Akademiska sjukhuset (Uppsala University Hospital) announced it would introduce, from early 2024, an AI system that automatically interprets stroke images taken by CT, with the main aim of making stroke diagnosis faster and more accurate — since rapid action limits brain damage and the risk of future disability. The software supports clinicians by analysing CT scans in several ways: assessing the size of an established infarct, examining CT angiography to identify clots in the large vessels supplying the brain, and analysing CT perfusion to find areas of impaired blood circulation. Erik Lundström, section head for neurology and stroke care and adjunct professor at Uppsala University, said advanced AI for automatic interpretation of brain CT would significantly improve the hospital's ability to handle stroke cases, in line with its goal of leading in stroke care. Christoffer Nyberg, consultant and neurointerventionist, noted that everyone involved in the decision process gets the same information within five minutes, saving time and easing pressure in stressful situations. The rollout was prepared by radiology (Bild- och funktionsmedicin) together with the Neuro division; similar systems were already in use in Region Skåne, Västra Götalandsregionen and Region Örebro.

Read the full press release (in Swedish)