PhD Studentship in AI and Data Science

Investing in excellence: studentship opportunities

The University of Derby has an opportunity for a full-time postgraduate research studentships in Data Science area of research in the College of Science and Engineering (SE).

The successful applicant will receive a maintenance stipend (based on the minimum stipend defined by UKRI, currently £19,237 for the academic year 2024/25) and home MPhil/PhD tuition fees (£4,786 - subject to amendment) only up to the target submission date.

Please note: if your application is successful and you are assessed as international for fees purposes, you will need to pay the difference between the home fees and the international fees.

The intended intake period is January 2025, or the next available intake.

The successful applicant will be expected to complete their MPhil/PhD within 3 years on the MPhil/PhD route, contribute to the College REF submission and get involved in the wider research activities of the College.

Applicants will become part of a friendly and welcoming team and will be supported and managed by their supervisors.

The vacancy details are as follows:

Purpose/objectives 

The primary aim of this studentship is to explore the potential of Artificial Intelligence (AI) by utilising Federated Learning (FL) and Generative Adversarial Networks (GANs) to advance predictive analytics and decision-making processes within the Blood and Transplant Unit. This project seeks to enhance the precision of transplant outcome predictions, detect possible complications at an early stage, and improve both pre- and post-transplant care. GANs will be leveraged for data pre-processing to generate synthetic datasets that supplement existing data, addressing challenges related to data scarcity, imbalance, and privacy issues. By combining FL and GANs, this studentship intends to establish a robust, privacy-preserving AI framework that can be extended across various healthcare settings within the NHS. The ultimate goal is to create a comprehensive strategy for integrating AI into primary care, leading to more efficient, personalised treatment plans and better patient outcomes.

In an era marked by rapid technological advancements, the healthcare landscape stands on the brink of transformation. Primary care, often considered the cornerstone of healthcare systems worldwide, faces an array of challenges, from the increasing demand for services to the need for more efficient resource allocation. This project embarks on a journey to chart a comprehensive roadmap for the application of Federated Learning (AI technique) in the realm of Blood and Transplant Unit, aiming to enhance predictive analytics for transplant outcomes, identifying potential complications and improving pre- and post-transplant care.

Federated learning (FL) is an innovative approach to machine learning that allows algorithms to be trained across multiple decentralised devices or servers holding local data samples, without exchanging them. This method will play a pivotal role to improve operations in the Blood and Transplant units of the NHS.

Project description

Advancing Research and Innovation in Transplant Matching

Objective: Drive progress in transplant matching research through AI to uncover new insights and enhance matching algorithms.
Details: Apply AI to large and complex datasets, identifying patterns and correlations that can optimise the matching process between donors and recipients.

Improving Data Quality and Accessibility with GANs

Objective: To employ GANs for data pre-processing for high-quality synthetic datasets that augment real-world data.

Details: GANs will be used to tackle challenges such as data scarcity and imbalance, common issues in transplant research. By creating synthetic data that closely mirrors the statistical characteristics of actual patient data this objective will strengthen the robustness and generalizability of predictive models.

Creating a Federated Learning Framework

Objective: To design and implement a Federated Learning (FL) framework specifically for the Blood and Transplant Unit within the NHS.

Details: Develop a decentralised AI model that enables the training of predictive analytics models on data from multiple sources (e.g., hospitals and clinics) without needing to transfer sensitive patient information. The framework will ensure that these models can be trained on a diverse and comprehensive dataset, improving the accuracy of transplant outcome predictions while protecting patient privacy.

Encouraging Collaboration and Knowledge Sharing Across Healthcare Units

Objective: To foster interdisciplinary collaboration and disseminate the knowledge gained from this project to other healthcare units within the NHS.

Details: The project is intended to serve as a model for the integration of AI into healthcare, offering insights and methodologies that other units can adopt. This objective includes sharing research findings through publications, presentations, and workshops to promote the use of AI-driven solutions in other areas of healthcare.

