AI predicts human disease risk up to 20 years before symptoms emerge

We’ve seen Artificial Intelligence (AI) models that can predict human disease, and even a rough time of death, before but now a new system can predict human disease risks up to 10 to 20 years before symptoms even occur. As Innovation Origins reported, researchers developed Delphi-2M using over 40,000 Biobank participants and 1.9 million Danish medical records. It can accurately predict the progression of diseases, including type 2 diabetes and heart attacks.

The researchers collaborated from the German Cancer Research Center, the University of Copenhagen, and the European Molecular Biology Laboratory. They published their scientific findings in the journal Nature.

The AI tool identifies high-risk individuals and enables doctors to implement early intervention measures.

The Future of Healthcare 2030, by Futurist Matthew Griffin

It does this by analyzing patients' medical history and relevant factors, such as obesity and age. It's a unique tool because it can predict over 1,000 diseases simultaneously with impressive accuracy. Previous models reportedly could only predict single, targeted diseases.

The researchers said their AI tool can predict diseases up to 20 years in advance, especially cancers, heart attacks, and other conditions with predictable disease progression.

This amazing technological development offers hope for improving people's health and empowering individuals to make positive lifestyle changes to halt the progression of disease.

There's no denying that AI has concerning environmental impacts due to its extreme energy usage and strain on electrical grids. However, the technology also holds great promise for predicting everything from diseases to dangerous storms to the clean energy transition.

As AI technology continues to advance, it has the potential to address many of the critical climate issues our world faces.

Regarding Delphi-2M, the researchers plan to conduct additional tests of its AI tool on different population groups and in various countries. They recognize the need to refine further and test the tool before it's ready for clinical use, hopefully within a few years.

In their publication, the researchers wrote, "In summary, transformer-based models appear to be well suited for predictive and generative health-related tasks, are applicable to population-scale datasets and provide insights into temporal dependencies between disease events, potentially improving the understanding of personalized health risks and informing precision medicine approaches."

"If our model says it's a one-in-10 risk for the next year, it really does seem like it turns out to be one in 10," Professor Ewan Birney from the European Molecular Biology Laboratory commented on the AI tool's reliability, per Innovation Origins.