New AI model predicts diseases 10 years ahead

Scientists have announced that with their newly developed artificial intelligence model, they can predict diseases people might develop in the next 10 years, much like weather forecasting.

The "Delphi-2M" AI model, developed by researchers from the European Molecular Biology Laboratory, the German Cancer Research Center (DKFZ), and the University of Copenhagen, can predict diseases a person may contract within 10 years by using data from their medical records.

The model uses a technology similar to ChatGPT, which is trained to understand language patterns to predict the sequence of words in a sentence.

Delphi-2M is trained to detect patterns in anonymous medical records and forecasts diseases people might develop within a decade.

The software was developed using anonymous medical records from over 400,000 people in the UK and later tested with data from other participants and 1.9 million people in Denmark.

Researchers noted that the AI achieves high accuracy in predicting progressive diseases such as type 2 diabetes, heart attacks, and sepsis.

Professor Ewan Birney, one of the lead scientists, said, "Just as we can say there is a 70% chance of rain in weather forecasts, we can do the same in healthcare."

The researchers emphasized that the model is not yet ready for clinical use but could be utilized for early diagnosis and preventive treatments to identify high-risk patients.

Additionally, the model is expected to help healthcare institutions plan for future patient loads.

The researchers pointed out that most of the data used in the study came from people aged 40-70, and the model needs further improvement and testing before clinical application.

The model is being updated to include more medical data such as imaging, genetics, and blood analysis.

The research findings were published in the journal Nature.