Story perspectives

New AI Tool Predicts 1,000+ Diseases for Future Healthcare

9/18/2025

42 7

Summary

Story summary

A new AI tool predicts the risk of over 1,000 diseases, targeting use by doctors in 5-10 years.
It utilizes anonymized data from 400,000 UK Biobank patients and has been tested on 1.9 million Danish patients.
The model evaluates medical history and lifestyle factors to forecast disease probabilities.
It is particularly effective for diseases with clear progression, such as cancer and heart attacks.
The study in *Nature* underscores its potential for personalized healthcare and early intervention, estimating health trajectories for up to 20 years.

History

Historical Context

The General Data Protection Regulation (GDPR) was enacted by the European Union in 2016 to enhance data privacy and ethical standards. This regulation has influenced AI research, notably by Yoshua Bengio's group, which advocates for ethical AI practices. As AI technologies like Delphi-2M emerge, ensuring diverse training data and ethical considerations is vital for their integration into healthcare and public health strategies.

1 of 2

Story summary

A new AI tool predicts the risk of over 1,000 diseases, targeting use by doctors in 5-10 years.
It utilizes anonymized data from 400,000 UK Biobank patients and has been tested on 1.9 million Danish patients.
The model evaluates medical history and lifestyle factors to forecast disease probabilities.
It is particularly effective for diseases with clear progression, such as cancer and heart attacks.
The study in *Nature* underscores its potential for personalized healthcare and early intervention, estimating health trajectories for up to 20 years.

1 / 2