London, 16 September 2026 – Researchers have developed a deep-learning system that predicts contrast-enhanced brain-tumour MRI images from scans taken without contrast agents, offering a possible route to reduce repeated gadolinium exposure. The technology is promising for patients needing frequent follow-up, but it is not yet a universal substitute for clinically indicated contrast imaging.
The model was evaluated on a held-out test group of 1,109 patients, an important step beyond training performance. A separate human-AI study involved 11 consultant neuroradiologists at six London hospitals and found that collaboration improved diagnostic accuracy and consistency compared with unsupported reading.
Unlock the Full Article
This article is exclusive to The Ledger Asia Subsribers / PAID members.
Already have an account? Log in here

