Lay Description
This project introduces a secure, open-source tool designed to address a critical data gap in the care of inflammatory bowel disease (IBD). By leveraging natural language processing (NLP), the project aims to automatically extract and structure key patient health information, such as comorbid conditions and patient-reported symptoms, from unstructured clinical notes that are often overlooked in standard electronic health records (EHRs). The system ensures patient privacy through de-identification, and its accuracy will be clinically validated. The primary goal is to create a rich, research-ready dataset that, when combined with genetic information, will accelerate discoveries of genotype-phenotype associations, supporting more personalised and effective treatments for IBD patients across the NHS.