Using AI to Codify Patient-Reported Data for Improved Clinical Documentation

This project explores how machine learning (ML) and natural language processing (NLP) can be used to understand and codify patient-reported free-text data for hospital admission. It focuses on converting free-text patient inputs into structured clinical codes such as ICD-10 and SNOMED-CT, helping to improve documentation, data quality, and hospital workflows.

In collaboration with industry partner EpiSoft, this research project will build and evaluate automated classification models using real-world data. The project also aims to assess the feasibility and acceptability of these tools for future clinical integration.

In the later stages, the project will explore real-world deployment of the NLP system into hospital workflows, including potential applications such as automated eligibility checks, personalised patient education, and integration with electronic medical records (EMRs).

Project Lead

Zhaoyuan (Derek) He (PhD Candidate)

Project Team

Prof John Grundy, Dr Chetan Arora