New research that appears in the journal PLOS ONE suggests that machine learning can be a valuable tool for predicting the risk of premature death. The scientists compared the accuracy of artificial intelligence prediction with that of statistical methods that experts are currently using in medical research.
close up of doctor's hands using computer with stethoscope
New research suggests that healthcare professionals should use deep learning algorithms to predict premature death risk accurately.

An increasing amount of recent research is suggesting that computer algorithms and artificial intelligence (AI) learning can prove highly useful in the medical world.

For instance, a study that appeared a few months ago found that deep learning algorithms can accurately predict the onset of Alzheimer’s disease as early as 6 years in advance.

Using a so-called “training dataset,” deep learning algorithms can “teach themselves” to predict if and when an event is likely to occur.

Now, researchers have set out to examine whether machine learning can accurately predict premature mortality due to chronic disease.

Stephen Weng, who is an assistant professor of epidemiology and data science at the University of Nottingham in the United Kingdom, led the new research.

How AI could help preventative care

Weng and colleagues examined health data on more than half a million people between the ages of 40 and 69 years. The participants had registered with the UK Biobank study between 2006 and 2010. The UK Biobank study researchers clinically followed the participants until 2016.

For the current study, Weng and team developed a system of learning algorithms using two models called “random forest” and “deep learning.” They used the models to predict the risk of premature death due to chronic disease.

The scientists examined the predictive accuracy of these models and compared them with conventional prediction models, such as “Cox regression” analysis and a multivariate Cox model.

“We mapped the resulting predictions to mortality data from the cohort using Office of National Statistics death records, the U.K. cancer registry, and ‘hospital episodes’ statistics,” explains the study’s lead investigator.

The study found that the Cox regression model was the least accurate at predicting premature death, while the multivariate Cox model was slightly better but was likely to overpredict death risk.

Overall, “machine learning algorithms were significantly more accurate in predicting death than the standard prediction models developed by a human expert,” reports Weng. The researcher also comments on the clinical significance of the findings.

He says, “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 computerized health risk assessment in the general population.”

“Most applications focus on a single disease area, but predicting death due to several different disease outcomes is highly complex, especially given environmental and individual factors that may affect them.”

We have taken a major step forward in this field by developing a unique and holistic approach to predicting a person’s risk of premature death by machine learning.”

Stephen Weng

“This 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,” explains Weng.

Furthermore, say the researchers, the results of the new study strengthen previous findings, which showed that certain AI algorithms are better at predicting heart disease risk than the conventional prediction models that cardiologists currently use.

“There is currently intense interest in the potential to use ‘AI’ or ‘machine learning’ to better predict health outcomes. 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,” says Prof. Joe Kai, a clinical academic who also worked on the study.

He continues, “These techniques can be new to many in health research and difficult to follow. We believe that by clearly reporting these methods in a transparent way, this could help with scientific verification and future development of this exciting field for healthcare.”