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 nanobubble-enhanced ultrasound imaging as an accessible alternative to MRI for cancer diagnosis. Tiny gas-filled nanoparticles amplify ultrasound signals and improve image quality, particularly in prostate cancer. The technology could reduce diagnostic delays, lower costs, and provide high-quality medical imaging to more patients worldwide.

This research develops a spatial “GPS” for lung cancer screening by mapping CT scans into a shared coordinate system. By identifying high-risk regions for malignant nodules, it supports radiologists and AI in improving diagnostic accuracy, decision-making, and interpretability, transforming screening from broad search to targeted, data-driven precision.

This research examines how AI is used in NHS radiology and challenges claims that it will replace radiologists. Instead of full automation, AI supports clinicians, helping manage workforce shortages while radiologists retain responsibility for diagnosis and treatment decisions. Evidence, not hype, should guide debates about AI and work.