17 Sep 2019

logo

Artificial intelligence outperforms traditional approach in predicting premature death: study

 

WASHINGTON, March 31 (Xinhua) — Computers could predict one’s chance of premature death more accurately than traditional approaches, according to a study recently published on scientific journal PLOS ONE.

Researchers from the British University of Nottingham have developed and tested a system of computer-based “machine learning” algorithms to predict the risk of early deaths due to chronic disease in a large middle-aged population.

This method uses computers to build new risk prediction models that take into account a wide range of demographic, biometric, clinical and lifestyle factors for each individual assessed, even their dietary consumption of fruit, vegetables and meat per day.

They found this artificial intelligence (AI) system was very accurate in its predictions and performed better than the current standard approach, known as the Cox model, developed by human experts.

The study recruited over 500,000 participants aged 40-69 from 2006 and 2010, and followed them up until 2016. During the 2006-2010 time period, nearly 14,500 people died, largely from cancer and other diseases.

The two machine-learning approaches, namely deep learning and random forest, correctly identified 76 percent and 64 percent of the premature deaths respectively. By comparison, the traditional Cox model identified only about 44 percent.

“There is currently intense interest in the potential to use ‘AI’ or ‘machine-learning’ to better predict health outcomes,” said University of Nottingham professor Joe Kai, one of the researchers of the study, in a statement.

“In some situations we may find it helps, in others it may not. In this particular case, we have shown that with careful tuning, these algorithms can usefully improve prediction,” he said.

The Nottingham researchers predict that AI will play a vital part in the development of future tools capable of delivering personalized medicine, tailoring risk management to individual patients.

Please follow and like us: