AI models are increasingly being used in healthcare not just to identify diseases that a person already has, but to assess how their health could change and which illnesses they may face in the future, according to a report by Longevitty.ai.
The report points to a broader shift in medical research, with scientists combining health records, blood proteins, and information about individual organs to understand how a person’s health changes over time.
Also Read: India Leads the World in Healthcare AI Adoption at 85%, Surpassing the US & UK
AI Models Look Beyond a Single Health Test
The report says researchers are shifting from asking whether a patient has a particular condition to examining the direction their health is heading.
Early identification remains a major challenge. The US Centres for Disease Control and Prevention estimates that 115.2 million American adults, more than two in five, have prediabetes, while eight in 10 do not know they have it.
One 2024 study used AI to analyse proteins from 45,441 people in the UK Biobank. Researchers found that 204 proteins could collectively estimate a person’s age. Those whose proteins indicated an older biological age faced a higher risk of 18 major diseases, including heart, kidney and lung disease, diabetes and cancer.
Stanford researchers, meanwhile, analysed 44,498 people by examining organs individually. Their study found the brain and immune system were among the strongest predictors of how long and how healthily people might live. Organ ages also varied with lifestyle and medicines.
Also Read: ‘Superhuman’ AI ECG Tool Spots Heart Disease in Under 2 Seconds; Speeds Up Diagnosis
Delphi-2M Forecasts More Than 1,000 Diseases
Another development highlighted is Delphi-2M, an AI model developed by European scientists to forecast possible future diseases.
The report compares its approach with a phone keyboard predicting the next word. Delphi-2M instead attempts to predict the next disease a person may develop and when it could occur.
It was trained on data from 400,000 people in the UK and tested, without modification, on 1.9 million people in Denmark. The model can predict more than 1,000 diseases and performed about as well as tools focused on individual diseases.
“81 per cent of doctors surveyed this year use AI in their work, compared with 38 per cent in 2023. However, doctors also expressed concern about patients using AI without medical supervision,” said Max Chopra, founder and CEO of Longevitty.ai.
For the healthcare market, the direction is clear: AI models are increasingly used to connect different parts of a person’s health history rather than treating each test as a separate event. The report said keeping health records together could become increasingly important as these systems develop.
The next phase of healthcare AI is increasingly about connecting scattered health signals over time. How reliably those systems translate prediction into useful medical decisions will remain an important area to watch.
Samarjit Kaur is a journalist and communications professional covering technology & emerging digital trends. With a focus on clarity and context, she reports on developments shaping industries and governance. When not reporting, she chooses to plug-in and relax on her playlists and plan her next bucket-list trips!
