AI-Powered Clot insights for improved stroke treatment

The MATRIX project is a new public-private partnership between Maastricht UMC+, Nicolab, and Erasmus MC, aiming to revolutionise stroke treatment using artificial intelligence (AI). The goal is to develop a smart system that can predict the type of blood clot causing a stroke using routine CT scans. This insight will help doctors select the most effective treatment from the very beginning—making stroke care faster, safer, and more personal.

Stroke is a leading cause of death and disability in the Netherlands and across the world. Every year, more than 40,000 people in the Netherlands suffer a stroke. Many of these patients require emergency treatment to remove a clot that blocks blood flow to the brain. However, current treatments are not always effective for every patient. Studies show that the type of clot—its structure and composition—can make a big difference in how well a treatment works. Right now, doctors can only see this after treatment, when it’s too late to change course. That’s why MATRIX is urgently needed.

MATRIX will use hundreds of stored clots and CT scans to train AI algorithms to recognise clot types before treatment begins. This will allow doctors to choose the right strategy—such as medication or mechanical removal—based on real-time information. By improving the chance of removing the clot in one go, MATRIX aims to reduce complications, speed up recovery, and lower healthcare costs.

By the end of the project, MATRIX will deliver four key results: a dataset of clot images and CT scans, an AI model to detect and classify clots, a system that integrates these insights into clinical workflows, and a user-friendly visualisation tool for doctors. Together, these will bring smarter, more personalised care to stroke patients across the Netherlands.

More information about Nicolab

More information about Cardiovascular Research Institute Maastricht