This research investigated whether AI-guided handheld ultrasound can help diagnose deep vein thrombosis (DVT) in primary care. Through a systematic review, a clinical study involving 565 patients, and stakeholder interviews, the research found promising results but highlighted challenges involving image quality, accountability, and integration into NHS healthcare systems.

This research develops a physics-based method for measuring lung elasticity from medical imaging to predict which emphysema patients will benefit from lung valve treatment. By creating detailed elasticity maps, the work aims to improve treatment selection, enhance patient outcomes, and provide new quantitative tools for assessing lung health.

This research develops an AI model that combines thyroid ultrasound imaging with genetic testing to improve diagnosis of indeterminate thyroid nodules. By integrating molecular and imaging data, the model helps distinguish benign from cancerous nodules more accurately, reducing unnecessary surgeries and improving clinical decision-making for thyroid cancer patients.

This research develops soft, tissue-like implantable sensors capable of monitoring molecular signals inside the body in real time. By combining high-performance electronics with flexible, biocompatible materials, these devices could detect inflammation, stress, or organ damage before symptoms arise, enabling earlier diagnosis and more personalized healthcare.

This research investigates whether artificial intelligence can help non-specialist clinicians diagnose deep vein thrombosis using AI-guided handheld ultrasound devices. By enabling faster point-of-care diagnosis in GP surgeries, the project aims to reduce hospital referrals, improve accessibility for vulnerable patients, and help healthcare systems manage increasing clinical demand more efficiently.

This research develops advanced brain-machine interface systems to improve life for spinal cord injury patients. Using neural networks such as FinNet and dynamic recurrent neural decoders, the work aims to better extract and translate brain activity into movement while creating low-power hardware capable of supporting long-term practical neuroprosthetic applications.

This research develops engineered ultrasonic reporters that allow ultrasound imaging to detect molecular activity rather than only anatomical structure. By targeting biological signals associated with cancer progression and cellular communication, the work aims to distinguish aggressive disease earlier and improve precision medicine through real-time, noninvasive monitoring of underlying cellular behavior.

This research uses artificial intelligence to predict the progression of Alzheimer’s disease and cancer using medical imaging data. By analyzing brain scans, tumor scans, and treatment responses, AI models can forecast disease development and treatment outcomes, enabling earlier intervention, more personalized care, and improved quality of life for aging populations.

This research improves photoacoustic imaging, a technique that uses light-generated sound waves to visualize tissue oxygenation deep inside the body. By calibrating measurements using highly oxygenated arterial blood, the method overcomes longstanding accuracy limitations and avoids skin-tone bias, potentially improving early tumor detection and non-invasive monitoring of tissue health.

This research redefines digital health literacy for an AI-driven world, emphasizing the alignment between users and technology. Using a Delphi method, it identifies three core components—knowledge, skills, and context. The resulting framework guides the design of digital health tools that better support behavior change by adapting to users’ real-world needs.