CDU data science team blog
UX and data science
Trialing UX techniques to evaluate data science products
Patient feedback text mining project update
The second phase of the pxtextmining and experiencesdashboard projects has begun! What have we achieved so far, and what are our plans?
When things go wrong in GitHub
Avoiding mistakes and what to do when accidents happen in GitHub
Data Scientist job
An accessible version of the Data Scientist band 7 job in the team
Shiny Developer job
An accessible version of the Shiny Developer band 7 job in the team
Extending issues in GitHub
Details on features used in GitHub for creating issue templates and labels, particularly the use of `wontfix`.
Statistical Process Control (SPC) R Packages
A quick introduction to the R packages available and how they are used by the team
Caseload over time
How to count open referrals in a given period of time.
Creating a branch and using issues
Steps to take when creating branches following the GitHub-Flow method with particular relation to how to tie these in with issues.
Keeping Git branches tidy
Following the GitHub SOP tidying branches needs to happen both on GitHub and locally and this details some of the ways to do that.
Settings in SSMS
Some of the options that may be useful in SSMS
Using {usethis} to set up .gitignore
How to ensure that certain files cannot be accidentally committed to GitHub (or any other version controlled area).
GitHub Standard Operating Procedure
How we use GitHub in the CDU data science team.
Open source within the NHS - benefits and possible pitfalls
What does open source mean for the average staff member in the NHS? What benefits does it bring to patients?
Package Workflow
From a team time session discussing the workflow to contributing to a (currently) private package
Personal Access Tokens
Connecting to the GitHub to install packages from a private GitHub repository requires security set ups and this blog details how to do it (and how not to do it).
Development of open tools for analysing healthcare data in R
***TLTR: (Too Long To Read)*** Our goal was to make it easier to work with healthcare data in a reproducible and collaborative way. We wrote lots of R functions for recurring data manipulations and analytical tasks that magically translate into SQL code and communicate with large databases. All our functions are grouped into R packages because this made it easier for us to: *(i)* write good documentation of our code and analytical tasks, *(ii)* easily distribute updates across all team members, *(iii)* formally test our code, and *(iv)* integrate common data manipulations (or analyses) into interactive dashboards in a modular way.
A Text Mining Pipeline for NHS Patient Experience Feedback
This blog post is a more technical description of the pipeline that we have built to analyse patient feedback text data from the NHS.
Pair programming, code review, journal club, and team time
What can the CDU data science team do to verify its outputs, disseminate learning, and support individual development in team sessions?
Age bands methodology
A blog compiling all the age bands methodology that can be used for comparing analysis populations against.
Population Projections
Links for population projection data.
Making presentation slides into a GitHub repository template
One of a series of posts relating to creating presentation templates using {xaringan}, GitHub and R Studio.
Making an RStudio presentation template
One of a series of posts relating to creating presentation templates using {xaringan}, GitHub and R Studio.
Index of Multiple Deprivation
The measure of relative deprivation in small areas in England called lower-layer super output areas
Mapping
Mapping using public health tools
Working in the open
What does it mean to work in the open? What is open source? What problems can we solve if we share more openly?
A new GitHub release and future projects
We have a new project out and would like to tell you about some more of our future work.
Data clinics in Nottinghamshire Healthcare
We have been working with teams to help them with their data problems. This post describes some of the clinics and what has come about as a result of this work.
Classification of patient feedback
An NHSE funded project to devise an application to automatically tag the content of patient feedback