Scientists have developed a new artificial intelligence tool that can predict people’s risk of more than 1,000 diseases including cancer, over a decade into the future.
Experts from the European Molecular Biology Laboratory (EMBL), the German Cancer Research Centre and the University of Copenhagen hope the new technology will help spot high-risk patients to prevent disease years ahead of time.
The generative AI tool works by assessing the likelihood of whether—and when—someone may develop potentially life-threatening diseases including cancer, heart disease and diabetes.
Writing in the journal Nature, Tomas Fitzgerald, an expert in molecular biology and study co-author, said: ‘Medical events often follow predictable patterns.
‘Our AI model learns those patterns and can forecast future health outcomes.’
Called Delphi-2M, the model has been trained to find patterns in anonymous medical records, looking for ‘medical events’ in a patient’s history alongside lifestyle factors such as smoking, alcohol consumption and being overweight, to predict what might happen up to 20 years into the future.
The model was trained and tested on patient data from 400,000 patients enrolled in the UK Biobank study and 1.9 million people on the Danish national patient registry.
Researchers hope patients will be able to benefit from the breakthrough within the next few years, which will tell them their likelihood of developing a disease over time—similar to forecasting a 70 per cent chance of rain at the weekend.
The breakthrough AI model uses patient history, ‘medical events’ and lifestyle factors such as smoking, alcohol consumption and weight to create forecasts for the next decade and beyond
Professor Ewan Birney, executive director of the EMBL, explained how the new technology could work in GP surgeries to help high-risk patients get the care they need, before progressive diseases take hold.
‘You walk into the doctor’s surgery and the clinician is very used to using these tools, and they’re able to say: “Here’s four major risks that are in your future and here’s two things you could do to really change that.”
He continued: ‘I suspect everyone will be told to lose weight, and if you smoke you will be told to stop smoking—and that will be in your data so that advice isn’t going to change remarkably—but for some diseases I think there will be some very specific things.
‘That’s the future we want to create.’
Whilst the model is best at forecasting disease that have a clear progression—like type 2 diabetes and heart attacks—rather than more random events like infection, the researchers are hopeful models like this could help anticipate healthcare demand at scale.
Professor Moritz Gerstung, expert in computational cancer biology at the German Cancer Research Centre, said: ‘This is the beginning of a new way to understand human health and disease progression.’
He added that the AI model could also help inform disease-screening programmes, preempting demand—such as how many people in London are likely to suffer a heart attack in the next year, to help allocate and plan resources.
But other experts have urged caution on the findings, saying the Delphi-2M is not without significant limitations.
Professor Justin Stebbing, a consulting oncologist at Anglia Ruskin University, warned against construing results as direct causal relationships.
He said: ‘The model reproduces biases found in its training data, including healthy volunteer and selection biases.’
Professor Peter Bannister, healthcare expert and fellow at the Institution of Engineering and Technology added: ‘This is still a long way from improved healthcare as the authors acknowledge that both datasets are biased in terms of age, ethnicity and current healthcare outcomes.
‘The immediate challenge for healthcare is to ensure there is a sufficient digital infrastructure and skills base for everyone, regardless of socioeconomic background, so the currently available technologies can be offered to those who most need improved access to treatments.’
Therefore, whilst the scientists behind the Delphi-2M are confident that their model can ‘do all diseases at once and over a long time period,’ setting it apart from existing single disease models, other experts in the field remain hopeful yet cautious.
Professor Gustavo Sudre, an expert in neuroimaging and AI at King’s College London, said: ‘While the current version relies solely on anonymised clinical records, it’s encouraging to see that the model architecture has been deliberately designed to accommodate richer data types, such as biomarkers, imaging and even genomics.
‘With these future integrations, the Delphi platform is well-positioned to evolve into a truly multimodal, precision-medicine tool.’
It comes as One Cancer Voice has warned that cancer cases will soar to record highs by 2040, with one patient told they have the devastating disease every two minutes.
The stark forecast shows the most common cancers – breast, prostate and lung – will hit unprecedented levels, while more than 63,000 cases are predicted in children and young people.
This article was originally published by a www.dailymail.co.uk . Read the Original article here. .