Potential project impact 

The incorporation of Federated Learning and GANs in this project is expected to lead to significant improvements in the operational efficiency and patient care within the Blood and Transplant Unit. GANs will be employed for data pre-processing, enabling the creation of high-quality synthetic datasets that enhance model training while safeguarding patient confidentiality. This approach is particularly valuable in addressing the challenges of limited and imbalanced datasets, which are common in transplant-related research. By enhancing the accuracy of predictive models, the project aims to minimise post-transplant complications, improve survival rates, and optimise resource allocation within the NHS. Additionally, the successful implementation of this AI framework could serve as a benchmark for other units, fostering widespread adoption of AI-driven solutions across healthcare systems. The project also holds the potential to accelerate research and innovation, contributing to the development of more effective and personalised patient care strategies.

Principal accountabilities and responsibilities

Principal Accountabilities and Responsibilities of the Student:

• Research and Development: The student will be responsible for designing and developing a Federated Learning framework integrated with GANs to pre-process and improve the quality of datasets used for predictive modelling. This includes selecting appropriate algorithms, training models across decentralized data sources, and optimising these models for transplant-related predictions.

• Data Management, Pre-processing, and Security: The student will manage and pre-process data using GANs to create synthetic datasets that enhance the existing data. This process involves ensuring data quality, addressing issues of data scarcity and imbalance, and implementing privacy-preserving techniques to protect patient information.

• Model Testing, Validation, and Deployment: The student will conduct thorough testing and validation of the developed models to ensure their accuracy, reliability, and applicability in real-world scenarios. The student will also work with healthcare professionals to implement these models in clinical settings and gather feedback for further improvements.

• Collaboration and Communication: The student will work closely with a multidisciplinary team, including data scientists, healthcare practitioners, and IT specialists. The student will also be responsible for reporting on project progress, addressing challenges, and sharing outcomes through detailed reports, presentations, and academic publications.

• Ethical and Regulatory Compliance: The student must ensure that all research activities comply with ethical guidelines and regulatory standards, particularly those related to patient data privacy and the use of AI in healthcare. The student will collaborate with NHS regulatory bodies

• Knowledge Dissemination: The student will contribute to the academic community by publishing research findings in high-impact journals and presenting at national and international conferences.

To apply

Please review our entry requirements before submitting your application and check out the 'Getting Started' section below.

Completed applications should be submitted via our online application system quoting funding reference: EXT_S&E_AI-DataScience_2425

Apply online

Closing dates for applications: Midday, Wednesday 6 November 2024.

(Please note: we encourage applicants to apply as soon as possible as we reserve the right to close before 6 November 2024 if a high volume of applications is received.)

Interviews: 27 November 2024.

If you have not been invited for an interview by the interview date, please assume your application has been unsuccessful.

For other enquiries which are subject-specific please contact:

Dr Aaisha Makkar - a.makkar@derby.ac.uk 

Find out more about our research degrees.

Getting started

Before you begin your application, make sure you have:

Studentship funding reference code

This is provided on the individual studentship advert and must be specified in your application. 

Personal statement

A 500-word personal statement outlining your suitability for the studentship project. This is a mandatory requirement and you must upload it into your application. You should include your reasons for applying for the studentship, your experience in the field, how you feel you would benefit from studying and relevant information about your previous studies.

Your CV

A CV outlining your academic and professional experience.

Qualifications

Your qualification details including grades and dates taken. You will have the opportunity to upload scanned copies of your qualification certificates and transcripts in the application. If you have no formal qualifications, you can also state this in your application.

Passport/birth certificate

A scanned copy of your passport or full birth certificate. This will help us verify your application to study with us. International applicants can provide a copy of their passport only for visa assessment purposes, and their current visa if residing within the UK.

Academic references

Two signed academic references. This is optional at application stage but highly encouraged. If successful in your application, two academic references will be a mandatory requirement of admission. The references should be in written format, signed and dated from either a supervisor or tutor from your most recent studies.