Lightweight AI Model for Medical Image Analysis

Overview

Artificial intelligence is transforming healthcare, with medical imaging at the forefront. From detecting early signs of cancer to supporting rapid triage in emergency care, AI has the potential to save lives. However, the most advanced models are often too large and resource-intensive to be deployed outside specialist centres.
This PhD project tackles the challenge of designing lightweight AI models that deliver accurate results while running efficiently on standard clinical hardware or even mobile devices.

Why it matters

•  Clinical Relevance: Expands access to advanced diagnostics in resource-limited hospitals and community care.
• Technological Innovation: Advances state-of-the-art methods in model pruning, quantisation, knowledge distillation, and novel neural network design.
• Global Impact: Democratizes healthcare by making AI-driven imaging tools more widely available.

Key Objectives

•  Develop methods to compress and optimise AI models without compromising diagnostic accuracy.
• Apply and test lightweight architectures across key imaging tasks such as disease diagnosis, organ segmentation, and anomaly detection.
• Collaborate with clinicians to validate approaches and ensure alignment with real-world healthcare needs.

Why Choose a PhD at Dundee?

•  Unique Collaboration: Work across computing and medicine to design clinically grounded AI.
• Expert Guidance: Benefit from supervision in cutting-edge computer vision and deep learning, with direct links to clinical research.
• Research Excellence: Join a university recognised globally for innovation in medical imaging and AI.
• Career Opportunities: Gain interdisciplinary skills valued in academia, healthcare, and industry.

Diversity statement

Our research community thrives on the diversity of students and staff which helps to make the University of Dundee a UK university of choice for postgraduate research. We welcome applications from all talented individuals and are committed to widening access to those who have the ability and potential to benefit from higher education.