Using AI/ML to automate clinical coding within the NHS
Research type
Research Study
Full title
Using AI/ML to automate clinical coding within the NHS
IRAS ID
366256
Contact name
Emily O'Brien
Contact email
Sponsor organisation
University Hospitals Plymouth NHS Trust
Clinicaltrials.gov Identifier
6840, University of Plymouth; Pending, Clinical trials.gov
Duration of Study in the UK
5 years, 11 months, 30 days
Research summary
This is a part time PhD project to automate clinical coding. Clinical Coding is the process of assigning structured medical codes to a patient’s notes to categorise the procedure/care they received, any diagnosis that was given and any comorbidities the patient had at the time of their hospital episode. This is currently done manually within the NHS by a team of trained clinical coders. However there is a shortage of qualified clinical coders in the UK and the backlogs of cases requiring coding are growing. The research project’s aim is to train machine learning or artificial intelligence models using existing NHS data to automate the classification of OPCS4 procedure codes and ICD10 primary diagnosis and comorbidity codes using patient’s hospital notes. These outputs will then be reviewed by the clinical coding teams to aid reducing the backlogs.
REC name
London - Stanmore Research Ethics Committee
REC reference
26/LO/0313
Date of REC Opinion
7 May 2026
REC opinion
Further Information Favourable Opinion