Dom Clarke - WWL NHS Trust

Former MSc student Dom Clarke shares his minidiary about his project at Wrightington, Wigan and Leigh NHS Trust

Before the placement began, I collected local weather data (from a public source) so that it could be implemented alongside the Trust’s internal data in a model predicting admissions.

The first week started with an induction into the team and with familiarising myself with the software and how to extract the Trust’s data. As the week progressed, I was introduced to different teams whilst also having the opportunity to go to off-site meetings to learn about current issues. Once I had settled in, I started using R to inspect the data and gain insight. By the end of the week I was able to produce a basic admission model.

In the following weeks, I attended more meetings in A&E and I looked further into improving the model, by observing a variety of potential covariates, some that worked, some that did not. I was able to develop a good benchmark admission model for each day and then I aimed for greater detail by looking at hourly admissions. I was able to build on the accuracy of existing models and my models have been handed over to the NHS for further use.

This is my first real experience of work and everyone has made me feel very welcome, with individuals from different backgrounds available to help whenever necessary. It has been a great decision to choose to work in data science, it has really justified my choice of study. I have had the opportunity to work alone and with teams, where every day, I have new enjoyable challenges and moments to make an impact.

The whole process was great, everyone at the office welcomed me with open arms and helped me with anything I wanted. I was provided total flexibility to use any methods I desired, it was really MY project. I had occasional meetings to manage progress with my boss but they were only to check up. In the end, the work performed, proved to be successful where improvement was made upon the current models. I was gifted with the reward of seeing the predictions implemented into the QlikView App and also seeing the work I had done presented and discussed in front of executives from trusts around the north west. I learnt lots from the placement, from presenting my findings to those without domain knowledge, to new applications of machine learning methods. Overall, it has been a perfect placement and provided the essential skills for a full time position within the data science area. In fact, I started as a full time data scientist today at Peak BI, a data science consultancy start up based in Manchester.