Research Engineer at the IT Innovation Centre, as part of the ECS Centre for Health Technologies.
From a background of Astrophysics (PhD - University of St. Andrews & Flatiron Institute, New York), Chris specialises in developing trustworthy and explainable machine learning algorithms. Closely aligned with University Hospital Southampton, Chris works directly with both clinicians and patients to develop tools that can improve healthcare.
- mySmartCOPD (NHSX £1.5 million): mySmartCOPD aims to develop a personalised AI-enhanced platform (building on mymhealth's app myCOPD) which informs users (who have Chronic Obstructive Pulmonary Disease - COPD) if they are likely to exacerbate in the near-future. By providing an early warning, people living with COPD have an opportunity to take action to prevent an exacerbation or lessen its severity. The project aims to empower users by having more control over their condition, enabling medication to be used more appropriately, and enable NHS resources to be used more efficiently. Chris leads the development and visualisations of the AI component for the project, which is to be used as part of a large clinical trial.
- AISmartCorps (InnovateUK - £626,367): Financial service companies acquire huge volumes of data which require checking for, and resolving, errors. This is a time consuming and often overly manual set of tasks that put staff under significant pressures which can lead to resourcing challenges and increased business risks. In addition, as data volumes increase, companies are finding it harder and harder to effectively check every line of data and have to choose where to focus their efforts to minimise risks. AISmartCorps aims to develop and deploy state-of-the-art AI algorithms for automating error detection and recommendations for appropriate resolutions, thus reducing this significant burden on staff and allowing them to focus on more valuable and skillful tasks for the company. Chris contributes to the development of AI components for the project.
- COdesigning Trustworthy Autonomous Diabetes Systems (TASHUB - £100,000): COTADS aims to develop trustworthy algorithms for people with type-1 diabetes during life transitions (i.e. as they transition from paediatric to adult care units). The project involves people with type-1 diabetes, clinicians, and data scientists who together explore where machine learning can be effective, and identify how best to communicate AI to better acceptance. Chris leads the development and visualisations of the AI component for the project.
- TriagED Decision support algorithms for emergency departments (Alan Turing Institute - £100,000): Emergency departments (EDs) in the UK are under sustained and increasing pressure. A key aspect of EDs is the ability to 'triage' patients (i.e. assess the immediate risk to the individual), however, under significant pressure it can be difficult to comprehensive assess every patient in a timely manner. TriagED investigated how machine learning algorithms could be used as decision support tools, to help improve the safety and efficiency within UHS's ED. Chris led a research piece which developed a model that predicted patient hospital admission ahead-of-time (i.e. from the ED), and explored the impact of the COVID-19 pandemic on patient admissions.
- CO@home (NHS): COVID-19 has placed unprecedented demands on hospitals. A clinical service, COVID Oximetry @home (HSJ Patient Safety Award 2021) was launched in November 2020 to support remote monitoring of COVID-19 patients in the community. Remote monitoring through CO@h aims to identify early patient deterioration and provide timely escalation for cases of silent hypoxia, while reducing the burden on secondary care. Chris contributed to the data analysis evaluating the efficacy of the intervention.