A new artificial intelligence (AI) tool is capable of predicting an individual’s risk of more than 1,000 diseases, in some instances projecting outcomes decades into the future, according to a recent study published on Wednesday.
The study, published in the scientific journal Nature, revealed that a modified language model called Delphi-2M uses health records and lifestyle factors to estimate the likelihood of conditions such as cancer, skin diseases, and immune disorders up to 20 years ahead.
Delphi-2M is considered capable of modeling multiple diseases, potentially allowing clinicians to identify high-risk individuals and roll out preventive measures sooner.
While previous AI tools predicted the risk of individual conditions like cancer or cardiovascular disease, the modified model was designed to forecast a person’s likelihood of developing 1,258 diseases based on their medical history, taking into account a person’s age, sex, body mass index, and lifestyle factors, including smoking and alcohol consumption.
For the majority of diseases, Delphi-2M outperformed or matched the accuracy of current single-disease risk models.
The tool was developed and validated using anonymized data from 400,000 participants in the UK Biobank and 1.9 million patients in Denmark’s national patient registry.