Revolutionizing Health Predictions: AI Forecasts Future Medical Risks

Fri Sep 19 2025 00:22:12 GMT+0300 (Eastern European Summer Time)

Utilizing advanced AI technology, researchers have created a model capable of predicting an individual's health risks a decade into the future, similar to weather forecasts. This innovative approach aims to transform early disease diagnosis and healthcare planning.

Researchers have introduced an advanced AI model named Delphi-2M, capable of predicting health risks for over 1,000 diseases using patterns from medical records. This technology mimics weather forecasting, aiming to identify high-risk patients and inform healthcare planning years in advance. While still in research phases, it opens doors to personalized medical interventions and better healthcare resource management.

Artificial intelligence can predict people's health problems over a decade into the future, say scientists. The technology has learned to spot patterns in people's medical records to calculate their risk of more than 1,000 diseases.

The researchers describe the capability as akin to a weather forecast that anticipates a 70% chance of rain—but for human health.

With this AI model, high-risk patients can be identified, enabling proactive disease prevention and improved healthcare management years ahead of time.

Known as Delphi-2M, the model employs similar technology to AI chatbots like ChatGPT, which are trained to identify patterns in data. Instead of predicting textual sequences, Delphi-2M analyzes medical records to estimate the likelihood of various diseases over specified periods.

According to Prof Ewan Birney, interim executive director of the European Molecular Biology Laboratory, Just like weather, where we might predict a 70% chance of rain, we can apply that same concept for healthcare. This model allows predictions for a multitude of diseases concurrently, a significant advancement over previous methodologies.

Initially developed using anonymous UK data—including hospital admissions and GP records from over 400,000 individuals participating in the UK Biobank project—the model showed promising results when tested against additional data from 1.9 million Danish medical records.

The AI model excels in predicting diseases with clear progression metrics, such as type 2 diabetes and heart attacks. Plans are underway to adapt the model for broader clinical applications, including the identification of patients who may benefit from lifestyle changes or early interventions.

This beginning of a new way to understand human health and disease progression could personalize care and anticipate healthcare needs at scale, stated Prof Moritz Gerstung from the German Cancer Research Centre.

The Delphi-2M model is still under refinement and testing. It faces challenges such as potential biases from its initial data set, primarily sourced from individuals aged 40 to 70. Future upgrades will integrate additional data types to enhance accuracy.

Despite being in a research phase, Prof Birney believes that with careful regulation and testing, this revolutionary technology could seamlessly augment healthcare practices in the years to come.