Claire Coffey (she/her) PhD Candidate
University of Cambridge (HDRUK-Turing Wellcome PhD Programme)
Contact:
Email: cic31@cam.ac.uk
Twitter: https://twitter.com/claireicoffey
Website: https://clairecoffey.github.io/
What does your job actually entail?
My workday always includes an endless supply of snacks, and usually includes an endless supply of data analyses to run. I’m currently finishing up a project on using algorithmic fairness metrics to quantify the fairness of disease risk prediction algorithms, and I’m really excited to be able to share this soon! Most of my work is done independently, but I also love collaborating with others. Therefore, meetings with the other organisers of the Statistical Methods theme here at DSxHE; lab meetings with Inouye Lab; or cohort events with the other students on the HDRUK-Turing Wellcome PhD programme are all a welcome disruption to my independent work.
When not at work you can be found…
Tending to my plants (I especially love growing my own food); singing - mostly to myself - with the occasional out-of-the-house performance; and partaking in too many activities including kung fu, rowing, dancing, and yoga. I also love being outdoors and preferably up a mountain - but these are hard to come by in Cambridge!
Why did you join the DSxHE community?
To meet like-minded individuals with a diverse range of experiences all interested in the same goal! My research is multi-disciplinary, yet largely independent. This independence, in combination with a worldwide pandemic, meant that I found it difficult to find others as excited about these topics as myself. DSxHE came at the perfect time to combat this, facilitating the creation of a community full of people from whom I can learn so much.
What’s your interest in data science and/or health equity?
My research is focused on algorithmic fairness in medical prediction algorithms. I’m interested in quantifying the fairness of algorithms that predict individuals’ disease risk; and using machine learning and artificial intelligence to develop new methods to predict disease risk with fairness at their core.
What’s a topic in data science/health equity that you know/care a lot about - why is it important/interesting, tell us about it!
It was during my Master’s in computer science that I first learnt about AI fairness - or rather - the lack of fairness in many AI applications. I was shocked by the extent of the discrimination seen in real-world applications of AI - with examples spanning from criminal justice to education, finance, and healthcare. From then, I felt like it was my responsibility to use my technical background to help improve these issues. I chose to focus on algorithms in healthcare due to their huge potential for positive impact. Yet, they also have a great potential for inequitable impact. The more these algorithms are used, the more imperative it is that their fairness is considered and improved. Which led me to my research today!
What is your favourite thing to do in the summertime?
Exploring nature in my campervan! Especially discovering new places and finding spots for wild swimming.
What is the best advice you ever received?
My Dad always tells me to not take myself too seriously, and I am often coming back to this little piece of advice. It’s easy to get carried away with the seriousness of the world - and of course there are times to be serious - but life’s a lot more fun when you can laugh at yourself.
What dish do you cook best?
I grew up in Birmingham, where I was spoiled by having the “balti triangle” right on my doorstep. Cambridge’s curry scene doesn’t even come close, so I’m always attempting to recreate the taste of home by working on my curry recipes! Even better if they include home-grown vegetables.