skills.rest
hiyenwong
hiyenwongCOMMUNITY
Explore all skills by this author →
𝕏@khonsu_51GitHub

explainable-gnn-eeg-neurological

Predict stroke severity from EEG connectivity using explainable graph attention.

2|Updated Feb 12, 2026
Search Tags:#eeg#brain-connectivity#explainable-gnn#stroke#nihss#gat#clinical-insights
One-click install
npx skills add https://github.com/hiyenwong/ai_collection --skill explainable-gnn-eeg-neurological
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: explainable-gnn-eeg-neurological
Source: https://github.com/hiyenwong/ai_collection/tree/main/collection/skills/explainable-gnn-eeg-neurological
Command: npx skills add https://github.com/hiyenwong/ai_collection --skill explainable-gnn-eeg-neurological

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Predicts stroke severity from EEG data using an explainable Graph Neural Network with attention, enabling clinical interpretation of brain connectivity patterns.

Core Features & Use Cases

  • Predict NIHSS from EEG connectivity with interpretable attention coefficients
  • Provide brain-network insights to support diagnosis, prognosis, and neurorehabilitation planning
  • Frequency-specific brain network analysis across delta, theta, alpha1, alpha2, and beta bands for clinical interpretation

Quick Start

Load EEG data aligned to Brodmann areas, compute LLC across five bands, construct brain graphs, train a GAT model, and inspect attention to derive clinical insights.

Frequently Asked Questions about explainable-gnn-eeg-neurological

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I predict stroke severity from EEG data using a Graph Neural Network?▼

You can predict stroke severity from EEG data by computing LLC connectivity across five frequency bands, localizing sources to Brodmann areas, constructing brain graphs, and training a Graph Attention Network to output NIHSS scores.

How does an explainable Graph Attention Network work for brain connectivity analysis?▼

An explainable GAT works for brain connectivity analysis by applying attention coefficients to brain graphs, quantifying the importance of specific neural connections across delta, theta, alpha1, alpha2, and beta frequency bands for clinical interpretation.

Do I need PyTorch Geometric to train an EEG-based Graph Attention Network model?▼

Yes, you need PyTorch Geometric to train the EEG-based Graph Attention Network model, as it provides the necessary framework for constructing brain graphs and interpreting attention mechanisms during acute stroke evaluation.

What is the best way to preprocess EEG signals for stroke severity prediction with a GNN?▼

The best way to preprocess EEG signals for GNN-based stroke severity prediction is to perform source localization to Brodmann areas and compute LLC connectivity across delta, theta, alpha1, alpha2, and beta bands to construct the input brain graphs.

Can I use EEG-derived brain graphs for acute stroke evaluation instead of traditional imaging?▼

You can use EEG-derived brain graphs for acute stroke evaluation by training a GAT model on LLC connectivity data to predict NIHSS scores, providing interpretable brain-network insights to support diagnosis and prognosis.

Related Skills

Explore More Education & Research Skills→
StreamPilotOrgStreamPilotOrg
1

metrics-calculator

Compute SaaS growth metrics and compare them against 2025 industry benchmarks.

Official
Basic
TuminhaTuminha
5

dental-image-generator

Generate dental illustrations and infographics for clinical communication.

Community
Intermediate
handsomelong922handsomelong922

tensorboard

Visualize and profile ML training runs with configurable logging.

Community
Advanced
Zeyuyang-0420Zeyuyang-0420

torch-geometric

Build and optimize Graph Neural Networks with PyTorch Geometric.

Community
Advanced