Researchers from the U.K.-based University of Nottingham have created algorithms capable of predicting premature death.

With implications for the future of preventive healthcare, the computer-based machine learning algorithms accurately predict the risk of premature death due to chronic disease among a large middle-age population better than current prediction models developed by humans.

Developed and tested by a team of healthcare data scientists and doctors, the algorithm was trained on health data from the U.K. Biobank, which holds the health data of roughly half a million people, ages 40 to 69, recruited between 2006 and 2010 from across the U.K. The team applied artificial intelligence (AI) machine learning models called random forest and deep learning to the data, outperforming the commonly used Cox regression prediction model, which is based on age and gender.

Study lead and Assistant Professor of Epidemiology and Data Science Dr. Stephen Weng said: "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.

"We mapped the resulting predictions to mortality data from the cohort, using Office of National Statistics death records, the UK cancer registry and 'hospital episodes' statistics. We found machine learned algorithms were significantly more accurate in predicting death than the standard prediction models developed by a human expert."

The team believes that the future of preventive healthcare will be impacted by AI, helping to improve health outcome predictions. Yet for this to happen, the team cautions that such algorithms will need to undergo further testing.

The research is published in the journal PLOS ONE.