Artificial Intelligence in Stroke Diagnosis: A Bibliometric Analysis

Abstract

Objectives: Stroke is one of the life-threatening conditions that is considered as 5th leading cause of death that requires urgent intervention. The objective of this study was to identify and review the importance of Artificial Intelligence (AI) in the diagnosis of stroke.

Methods: A review that includes AI and stroke-related studies which are conducted in accordance with the PRISMA chart. A variety of search engines were used to collect 121 articles then a master Excel sheet was used to extract the necessary data which underwent many steps of filtration to exclude the unrelated articles before starting the analysis process.

Results: Out of 121 studies published between 2013 to 2022 identified at the beginning of the study, only 39 studies were used in the final analysis. The majority of studies, thirteen (32.5%), were published in 2021 compared to 2014 which represents the least year of publications of such studies with only three (7.5%) studies. More than half of the studies, 22 (56.41%) of studies were retrospective type of studies. Seven (17.5%) of the studies were conducted in China which represents the highest number of studies to be published in a country. Among all studies included, the most common modality of AI used was machine learning 15 (38.5%). Conclusion: The number of AI studies in the past 10 years is increasing year after year and most of these studies are retrospective.