In 2024, researchers used AI to study proteins from 45,441 people in the UK Biobank. They found that 204 proteins, taken together, could estimate a person’s age. More importantly, people whose proteins appeared older than their actual age had a higher risk of 18 important diseases, including heart, kidney and lung disease, diabetes and cancer.
New Delhi: Artificial intelligence (AI) models are increasingly being used to study how diseases may develop in the future, pointing to a shift in healthcare from identifying existing conditions to predicting possible health risks, according to a report by Longevitty.ai.
The report highlighted that scientists are using AI to look beyond individual health tests and understand how a person’s body may be changing over time. Instead of focusing only on whether a person currently has a particular condition, researchers are studying whether health data can indicate the direction in which a person’s health is moving.
It stated, “Scientists are using Super Intelligence to look beyond a single test and ask a more useful question: which way is your body heading, and how fast?”
For millions of people, early identification of health risks remains important.
According to the US Centres for Disease Control and Prevention, 115.2 million American adults, more than two in five, have prediabetes, and eight in ten of them do not know it.
The report said researchers are now using AI to study blood proteins and biological ageing as part of this broader approach.
In 2024, researchers used AI to study proteins from 45,441 people in the UK Biobank. They found that 204 proteins, taken together, could estimate a person’s age. More importantly, people whose proteins appeared older than their actual age had a higher risk of 18 important diseases, including heart, kidney and lung disease, diabetes and cancer.
The report also highlighted research by scientists at Stanford, who studied the body one organ at a time. After analysing 44,498 people, they found that the brain and immune system were among the strongest predictors of how long, and how healthily, someone might live.
The researchers also found that the ages of these organs changed depending on how people lived and what medicines they took.
Another development highlighted in the report is Delphi-2M, an AI model developed by scientists in Europe to forecast possible future diseases.
The model works in a way similar to how a phone keyboard predicts the next word a person is likely to type. Instead of predicting a word, Delphi-2M tries to predict the next disease a person may develop and when it might happen.
The model was trained using data from 400,000 people in the UK and was then tested on 1.9 million people in Denmark without being changed. According to the report, it can predict more than 1,000 diseases.
The report said Delphi-2M performed about as well as tools that are designed to look at one disease at a time.
Max Chopra, Founder and CEO of Longevitty.ai, said “Super Intelligence is already becoming part of everyday medical work. In a survey this year, 81 per cent of doctors said they use it in their work, compared with 38 per cent in 2023. But those same doctors said they worry about patients using Super Intelligence without a doctor helping them”.
He added, “Someday, one Super Intelligence may be able to look at your health history and your blood together. To me, that makes another point especially important: your health records are part of your story, and it’s worth keeping them together so that story doesn’t get lost from one test to the next.”
The report noted that this shift is already underway, with researchers using blood proteins, organ-level information and health records to study ageing and disease risk.
ANI
Leave a Reply