Dr Matthew Watson

Post Doctoral Research Associate in the Department of Computer Science

What do you do?

I am a Postdoctoral Research Associate in the Department of Computer Science at Durham University, researching the application of machine learning to healthcare. My research interests lie in the area of explainable machine learning and how this can be used to improve the quality of, and trust in, machine learning models. I received my undergraduate degree in Computer Science from Durham University in 2019, and my PhD also from Durham in 2023.

How are you involved in this area of science? 

I create new machine learning techniques that are robust, trustworthy and explainable, and apply these to novel problems in healthcare. This involves working with numerous clinical partners to identify clinical need and propose possible solutions. For example, in collaboration with Evergreen Life and University College London, I have created machine learning models that can predict chemotherapy patients that are at high-risk of complications - this allows high-risk patients to have more increased monitoring, while low-risk patients do not need as many trips to the hospital. I have also worked with the Northern Care Alliance NHS Foundation Trust to use advanced text modelling techniques to improve the monitoring of patients for significant clinical deteriorations. This significantly reduces the number of unnecessary alerts when compared to current techniques, meaning clinicians could better focus their time.

What do you love about this topic?

From a Computer Science perspective, working with healthcare data introduces many challenges that we must address: explainability, trustworthiness and model bias are all interesting areas that I love finding solutions to. I also love working with clinicians, each of which is an expert in their field; I find the challenge of bridging that gap between my computer science and research skills, and their clinical expertise, extremely rewarding.

How does this work deliver real-world impact?

All of my healthcare projects have the end goal of being deployed in hospitals, even if they take a long time to get there! This means there is always a direct link to patient benefit: be that through better outcomes, higher quality and more personalised care, or reduced cost to the NHS. We also always include patient groups in our research, to understand their thoughts on how our work might impact them - not only does this improve the quality of our projects, it also means we can clearly understand how we can improve their care experiences!

Find out more:

• Learn more about how researchers developed AI to personalise chemotherapy care.