Title: Machine learning imaging-based differentiation of clot composition in acute ischemic stroke and its potential use in decision-making in mechanical thrombectomy.

This research is the first of its kind. We integrated CT-source data with advanced histological analysis, classifying the clots and exploring their characteristics. From this analysis a trend emerged, indicating that clots with extreme perviousness characteristics may be predominantly composed of a single component. This innovative partnership highlights the potential of combining machine learning and medical expertise to further advance thrombectomy and improve patient outcomes.