Keywords: MachineLearning, Prediction, Tabular
Need: Respiratory admissions are the principal cause of seasonal pressures in emergency departments. The NHS would benefit from better predictions of the peak levels and timing of increases in respiratory emergency admissions to respond better to seasonal pressures and local areas need adaptable models that can be tailored to influence their planning processes.
This project, hosted by the Suffolk and North East Essex (SNEE) Integrated Care System (ICS), seeks to deliver insights on the characteristics of individuals and exogenous factors that can explain the likelihood of future respiratory admissions. The focus is to demonstrate how a range of data from across the ICS, including primary care records, as well as data science approaches, can provide the basis for three-month predictions of emergency demand. Ideally a generalist approach will be developed so that it may be applied to other predictions in future.
It is envisaged that the work will be used to:
Current Knowledge/Examples & Possible Techniques/Approaches:
Related Previous Internship Projects: A previous project in SNEE ICS investigated inequalities specific to diabetes. The East Suffolk and North East Essex FT is advertising an internship on Inequalities in Population Health Data.
Enables Future Work: Both approach and generalised components of the code can be reused across other datasets and comorbidities.
Outcome/Learning Objectives: Open code and report demonstrating potential and evaluation of these techniques.
Datasets: Population health datasets held by SNEE Integrated Care System (primary, acute, community, mental health and more).
Desired skill set: When applying please highlight any experience around work with big data analysis, forecasting, coding experience (R/Python preferably) and software development (including any coding in the open), and any other data science experience you feel relevant.
Return to list of all available projects.