Real-Time Imaging of Platelet Activity Predicts Blood Clots Before They Form

Researchers from the University of Tokyo have developed an innovative method to observe blood clot formation as it happens without invasive procedures. Utilizing a cutting-edge microscope combined with artificial intelligence (AI), this technique enables precise tracking of platelet behavior in patients with coronary artery disease (CAD). The study demonstrates that capturing the dynamics of platelet aggregation can help predict clot formation risks, potentially leading to better personalized treatments.

Platelets are small blood cells that play a vital role in stopping bleeding by forming clots. However, in individuals with heart disease, these cells may overreact, creating dangerous clots within arteries that can cause heart attacks or strokes. Understanding their activity in real time has thus been a significant challenge.

Lead researcher Dr. Kazutoshi Hirose explains that the new approach employs a frequency-division multiplexed (FDM) microscope—a device that functions like a high-speed camera to record detailed images of flowing blood cells. Artificial intelligence then analyzes these images, distinguishing between single platelets, platelet clumps akin to traffic jams, and white blood cells caught in the process.

The team tested this technology on blood samples from over 200 patients, revealing that those with acute coronary syndrome exhibited more platelet clumping than patients with chronic conditions. Interestingly, the study found that blood drawn from a vein in the arm provides similar insights into platelet activity within coronary arteries, simplifying the testing process. This non-invasive approach could revolutionize how clinicians monitor and manage blood clot risks.

Dr. Keisuke Goda remarks that advancements in high-speed imaging and AI have unlocked new possibilities for visualizing blood cell movement, enabling pattern recognition beyond human capabilities. The ultimate goal is to customize antiplatelet therapies based on individual platelet behavior, which varies from person to person. Such personalized treatment could improve outcomes and reduce complications, like recurrent thrombosis or bleeding.

This research represents a significant step toward safer, more effective management of heart disease, leveraging technology to predict and prevent life-threatening blood clots in real time.