Artificial Intelligence can help predict premature deaths

Researchers have developed an Artificial Intelligence (AI)-based system to predict the risk of early deaths due to chronic disease in middle-aged adults.

The study, published by PLOS ONE journal, found that the new AI Machine Learning models known as "random forest" and "deep learning" were very accurate in its predictions and performed better than the current standard approach to prediction developed by human experts.

Such new risk prediction models take into account demographic, biometric, clinical and lifestyle factors for each individual, and assess even their dietary consumption of fruit, vegetables and meat per day, said Stephen Weng, Assistant Professor at the University of Nottingham in Britain.

The traditionally-used "Cox regression" prediction model, based on age and gender, was found to be the least accurate at predicting mortality and also a multivariate Cox model which worked better but tended to over-predict risk.

"Preventative healthcare is a growing priority in the fight against serious diseases so we have been working for a number of years to improve the accuracy of computerised health risk assessment in the general population," said Weng.

For the study, the team included over half a million people aged between 40 and 69. Although these techniques could be new to many in health research and difficult to follow, clearly reporting these methods in a transparent way could help with scientific verification and future development of AI for health care, said Joe Kai, Professor at the varsity.