Abstract
This study presents a novel method for classifying ischemic stroke aetiology using deep learning models, focusing on cardioembolism and large artery atherosclerosis. Recognizing the global burden of ischemic strokes and the complexity of their causes, we utilize convolutional neural networks (CNNs) to analyze medical imaging data. The method aims to accurately identify the origins of blood clots in stroke patients, enhancing diagnostic accuracy and treatment efficiency. Our approach demonstrates promising results in stroke classification, offering significant implications for medical practice, particularly in regions like Kazakhstan with high stroke prevalence. This research contributes to the evolving field of applying artificial intelligence in healthcare, specifically in stroke management.