Consultant Gastroenterologist and Data Scientist
Dr Matt Stammers is a Consultant Gastroenterologist at University Hospital Southampton and holder of a Research Leaders Programme (RLP) fellowship.
His research focuses on how we can better use clinical data, particularly to find patients for research cohorts, predict their outcomes and develop useful algorithms to help streamline clinical care.
Algorithm development and validation
Matt first became involved in research in 2016, working on algorithm development to predict which patients would be most likely to benefit from having a colonoscopy. This was when he first started programming. Over the years, he’s built on this experience, becoming more involved in software development and research using clinical data.
During the COVID-19 pandemic, Matt was employed by the Trust to use data modelling to predict things like the number of beds required and the number of deaths due to the disease. He then collaborated with King’s College to develop an algorithm to predict Covid-19 mortality and worked with Dr Tristan Clark locally to convincingly demonstrate that point-of-care testing substantially reduces hospital acquired infection rates.
He has since developed an algorithm that can estimate invasive cancer depth at endoscopy, proved the benefit of local inflammatory bowel disease service re-design and validated several algorithms.
Future direction of travel
For his Doctor of Medicine (DM) degree, Matt is looking at ways to identify patients using unstructured clinical information contained in clinical free text. The use of a combination of structured, semi-structured and unstructured clinical and administrative information to develop useful algorithms continues to be his main focus of research interest.
“In the future, clinical algorithms will help us decide what to do and, in some cases, make clinical and reporting decisions for us,” he explains. “We are starting to see this in some areas like radiology already – in the future this reality will affect all clinical specialties.
“My goal is to help shape that future by getting very engaged in the development of the underlying technology at an early stage.”
Developing a research data sciences team
“I’m acutely aware I’m just one person, and I’ve only got so many hours a week. The only way I can scale the mission of using technology to help clinicians and researchers is by becoming a leader,” says Matt. “That’s going to require training, experience and networking - these are all things the programme should deliver.”
Matt has started building a team of people at the Trust who work with clinical data, and he hopes to use his fellowship to learn to be a better leader by investing in others. When he started in 2019, it was just him and a couple of others. Now there are multiple fellows and a core team of developers and analysts, with more being recruited soon.
His long-term goal for the team is to make UHS a national leader for data science, as part of the new Southampton Emerging Therapies and Technologies (SETT) Centre and NIHR Biomedical Research Centre data science theme.