Predictive value of clot imaging in acute ischemic stroke: A systematic review of artificial intelligence and conventional studies

Abstract

The neuroimaging signs of the clot in acute ischemic stroke are relevant for clot biology and its response to treatment. The diagnostic and predictive value of clot imaging is confirmed by conventional studies and emerges as a topic of interest for artificial intelligence (AI) developments. We performed a systematic review to evaluate the state of the art of AI in clot imaging, how far AI is from becoming clinically beneficial, and what are the perspectives to consider for further developments. In parallel, the review is examining the evidence brought by conventional studies concerning the relevance of clot imaging, from 2019 to August 2022. The automatic detection and segmentation of the clot are the most important advances towards AI implementation in the clinic. Predictive radiomics models require further exploration and methods optimization. Future AI approaches could consider conventional clot imaging characteristics and patient specific vascular features as variables for model development.