List of PhD projects where we currently take students this year.
Click on the links below or scroll down for details.
- AI-Driven Multimodal Biomedical Data Fusion
- Deep Learning for Medical Imaging
- Generative AI for Digital Marketing
- Natural Language Processing for Business Intelligence
- Deep Learning for Personalised Remarketing
- AI-Assisted Tax Assessment
- Human augmentation through Artificial Intelligence
- UX/UI Design with Interactive and Generative AI
- AI-Driven Multimodal Biomedical Data Fusion
Technological advances in biomedicine now enable the collection of high-dimensional, heterogeneous data across multiple platforms, including imaging, genomics, transcriptomics, metabolomics and clinical information, providing a comprehensive view of disease biology. Despite this, integrating such multimodal data remains challenging due to the complex interactions across data types. Artificial intelligence (AI), particularly transformer-based models behind the modern interactive and generative AI, provides new opportunities to contextualise within-modality features and uncover latent relationships across modalities through self-attention and cross-attention mechanisms. This project aims to develop advanced AI models to fuse multi-modal imaging, multi-omics, and preclinical/clinical datasets to predict immunotherapy outcomes and uncover mechanisms of response and resistance. You will have opportunities to gain experience in research into representative AI architectures including the transformer, other state space models (e.g. Mamba) and hybrid models (e.g. Jamba and Griffin).
You may focus on one or more of the following key research directions, with potential outcomes, including:
(1) developing and validating models for accurate patient stratification, including early identification of non-responders to immunotherapy.
(2) revealing key spatial and molecular determinants of therapy resistance, integrating imaging and omics signatures.
(3) developing composite non-invasive biomarkers that fuse imaging and multi-omics data for precise prediction of treatment response,
(4) informing new clinical decision protocols, enabling better patient selection and targeted combination therapy,
(5) establishing mechanistic links between tumour microenvironmental factors like hypoxia and metabolic reprogramming and immunotherapy resistance.
(6) guiding drug development by identifying critical molecular targets associated with immunotherapy response, with the long-term goal of overcoming resistance and improving patient outcomes in immunotherapy.
Project brief:
PhD Project - Medical Imaging.pdf
- Deep Learning for Medical Imaging
Medical imaging seeks to reveal the structures and activities inside the human body that are normally
invisible behind other body parts such as skin and bones. With a wide range of imaging technologies
such as radiography, magnetic resonance imaging, ultrasound, echocardiography and tomography, it
has long been of great importance in clinical practices for measurement, identification, location and
structural analysis of targeted organs, tissues or areas of abnormality. In this project, we will address
the key problems of medical imaging such as segmentation, image registration, image generation, classification and diagonose support, using deep learning methods including Convolutional Neural Network (CNN), U-Net and its variants (ResUNet, nnUNet, etc), and Generative Adversarial Network (GAN), Diffusion Models, Large Language Models (LLM) and Multi-Modal Models.
Project brief:
PhD Project - Medical Imaging.pdf
- Generative AI for Digital Marketing
In the domain of digital marketing, the integration of Generative AI presents an opportunity to
enhance customer engagement, personalise marketing strategies, and optimise decision-making
processes. By utilising Generative AI technologies, businesses can create tailored content, automate
marketing tasks, and derive insights from datasets to drive more effective marketing campaigns. This
project aims to explore the impact of incorporating Generative AI into digital marketing practices,
focusing on enhancing customer experiences, improving marketing efficiency, and staying
competitive in the digital landscape.
The research may focus on one or more of the following aspects of text creation,
image/graphics/video generation, and enhancing customer interactions. You will have opportunities to
gain experience with the industry-demanding Generative AI technologies including Generative
Adversarial Network (GAN), Diffusion Model (DM), Large Language Model (LLM), Large Vision
Model (LVM), and Multimodal Model.
Project brief:
PhD Project - Generative AI for Digital Marketing.pdf
- Natural Language Processing for Business Intelligence
Business operations involve a large amount of data from their internal communications, presentations, reports, accounting and marketing, and external information from governments, industry sectors, customers, suppliers, competitors etc. These data are often available in unstructured or semi-structured formats, and therefore make the data analytics a challenging task. In this project, we aim to develop novel methods using the modern artificial intelligence (AI) and natural language processing (NLP) technologies to support decision making for both business operations and strategic planning.
