A new AI model is being trained using millions of NHS patients’ data in an exciting project that is a world first. This pilot, which is being run by a research team at King’s College London and UCL, has the aim of seeing whether the model, trained on anonymised data, will eventually be able to transform the patient care experience and flag up opportunities for early interventions to secure better outcomes and save lives.

Harnessing the power of generative AI

The generative model, which is called Foresight, uses data and predictive analytics to assess previous medical events and what is likely to happen next. In this way, it is similar to other generative AI models such as ChatGPT, which uses sentences to predict the next word, and uses the sort of data a data analysis company such as https://shepper.com would use.

The power of national data

Foresight is being collected on de-identified data that is collected as a matter of routine, including Covid-19 vaccination rates and hospital admissions, to predict the types of health outcomes that patient groups can expect, including events such as a new diagnosis, heart attack, or hospitalisation. By predicting these events before they occur, the system could support doctors to make early interventions and move toward a more preventative model of healthcare on a large scale.

How the work is being developed

The study is being run on a platform within NHS England. This allows data to be used from 57 million people who live in the UK and use the NHS, with the data strictly controlled under NHS guidelines. The technology will only be as good as the data it is trained on, which is why this vast pool of anonymous data is so important. By using data on a national scale, the full picture can be assessed, particularly where rare diseases and minority groups are concerned. These demographics tend to be left out of research.

This tool could well revolutionise the NHS and its ability to prevent and treat disease in future.