St Bartholomew's Hospital used machine learning to analyse thousands of outpatient letters within the congenital heart disease service and detect whether the correct action had been taken by either the patient or the clinical team.

Shane Cashin, the information business partner at St Bartholomew's, came up with the idea and worked alongside representatives from our clinical services, Queen Mary University of London and Andrew Houston from the Barts Life Sciences programme.

He said: 'If used correctly, AI has the potential to ease workloads for our admin teams, reduce waits for patients and improve safety.'

The machine was taught to identify key phrases within patient letters and check them against electronic health records.

Over six months the system read 1,500 letters and identified 16 so-called ‘high risk' cases where action was needed to prevent delays in care or even serious harm.