This past summer was the fourth year of our annual internship programme here at Health Data Insight. Five interns from across the country joined us for 10 weeks over the summer to work on projects with healthcare data.

Read on to hear about what each of them has been working on and their favourite things about the internship programme.

 

Tobby visualised data

Tobby

This is “Tobby” by Kim Whittlestone on Vimeo, the home for high quality videos and the people who love them.

Tobby spent his internship creating visualisations of cancer data from Public Health’s England ‘Get Data Out’ programme.  The Get Data Out programme produces open, anonymous statistics about the incidence, survival, diagnosis and treatment of different types of cancers.  Tobby helped to transform the plain data tables into easy to use interactive graphics to help unlock the most important insights for a public audience.  After some final adjusting, the visualisations will be available to see next year.

“It’s been a really good experience. Even though it’s only been 7 weeks I have gotten a lot done and have a lot to say for myself and for my work here. I have made something that is almost ready to go online and I set the stage for future developments in the future”
  

 

Andrew built an app

Andrew

This is “Andrew” by Kim Whittlestone on Vimeo, the home for high quality videos and the people who love them.

 

Over the summer Andrew built an application that compares how different NHS hospital trusts care for cancer patients. Different hospital trusts care for different sorts of people, and so comparing how care is given between trusts is difficult. But comparing trusts is useful to help commissioners decide how best to deliver cancer services. Andrew’s app will help trusts make these decisions and will be available to the NHS soon.

“It’s been fun working at HDI because the company has a real sense of purpose to improve patient care”

 

Roan automated a process

Roan Final

This is “Roan Final” by Kim Whittlestone on Vimeo, the home for high quality videos and the people who love them.

 

Roan spent her time in Cambridge working to automate an existing process for extracting data.  When external researchers apply to use data from Public Health England’s National Cancer Registration and Analysis Service (NCRAS), NCRAS analysts often spend a lot of time finding and extracting the data. Over the summer Roan built a tool to make this process quicker and easier.

“There has been a lot of skills which I’ve picked up since coming here which have been really valuable and a lot of fun. I was new to R when I got here and it’s been very valuable. And seeing how I can make something that will help and make a difference has been really interesting and enjoyable”

 

Edward tested simulated data

Edward Pearce

This is “Edward Pearce” by Kim Whittlestone on Vimeo, the home for high quality videos and the people who love them.

 

Edward spent his internship testing the Simulacrum data.  The Simulacrum contains artificial data which imitates some of the data held securely by Public Health England’s National Cancer Registration and Analysis Service. Edward spent his time comparing simulated data outputs to real data outputs to make sure the simulated data can give accurate answers about cancer whilst also protecting patient confidentiality.

“HDI is a friendly and relaxing place to work and the internship projects on offer are an interesting way to work with anonymised cancer data. I’ve enjoyed learning lots of new and useful programme packages and techniques”

 

David tested a new methodology

David

This is “David” by Kim Whittlestone on Vimeo, the home for high quality videos and the people who love them.

During his internship, David worked on two different methodological problems using cancer data. Firstly, David examined all the current methodologies for calculating cancer survival and tested a new method to see if it would work better for rarer cancers. Then David spent the second half of his internship modelling cancer incidence to see how it has changed over time and see if it could project incidence into the future.

“If I’m stuck on anything I can send an email and ask them, and they are happy to help. I was really surprised by the amount I learnt, and it will really prepare me for next year’s module in my maths degree about medical statistics”

 

Amine experimented with natural language processing

Amine spent his time in Cambridge over the summer developing his skills in natural language processing.  Amine experimented with using NLP on cancer pathology reports to see if a computer algorithm can be used to analyse the free text in the reports. Using NLP could help the National Cancer Registration and Analysis Service (NCRAS) in Public Health England process cancer data faster and help to draw out insights from the data quicker.