A recent study has highlighted the potential of artificial intelligence (AI) in predicting future health conditions for patients, providing valuable support to doctors in monitoring and diagnosing ailments. The AI tool, named Foresight, is part of the same AI family as ChatGPT but has been specifically trained on data from NHS electronic records.
Developed by researchers from King’s College London (KCL), University College London (UCL), King’s College Hospital NHS Foundation Trust, and Guy’s and St Thomas’ NHS Foundation Trust, Foresight underwent training using data from over 811,000 patients from NHS trusts in London as well as a publicly available dataset from the US.
The study, published in The Lancet Digital Health, revealed promising results. Foresight accurately identified potential health conditions a patient might experience next with rates of 68% and 76% using NHS hospital data, and 88% accuracy with US data.
Zeljko Kraljevic, a research fellow at KCL specializing in health informatics and biostatistics, emphasized the tool’s precision in predicting patient health trajectories, highlighting its potential to aid clinical decision-making and inform research.
Senior author Richard Dobson, a professor of medical informatics at KCL and UCL, underscored the importance of using appropriate data to train AI models effectively in healthcare settings. He expressed optimism about the role of AI in healthcare and emphasized the need for collaborative efforts to develop tools that support healthcare systems in serving patients better.
Looking ahead, the research team aims to involve more hospitals in the development of Foresight 2, which is envisioned to be an even more accurate language model.
Discussion about this post