Experts state that a new AI tool predicts your disease risk.
Delphi-2M diagnoses an individual’s medical history and lifestyle factors (e.g. past obesity, smoker, or drinker) and basic factors (e.g. sex and age) to facilitate health-related predictions, including one’s specific risk for more than 1,000 diseases, for a decade or more into the future.
“Medical events often follow predictable patterns,” Tomas Fitzgerald, a staff scientist at EMBL’s European Bioinformatics Institute (EMBL-EBI), told The Guardian. “Our AI model learns those patterns and can forecast future health outcomes.”
For example, you’re a trim 26-year-old who eats whatever you want and doesn’t exercise. This tool might tell you that there’s a 75 percent chance you’ll be overweight and at-risk for heart disease and diabetes by 50 if you don’t change something. And even if you’re 26 and have a healthy lifestyle, it still may predict, based on your medical history, you have a fair chance at getting skin cancer in your mid-40s. Or maybe you’re a 65-year-old who is in good shape now but will be vulnerable to a heart attack in 10 years. It’s better to know now and take the necessary precautions.
In sum, such predictions will allow patients to get ahead of the curve and address health vulnerabilities well before they happen. Think of it like a weather forecast that tells you there’s a 70% chance of a slow blizzard this weekend.
If you live in my area, you might cancel that weekend getaway in the country and stock emergency supplies like food, water, blankets, and batteries, and anything else needed to stay safe and warm.
Ewan Birney, the EMBL interim executive director, told The Guardian: “You walk into the doctor’s surgery and the clinician is very used to using these tools, and they are able to say: ‘Here’s four major risks that are in your future and here’s two things you could do to really change that.’
“I suspect everyone will be told to lose weight, and if you smoke you will be told to stop smoking – and that will be in your data so that advice isn’t going to change remarkably – but for some diseases I think there will be some very specific things. That’s the future we want to create.”
The generative AI tool, which was tested on anonymised patient records from 400,000 400,000-person cohort in the UK Biobank study, as well as 1.9 million patients in the Danish national patient registry, was custom-built by experts from the European Molecular Biology Laboratory (EMBL), the German Cancer Research Centre, and the University of Copenhagen. It is among the most well-developed models to date in terms of how generative AI can be used at scale to model human disease progression, having been trained on data from two entirely separate health systems.
When will this tool become available?
Ewan Birney, the interim executive director of EMBL, suggested that patients may be able to benefit from this tool within a couple of years.
