Medicine
| Location: | UK Other |
| Closing Date: | Thursday 31 August 2023 |
| Reference: | MED1955 |
Title: PhD Project: Federated Learning for Big Healthcare Data Infrastructures
Your Project ID (optional)
Application Deadline or accept year round applications: 2023
Funding Information
(UK Students Only) Funded Project
Funding Notes (optional): Funding received from Health Data Research – UK
Project Description:
An exciting new research collaboration between Health Data Research UK and the University of Nottingham is funding a UK PhD studentship within the School of Medicine.
Health Data Research UK (HDR UK) is a national institute dedicated to improving health outcomes and advancing medical research through the use of data. Established in 2018, HDR UK is a partnership between leading universities, research institutions, and the National Health Service (NHS) in the UK. The primary mission of HDR UK is to harness the power of health data to drive scientific discoveries, develop innovative treatments and interventions, and ultimately improve patient care. By integrating and analysing vast amounts of health data, including electronic health records, genomic data, and other relevant information, HDR UK aims to generate insights that can transform healthcare delivery, policy, and research.
We need to better understand how to harness the power of federated learning to design and deploy predictive models within HDR-UK federated data infrastructures, especially focussing on technology potential, limitations, and ethical concerns.
We are looking to recruit a PhD student with an interest in developing and applying federated modelling approaches to big data healthcare problems, as well as providing a critical appraisal of the potential of the models to underpin the analytics capabilities of federated data architectures. An interest in artificial intelligence applied to better care is important.
Project objectives:
The project will aim to meet the following objectives. These are broadly outlined below, and will be finalised and made more specific during the PhD.
- Systematic review exploring existing literature on federated learning approaches for healthcare data
- Systematic review exploring on limitations, opportunities, ethical and privacy concerns regarding the application of federated learning in healthcare
- Development of overarching federated approaches, deployable to multiple domains in healthcare and testing in multiple case studies
- Development of a framework for federated model prediction explainability
Further Information
You will work with Professor Philip Quinlan (Professor and Director of Health Informatics) and Dr Grazziela Figueredo (Associate Professor in Health Data Science)
Qualifications:
Applicants should hold a 2.1 undergraduate degree and a good masters’ degree in a subject relevant to data science, machine learning and AI focussed on healthcare.
Applications:
For enquiries in relation to this project, please email Philip Quinlan, using the email below. Applications for this studentship should consist of a full detailed CV and a covering letter outlining relevant experience and interest in the PhD area. A statement outlining your suggested approaches to answering the PhD brief should also be included in the covering letter. These should be submitted directly by email to Prof Philip Quinlan (Philip.Quinlan@nottingham.ac.uk). This project is also advertised via Find a PhD.
Details of how to apply on-line can be found via this link:
https://www.nottingham.ac.uk/pgstudy/how-to-apply/apply-online.aspx