Graph-based models improve epilepsy zone localization from sEEG data

Researchers have developed and studied graph-based models for localizing the epileptogenic zone (EZ) using stereo-EEG (sEEG) recordings. The study compared various graph topologies, including dense graphs, anatomy- and geometry-informed priors, and proposed Region-Bridge-c, to assess their impact on EZ localization accuracy. At approximately 30% edge retention, the Region-Bridge-c topology achieved the highest performance in terms of PR-AUC and ROC-AUC, significantly outperforming dense graphs while using fewer edges. The findings suggest that explicit evaluation of graph construction is crucial, as the optimal topology can vary significantly between patients. AI

IMPACT This research could lead to more accurate and efficient methods for identifying the epileptogenic zone, potentially improving outcomes for epilepsy surgery patients.

RANK_REASON The cluster contains an academic paper detailing a novel approach to a scientific problem. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• IArxiv
• Region-Bridge-c
• ScienceCast
• Spatial Knowledge Spillovers and University Research: Evidence from Austria

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →