In Novo Nordisk Development, we manage clinical drug development worldwide, ensuring that the process lives up to uniform global standards, regulations and business ethics while delivering viable products that make a difference to patients and ultimately benefit society.
In Data Management – part of Development – we are responsible for capturing and cleaning all the incoming data from our clinical trials, and preparing it for statistical analysis within a very heavily regulated (GxP) environment. This includes identifying unusual datapoints, or patterns in datasets containing typically several hundred features, and interacting with the medical staff who treat our patients to query anomalous data. Currently, this is a very manual, labour-intensive task.
With the expected increase in data volumes over the coming years, there is an urgent need to increase the level of automation in our data cleaning, through the use of automated outlier-detection algorithms and AI-enabled insights to allow us to focus our efforts where they are most needed. In this project, you will be working with historic (labelled) data from some of our concluded trials, with the purpose of training a model to:
Recognize unusual patterns in the data
Based on prior learnings, recommend whether querying these data would produce an impactful data change or not.
This project builds on the work of previous Masters students and Novo Nordisk Graduates, and would be carried out in collaboration with the Data Science Automation Group, a diverse team of 7 Specialists who combine business and technical knowledge, with a mandate to identify and eliminate manual work in our drug development pipeline.
Contact Kian Norouzi (Data Management Process & Innovation Specialist) or Alasdair Thomson (Principal Data Management Process & Innovation Specialist).