Introduction
A regular series of open invitation talks from leading voices in data science, AI, healthcare and those with lived experience dealing with multiple long-term conditions (MLTC). This seminar series is part of the AI for multiple long-term conditions: Research Support Facility (RSF) project.
Event title: Developing and publishing code for trusted research environments
Sub-title: Best practices and ways of working - lessons from the Wales Multimorbidity Machine Learning Project
Speaker: Ed Chalstrey, Research Data Scientist, The Alan Turing Institute
About the event
Trusted Research Environments (TREs) are becoming commonly used for the analysis of data from a range of sources, particularly electronic health records. Data within TREs are kept secure and are only accessible following appropriate approvals and access being granted, to comply with the legal requirements of data providers (e.g., the NHS), allowing research to be carried out safely.
But how can researchers ensure that analyses carried out with TREs are as reproducible as possible, given the constraints of working with sensitive data in a largely closed environment? What are the considerations when it comes to publishing results and crucially methods code, when that code has been developed to run on sensitive data that cannot itself be published?
In this talk, Ed Chalstrey will discuss some of the ideas around beginning to answer these questions that came out of the Wales Multimorbidity Machine Learning (WMML) project. This is in collaboration with Swansea University, The University of Manchester and The Alan Turing Institute. The talk will begin with an introduction to TREs, with some examples of what these platforms look like. Ed will then move on to discuss the report he has authored offering practical advice that researchers using TREs can follow.