1. Home
  2. >>
  3. death
  4. >>
  5. Artificial Intelligence Could Predict Premature Death, Study Finds

Artificial Intelligence Could Predict Premature Death, Study Finds

Artificial Intelligence Could Predict Premature Death, Study Finds

Computers are capable of teaching themselves to predict premature death, which could greatly improve preventive health care in the future, a new study by experts from the University of Nottingham in the UK suggests.

The healthcare data team of scientists and physicians has developed and tested a machine learning computer algorithm system to predict the risk of premature death due to chronic disease in a large middle-aged population.

The researchers found that this artificial intelligence system was very accurate in its predictions and performed better than the current standard approach to prediction developed by human experts, as revealed in an article about the study published by “PLOS ONE” in a edition of special collections of «Machine Learning in Health and Biomedicine».

The team used health data from just over half a million people aged 40 to 69 recruited into the UK Biobank between 2006 and 2010 and followed through 2016.

Leading the work, Assistant Professor of Epidemiology and Data Science, Dr. Stephen Wengsays: “Preventive health care is an increasing priority in the fight against serious diseases, which is why we have been working for several 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 different disease outcomes is very complex, especially due to the environmental and individual factors that can affect them. ‘

He adds: “We have taken a great step forward in this field by developing a unique and holistic approach to predict a person’s risk of premature death using machine learning. It uses computers to create new risk prediction models that take into account a wide range of demographic characteristics, biometric, clinical and lifestyle factors for each individual tested, including their dietary intake of fruits, vegetables and meat per day.

«We assigned the resulting predictions to the mortality data of the cohort, using the death records of the Office for National Statistics, the UK Cancer Registry and ‘hospital episodes’ statistics. The machine-learned algorithms were significantly more accurate in predicting death than the standard prediction models developed by a human expert, ”he stresses.

The future of health

The machine learning models of AI used in the new study are known as “random forest” and “deep learning.” These were compared with the traditionally used ‘Cox regression’ prediction model based on age and gender (found to be the least accurate in predicting mortality) and also a multivariate Cox model that performed better, but tended to predict risk.

The teacher Joe Kai, one of the clinical academics working on the project, says that “there is currently great interest in the potential of using ‘AI’ or ‘machine learning’ to better predict health outcomes. In some situations, we may find it helpful. In other cases, it may not. In this particular case, we have shown that, with careful tuning, these algorithms can improve prediction. ‘

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

This new study builds on previous work by the Nottingham team which showed that four different AI algorithms, ‘random forest’, ‘logistic regression’, ‘gradient augmentation’ and ‘neural networks’ were significantly better at predicting cardiovascular disease than an established algorithm used in current cardiology guidelines.

Nottingham researchers predict that AI will play a vital role in the development of future tools capable of delivering personalized medicines, adapting risk management to individual patients. Further research requires verifying and validating these AI algorithms in other population groups and exploring ways to implement these systems in routine healthcare.