Project brief:
PhD Project - NLP for Business Intelligence.pdf
- Deep Learning for Personalised Remarketing
The project is aimed to investigate and develop new methods to manipulate and generate images and
videos for personalised remarketing. Realistic image and video contents will be automatically
generated from a user’s online profile (preferences, browsing history, shared pictures etc.) by
embedding compatible items (e.g. clothes, hats, sunglasses etc) to the original pictures or videos of the
user.
This is an example of how the system works. From a simple picture of a user, new images will be
automatically generate with “add-on” promotional items (e.g. clothes) specifically for the user only.
Also different poses, gestures or background fitting to the user’s geolocation and temporal context
will be incorporated into the process.
The key methods and algorithms of the project include semantic segmentation from an input image,
items (e.g. clothes) prototyping, person detection and pose estimation, effective training of generative AI models auch as Generative
Adversarial Networks (GANs) or Diffusion Models (DMs).
Project brief:
PhD Project - Personalised Remarketing.pdf
- AI-Assisted Tax Assessment
Identifying potential capital allowances involves large volumes of construction and fixed asset cost
data to be analysed to assign the correct tax treatment, requires an understanding of tax legislation,
complex case law and construction terminology. This process is time-consuming and complex, and as
a result, often available tax relief is overlooked. In this project, we will develop an automated solution
to this problem by using advanced techniques of Artificial Intelligence and Natural Language
Processing. It involves a combination of the traditional AI technologies such as rule-based expert
system and the modern natural language processing and gnerative AI.
Project brief:
PhD Project - Tax Assessment.pdf
- Human augmentation through Artificial Intelligence
With the rapid advancements in AI, human augmentation has moved from science
fiction to real-world applications, providing solutions that empower individuals in their
daily lives, workplaces, and critical decision-making contexts. As AI systems evolve
from tools to collaborative partners, there is now a need for solutions that support
and amplify human capabilities rather than replace them by acting as cognitive,
emotional, and physical partners - helping users make better decisions, improve
skills, and even overcome physical disabilities. However, the rapid growth of AI
raises critical questions about reliability, safety and trustworthiness. Ensuring that AI
systems are responsive to diverse user needs is therefore crucial for achieving
meaningful human-centred augmentation, whilst respecting human autonomy and
maintaining trust. This research aims to explore these intersections between
technology and humans, advancing the design and development of human-centred
AI applications that act as collaborative partners, providing personalised, intuitive
support in real-world diverse domains, such as healthcare, education, social support,
creative industries, and many more. By designing AI systems that adapt to user
needs, explain their processes, and build trust, PhD candidates will contribute to AI
technologies that ultimately augment human capacity.
Project brief:
PhD Project - Human Augmentation through AI.pdf
- UX/UI Design with Interactive and Generative AI
User experience/interface (UX/UI) design is a complex and time-intensive process
that traditionally relies on highly skilled professionals to create intuitive, engaging,
and effective solutions. This labour-intensive and time-consuming process can be a
bottleneck in product development cycles. The emergence of artificial intelligence
(AI), particularly interactive AI and generative AI, can significantly accelerate design
workflows by automating repetitive tasks, generating design variations, providing
data-driven insights, and enabling a wider range of users to engage in the design
process. This project aims to explore how AI can augment human designers'
capabilities in the above-mentioned area of UX/UI design. You may focus on one or
more of the following key aspects: (1) new AI models or frameworks to create unique
design elements, patterns, and illustrations, (2) generative user interfaces (genUI) for
personalised individual experiences, (3) automated or AI-assisted user persona and
profile generation, (4) transforming from low fidelity to mid and high fidelity
prototypes, (5) AI-assisted user insight analysis for both user requirement
identification and design evaluation, (6) AI-enhanced accessibility design and/or
assessment, to name a few.
Project brief:
PhD Project - UX-UI Design with Interactive and Generative AI.pdf