart attacks in individual patients,
AI can exploit large amounts of data and identify complex patterns that may not be evident to humans.
The team hopes their new finding can help develop personalized treatment and ultimately lead to better outcomes for patients.
One author of the study is Dr. Luis Eduardo Juarez-Orozco, of the Turku PET Centre, Finland.
The study was presented at ICNC 2019.
Copyright © 2019 Knowridge Science Report. All rights reserved.
