Advancements in Artificial Intelligence for Stroke Management: Enhancing Diagnostics, Treatment, and Rehabilitation

Abstract

Background:  Stroke is a major contributor to disability and mortality worldwide, necessitating swift medical attention to minimize brain damage and avoid complications.

Objectives: This systematic review evaluates how Artificial Intelligence (AI) can be integrated into stroke management to enhance diagnostic precision, treatment efficacy, and personalized care. Our primary goals include determining AI's effect on improving diagnostic accuracy through imaging, optimizing treatment decision-making processes, and developing rehabilitation strategies. Moreover, the review also investigates any challenges or opportunities associated with its use in clinical settings.

Methods: An deep literature search was held using PubMed, WOS, Google Scholar, and IEEE Xplore databases from January 2010 through December 2023, using keywords such as "Artificial Intelligence," "Stroke Management," Diagnosis," Treatment," and Rehabilitation". Two independent reviewers screened and selected articles that met our eligibility criteria; ultimately, 62 met this threshold.

Results: This review highlights AI's significant contributions to stroke management, particularly improving MRI stroke detection accuracy and diagnostic efficiency and optimizing treatment plans for thrombolysis and mechanical thrombectomy. Furthermore, AI-driven rehabilitation programs offer tailored therapy programs to boost motor function recovery rates and patient outcomes, yet data privacy concerns and large dataset requirements persist as significant obstacles.

Conclusion: Artificial intelligence's rise in stroke management holds promise for further developing predictive models and tailoring treatment plans, leading to more effective healthcare interventions worldwide.