UK Scientists/Researchers Computer Can 'Predict When You Are Going to Die With 95% Accuracy'
University of Nottingham UK Scientific Rearchers have used Artificial Intelligence to predict early death in middle age to 95 per cent accuracy
Daily Mirror UK, 29 MAR 2019.
UK Scientists say they have found a way to predict the risk of early death with 95 per cent accuracy...
The dream of being able to predict our own life-spans has been explored in the realms of science fiction for decades.
But in a concept straight out of the plot of 1997 Ethan Hawke dystopian flick Gattaca, researchers now believe computers can predict early death - with 95 per cent accuracy.
University of Nottingham scientists say being able to predict who is at risk of dying early could soon be a reality, thanks to leaps and bounds in artificial intelligence (AI) technologies.
Their study , published this week, claims the AI prediction models they used proved significantly better at predicting early death among middle-aged people than models designed by human medical experts.
The future of Artificial Intelligence could see machines understand more about our health than our own doctors ......
Clever computer machine-learning models allowed the researchers to take into account middle aged people's lifestyle factors that could contribute to premature death.
Traditional predictions are based on demographics - focused on age and gender.
But by using machine-learning, the researchers were able to go into much great detail - even taking into account the amount of vegetables a person included in their diet in order to predict their mortality.
The UK researchers say the study has the potential to be used to improve future health-care to prevent people dying young from chronic disease, like cancer.
The study, published in medicine journal PLOS ONE, looked at the health data of just over half a million people aged between 40-69 years.
The participants were recruited into the study via the UK Biobank, Britain's long-term genetic health research resource.
Traditionally, scientists have predicted early death using models based on age and gender, which is said to have a weak record of prediction mortality.
The new study used AI machine learning models known as "random forest" and "deep learning".
Machines may one day know us better than we know ourselves (Image: Getty Images)
Previous research by the same Nottingham team found using the newer models more accurately predicted early death from cardiovascular disease than age-and-gender based algorithms medical experts currently use.
Using computer machine-learning, the AI system was able to build upon the age-and-gender based predictions, by also exploring biometric, clinical and lifestyle data.
Just over 14,000 of the people in the study died during the research period between 2006 and 2010.
The researchers were able to analyse those deaths using mortality data from the Office of National Statistics, the national cancer registry, and "hospital episodes."
When they compared the outcomes to the AI predictions, they found their models were 95 per cent accurate.
The researchers used computer machine-learning models to analyse middle-aged people's risk of early death (Image: Getty Images)
AI is being hailed as a potential lifesaver in the field of UK medicine, as experts look for ways to help us live longer, healthier lives and prevent unnecessary illness.
But critics fear the rise of Big Data in the health sector could lead to our health risks being used against us by the likes of US insurance companies - or, such as in the future imagined in Gattaca - US employers, and even prospective romantic partners.
However the researchers predict future medicine will focus on tailoring health-care to individual patients based on their personal risk profiles.
The study's lead researcher, Assistant Professor of Epidemiology and Data Science, Dr Stephen Weng, said the study could help fight early death from chronic disease.
“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 computerised health risk assessment in the general population," he said.
" 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."
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."
The UK researchers say further study is required to explore if the predictions work in other population groups.
They pointed out the UK Biobank tends to attract healthier volunteers from, educated, and wealthier backgrounds than the average Brit.
The researchers say they hope their findings can be used to investigate ways to use what they have found in everyday health-care to prevent early death in the near future.
.
UK Scientists' Can Predict WHEN You are Going to DIE >95% Accuracy
PUTIN TRUMP & Netanyahu Will Meet in HELL
1.. 
Linear Mode
