News

Knowledge Transfer Partnership innovation success

Knowledge Transfer Partnership innovation success

A joint project with Anglia Ruskin University to see how artificial intelligence (AI) and machine learning can improve the way we extract patient data to drive better patient safety and outcomes, has been rated as Outstanding by Innovate UK, the UK’s national innovations agency.

The project, which began in February 2019, was funded as a Knowledge Transfer Partnership (KTP) by Innovate UK. The KTP brought together our clinical and tech experts with academics from the University. Our colleagues included: Amirali Shirazibeheshti, Dr Tarek Radwan, Dr Alireza Ettefaghian and Hasib Aftab.

This is the second KTP fund we have received from Innovate UK, and the second to be rated as ‘Outstanding’. The first fund was in 2015, leading to the development of our in-house data-management tool EZ Analytics, known internally as EZA. This second project added an AI element to EZA.

The team deployed an AI module in EZA, focusing on a target group of patients, in this case older people prescribed anticholinergic drugs. This module optimised the identification of patients in this group with significant risk-factors, supporting our clinical teams to de-prescribe where appropriate. Extracting the data in a way that individuals or existing searches could not do, this new AI-driven approach will reduce the risk of serious illness or death.

“I would like to thank everyone involved for their efforts in making this project a success. We are proud to be one of the very few organisations in the country to have been granted two KTP funds and two rare “Outstanding” grades. This shows our commitment to constant innovation, to always going above and beyond to maintain our position as a leading provider of digitally driven primary care in England. The next step now is to embed this AI into EZA consistently across our practices to drive better outcomes for our patients.” Dr Tarek Radwan, Primary Care Director

The team will now be exploring the options for applying this level of sophisticated data interrogation to cover other areas where it is possible to calculate risks and suggest improvement plans.

The outcome of this work has been published in the International Journal of Medical and Health Sciences.