Artificial Intelligence in Neurovascular Surgery: Emerging Applications in Intracranial Aneurysm Management

Background/Objectives: Artificial intelligence (AI) has emerged as a promising adjunct in intracranial aneurysm (IA) care, with potential applications in diagnosis, rupture risk assessment, treatment planning, and outcome prediction. This review summarizes current AI-based approaches in the management and prognostication of IAs. Methods: A structured search of PubMed, Embase, and Scopus was conducted for English-language studies published from January 2001 to 14 April 2026, evaluating machine learning (ML), deep learning, or large language model (LLM) applications in IA diagnosis, treatment, or prognosis. After screening and full-text review, 45 studies were included. Results: AI models demonstrated utility across several domains of aneurysm care. Deep learning approaches supported aneurysm detection and morphological segmentation, while traditional ML models were applied to rupture status classification, aneurysm stability, treatment selection, occlusion prediction, and post-treatment outcomes. Emerging LLMs showed early promise in treatment recommendation and prognostication, although evidence remains limited. Conclusions: AI may improve individualized aneurysm management from initial diagnosis to long-term outcome prediction by integrating clinical, radiographic, procedural, and hemodynamic data beyond conventional scoring systems. However, current models are limited by retrospective designs, heterogeneous outcomes, limited external validation, and concerns regarding interpretability, bias, and clinical implementation. Prospective multicenter validation and standardized reporting are needed before AI tools can be reliably incorporated into routine aneurysm care.