Helping the NHS cut waiting lists
With East Suffolk and North Essex NHS Foundation Trust (ESNEFT), we used machine learning to predict how long operations will take, so that theatre lists can be planned more accurately.
The challenge
Waiting lists for planned operations have grown a great deal since the pandemic. Operating theatres are expensive and in short supply, so using them well matters. Theatre lists are usually planned from the surgeon's own estimate of how long each operation will take. When the estimates are wrong, theatres sit idle or lists overrun and operations are cancelled.
What we did
In a Knowledge Transfer Partnership with ESNEFT, we took the Trust's records of past orthopaedic operations and taught a computer program to predict how long a new operation will take, using only what is known when the list is planned, such as the type of operation, the surgical team and the theatre.
The predictions were clearly more accurate than the surgeons' own estimates. They are a first step towards a digital twin of the hospital, a computer model in which changes to scheduling can be tried out before they are made. The project won Best Partnership: Public Sector at the University of Essex Innovation Awards 2024.