The application of artificial intelligence in human activities has been growing in recent years. From government operations, aviation and the manufacturing industry, the world continues to seek ways to partner robots with humans. The medical community is also trying to keep up with the tech trend. A new study published in the journal PLOS ONE suggests that doctors may soon have the opportunity to work with AI technology to predict a patient’s death and prevent it. Researchers from the University of Nottingham in England found that a recently developed system of computer-based machine learning algorithms built by healthcare data scientists and doctors was able to accurately predict the risk of early death based on the chronic diseases in a large population. The AI system also performed better than the current standard approach used by doctors to observe a patient’s risks, EurekAlert reported. For the study, the researchers provided the system with health data gathered from more than half a million people aged between 40 and 69 across the United Kingdom between 2006 and 2010 and in 2016. "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,” researcher Stephen Weng, an assistant professor of epidemiology and data science at Nottingham, said. “We found machine learned algorithms were significantly more accurate in predicting death than the standard prediction models developed by a human expert." The AI tool evaluated diet, demographic, biometric, clinical and lifestyle factors for each individual to provide predictions. Professor Joe Kai, one of the clinical academicians working on the project, said the study comes amid the “intense interest” of the medical community in the potential use of AI or machine learning to predict health outcomes. He added their findings will guide future research on application of the technology in healthcare. The Nottingham researchers said AI has the potential to help improve delivery of personalised medicine and tailor risk management for patients in the future. However, they noted further studies are required to confirm the accuracy of the technology. Source
