Artificial intelligence can predict people's health problems over a decade into the future, say scientists. The technology has learned to spot patterns in people's medical records to calculate their risk of more than 1,000 diseases. The researchers say it is like a weather forecast that anticipates a 70% chance of rain – but for human health.
The vision is to use the AI model to spot high-risk patients to prevent diseases and help hospitals understand demand in their area years ahead of time. The model – called Delphi-2M - uses similar technology to well-known AI chatbots like ChatGPT. Delphi-2M has been trained to find patterns in anonymous medical records, helping predict health risks over time.
According to Prof Ewan Birney, the interim executive director of the European Molecular Biology Laboratory, just like weather predictions, we can provide healthcare forecasts. This is the first time we've achieved this across multiple diseases. The AI tool, which initially used UK data from over 400,000 individuals, has also shown promise with medical records from Denmark.
The AI predictions could lead to early interventions, like prescribing medications or recommending lifestyle changes to those at high risk. However, while the model shows great potential, it's not yet ready for clinical application, necessitating further research and refinement to mitigate potential biases.
Prof Birney emphasizes the need for thorough testing and regulation, anticipating that AI's integration into healthcare will take time, much like genomics has over the past decade. This collaborative research, involving esteemed institutions, marks a significant advancement toward responsible predictive modeling in the medical field.
