Application of Machine Learning and Deep Learning in Imaging of Ischemic Stroke

ABSTRACT

Timely analysis of imaging data is critical for diagnosis and decision-making for proper treatment strategy in the cases of ischemic stroke. Various efforts have been made to develop computer-assisted systems to improve the accuracy of stroke diagnosis and acute stroke triage. The widespread emergence of artificial intelligence technology has been integrated into the field of medicine. Artificial intelligence can play an important role in providing care to patients with stroke. In the past few decades, numerous studies have explored the use of machine learning and deep learning algorithms for application in the management of stroke. In this review, we will start with a brief introduction to machine learning and deep learning and provide clinical applications of machine learning and deep learning in various aspects of stroke management, including rapid diagnosis and improved triage, identifying large vessel occlusion, predicting time from stroke onset, automated ASPECTS (Alberta Stroke Program Early CT Score) measurement, lesion segmentation, and predicting treatment outcome. This work is focused on providing the current application of artificial intelligence techniques in the imaging of ischemic stroke, including MRI and CT.