Edinburgh University makes breakthrough in predictive healthcare

Artificial intelligence could help predict those at risk of major diseases before they are even diagnosed, researchers have found.

The University of Edinburgh, in collaboration with Optima Partners and Biogen, used machine learning to identify protein patterns linked to a range of diseases, which allowed them to predict a person’s risk of illness up to 10 years before diagnosis.

The team explored a list of more than 20 diseases, including heart disease, which is responsible for around 20 deaths per day north of the border, and type 2 diabetes, present in around six per cent of Scotland’s population.

Mental health conditions were also studied, including Alzheimer's, which is the UK's biggest killer, according to the latest findings from the Office for National Statistics.

The study suggested current patient blood tests could be compared with the identified proteins, meaning practitioners could detect the possibility of a particular disease developing later in life, allowing for an earlier treatment and better outcomes.

Danni Gadd, first author of the study, said: “Our research represents a promising step forward in risk prediction.

“Being able to detect early warning signs for a broad set of conditions may lead to opportunities for early intervention and prevention, marking a significant moment for the healthcare industry.” 

Danni Gadd, first author of the study

Analysing data from almost 50,000 individuals who had a blood sample taken between 2006 and 2010, the research improved the prediction for disease outcomes diagnosed up to 15 years after the initial blood sample was taken.

However, Chris Foley, chief executive of Optima Partners, said "more work is still needed" before research is used in clinical settings.

The research comes after former prime minister Tony Blair urged Labour to roll out AI tech across the NHS so it could adopt a preventative health care approach.