Deep Learning in the Management of Intracranial Aneurysms and Cerebrovascular Diseases: A Review of the Current Literature

Abstract

Intracranial aneurysms are a common asymptomatic vascular pathology, the rupture of which is a devastating event with a significant risk of morbidity and mortality. Aneurysm detection and risk stratification before rupture events are, therefore, imperative to guide prophylactic measures. Artificial intelligence has shown great promise in the management pathway of aneurysms, through automated detection, the prediction of rupture risk, and outcome prediction after treatment. The complementary use of these programs, in addition to clinical practice, has demonstrated high diagnostic and prognostic accuracy, with the potential to improve patient outcomes. In the present review, we explored the role and limitations of deep learning, a subfield of artificial intelligence, in the aneurysm patient journey. We have also briefly summarized the application of deep learning models in automated detection and prediction in cerebral arteriovenous malformations and Moyamoya disease.

Keywords: Arteriovenous malformations; Cerebrovascular disease; Deep learning; Intracranial aneurysms; Machine learning; Moyamoya disease.