AI enabled decision support systems in epilepsy surgery a scoping review.

Area of Science:

• Neurology
• Medical Informatics
• Artificial Intelligence

Background:

• Epilepsy surgery decision-making is complex.
• Artificial intelligence (AI) offers potential support.
• Evidence for AI implementation across the epilepsy surgery pathway is limited.

Purpose of the Study:

• To map AI-enabled decision support systems in epilepsy surgery.
• To characterize datasets, modeling, validation, and integration.
• To identify gaps for scalable AI adoption.

Main Methods:

• Scoping review of 145 studies (Jan 2018 - May 2025).
• Analysis of AI systems across surgical stages and clinical tasks.
• Characterization of datasets, modeling, validation, and workflow integration.

Main Results:

• Literature concentrated on pre-operative AI; no intra-operative studies found.
• Most studies used small, single-center, non-public datasets with supervised CNN models.
• External validation and workflow integration were uncommon.

Conclusions:

• Significant gaps exist in AI generalizability, workflow readiness, and equity.
• Priorities include multi-center data, rigorous cross-site evaluation, and clinically meaningful endpoints.
• Safe and scalable AI adoption requires addressing these limitations.