Research
My research is centered around automating information extraction from large volumes of veterinary clinical notes using text mining methods that include machine learning and AI. We are evaluating the use of transformers (neural network language models that capture subtle details of word context) to help automate capturing meaning from messy clinical notes. We are also evaluating large language models (like ChatGPT) for their utility in extracting data from clinical text. There has never been a more exciting time to work in this field.
Text-mining clinical records
Every day, veterinary surgeons across the country write notes about their patients into electronic health records using a variety of practice management systems (PMS). These notes provide a potential resource for developing our understanding of companion animal health but, additionaly, contain the most timely information about changes of clinical syndromes that might occur as part of a disease outbreak. The Small Animal Surveillance Network (SAVSNET) has designed a system that can be implemented in the PMS to allow these clinical notes to be sent immediately to the University of Liverpool for analysis.
To date SAVSNET is collecting data from over 400 veterinary premises from across the UK and the data sent includes the free text that the veterinary surgeon wrote during the patient visit. This text is typed free hand and contains a variety of spelling, grammar and punctuation variations that challenge modern text-mining software but when the software approaches are developed in close collaboration with clinicians conversant in veterinary language, this potent partnerships provides an efficient environment to improve the accuracy of text mining.
SAVSNET has made good use of simple test mining tools such as key word recognition for words like 'Tick' or 'Diarrhoea' but we are also using more complex pattern matching tools to filter records for specific features. For instance in records where the word 'vomiting' is mentioned, this may actually allude to vomiting being absent (e.g. if the vet wrote 'no vomiting seen') to correct for this a more complex filter (a regular expression) is used to screen records. Keyword and regular expression searches are incorporated into our text mining suite (SAVSNET Datalab) which allows students to develop and test regular expressions, select a filtered set of records and then read and manually annotate them for relevance to their given project prior to collating the text-mining findings with other patient data (breed, species, sex, age, geographical data). This has been used for numerous undergraduate and post-graduate projects and publications.
We are in the process of implementing various machine-learning methods into our text mining that include neural word embedding, Neural network analysis and topic modelling.
Research grants
LEVERAGING HEALTH RECORDS AND MACHINE LEARNING TO EXPAND THE LARGEST PET ANIMAL CANCER REGISTRY
PET PLAN CHARITABLE TRUST (UK)
April 2023 - March 2025
Understanding and improving behaviour consults in veterinary practice
DOGS TRUST (UK)
September 2022 - August 2026
Targeting health informatics research to promote canine welfare research, surveillance and education
BRITISH SMALL ANIMAL VETERINARY ASSOCIATION (UK)
September 2018 - April 2019
Small Animal Veterinary Surveillance Network
BIOTECHNOLOGY & BIOLOGICAL SCIENCE RESEARCH COUNCIL
August 2016 - April 2020
Research and surveillance of vaccine preventable diseases and vaccination uptake in companion animals in the UK
PET PLAN CHARITABLE TRUST (UK)
April 2021 - March 2024
Can the SAVSNET informatics systems be used to improve signal identification and reporting of adverse drug reactions?
DEPARTMENT FOR ENVIRONMENT, FOOD AND RURAL AFFAIRS (UK)
February 2019 - October 2022
Working with commercial companies for the sustainability of health informatics research in the veterinary sector.
BIOTECHNOLOGY & BIOLOGICAL SCIENCE RESEARCH COUNCIL
June 2019 - November 2019
National surveillance of antimicrobial consumption and resistance in companion pet animals.
DEPARTMENT FOR ENVIRONMENT, FOOD AND RURAL AFFAIRS (UK)
June 2014 - March 2019
Serum citrulline as a novel marker of chronic enteropathy in dogs
ZOETIS (UK)
December 2013 - November 2015
Metabolic Profiling of Dogs with Portovascular Anomalies.
PET PLAN CHARITABLE TRUST (UK)
August 2003 - July 2004
Metabolomics: a post-genomic tool to predict outcomes in canine lymphoma.
BRITISH SMALL ANIMAL VETERINARY ASSOCIATION (UK)
March 2005 - September 2008
Influences of glucocorticoid receptor activation on gastric epithelial cell migration.
PET PLAN CHARITABLE TRUST (UK)
November 2008 - April 2010