The aim of our group is to facilitate the use of healthcare free-text (clinical notes, reports, letters) in medical research. Our research is founded on a number of key projects.
Whole population automated reading of brain imaging reports in linked electronic health records (WARBLER)
We will analyse anonymised data from brain scan reports to study the effect of brain appearance on future health.
Turing Project on Large-Scale and Robust Text Mining of Electronic Healthcare Records
We aim to compare machine and deep learning algorithms to rule-based methods for mining Scottish brain imaging reports.
HDR UK National Text Analytics Implementation Project
We will apply NLP methods to generate a more comprehensive picture of patients and their health.
HDR UK Human Phenome Project on Deriving and Applying Health-Related Phenotypes at Scale
We aim to extend the phenotyping potential of disease status algorithms with linked health data and unstructured data from electronic medical records.
MRC Mental Health Data Pathfinder Project
We focus on an exemplar study on linking routine neuroimaging assessments and electronic health records to identify the relationships between common physical and mental health conditions.