SEED-VIG Skill (Dataset-Orchestration Layer)

Overview

seed-vig-skill is the NeuroClaw orchestration skill for the SEED-VIG (SJTU Emotion EEG Dataset - Vigilance) dataset, developed by the BCMI Lab at Shanghai Jiao Tong University for vigilance/fatigue detection research.

It strictly follows the NeuroClaw hierarchical design principles:

• This skill only describes WHAT needs to be done and which tool skill to delegate to.
• It contains no implementation code or concrete commands.
• All concrete execution is delegated to existing base/tool skills via claw-shell.
• Companion scripts in scripts/ provide reference implementations for EEG validation, feature extraction, and vigilance classification.

Core workflow (never bypassed):

1. Identify input SEED-VIG data and target analysis.
2. Generate a numbered execution plan clearly stating WHAT needs to be done and which tool skill will handle each step.
3. Present the full plan, estimated runtime, resource requirements, and risks to the user and wait for explicit confirmation ("YES" / "execute" / "proceed").
4. On confirmation, delegate every step to the appropriate skill via claw-shell.
5. After execution, save all outputs in a clean directory structure (seed_vig_output/).

Research use only.

---

Quick Reference

Task	What needs to be done	Delegate to	Expected output
EEG validation	Validate SEED-VIG BIDS structure	scripts/validate_seed_vig.py	Validation report
EEG preprocessing	Filtering, artifact removal	eeg-skill	eeg_output/ preprocessed EEG
Feature extraction	Band power, DE, connectivity	scripts/extract_seed_vig_features.py	Feature matrices
Vigilance classification	Binary/multi-class vigilance detection	scripts/classify_seed_vig.py	Classification results

---

Dataset Characteristics

• Cohort: 23 healthy subjects
• Task: Simulated driving task (vigilance decrement paradigm)
• EEG System: 17-channel EEG (ESI NeuroScan or dry electrodes)
• Sampling rate: 200 Hz
• Reference: Linked mastoids (M1/M2)
• Labels: Vigilance levels (KSS scale or EEG-derived)
• Duration: ~2 hours per subject
• Access: BCMI Lab (bcmi.sjtu.edu.cn/~seed/)
• Format: MATLAB .mat files (community BIDS conversion available)

---

Supported Modalities

Modality	Description	Details
EEG	17-channel EEG	ESI NeuroScan, 200 Hz
Eye tracking	Eye movement data	Blinks, gaze position
Peripheral	EOG, EMG	Eye/muscle artifacts

---

SEED-VIG Vigilance Labels

Label	Description	Method
KSS	Karolinska Sleepiness Scale	Self-report (1-9)
EEG-based	Theta/alpha/beta power ratios	Spectral analysis
Binary	Alert vs. Drowsy	Threshold-based

---

BIDS Preparation

Script: scripts/validate_seed_vig.py

Validates SEED-VIG BIDS structure and generates a compliance report.

python skills/seed-vig-skill/scripts/validate_seed_vig.py \
  --input /path/to/SEED-VIG/bids \
  --output /path/to/seed_vig_output/qc/bids_validation.csv

Features:

• BIDS directory structure validation
• Subject completeness check (23 subjects)
• EEG file presence verification
• Vigilance label availability check

---

Core Workflow (Never Bypassed)

1. Identify user target: full SEED-VIG pipeline, feature extraction only, or classification only.
2. Generate a numbered plan with tools, outputs, runtime, storage, and risks.
3. Wait for explicit confirmation (YES / execute / proceed).
4. On confirmation, run BIDS validation using scripts/validate_seed_vig.py.
5. Delegate to eeg-skill for EEG preprocessing.
6. Run scripts/extract_seed_vig_features.py for feature extraction.
7. Run scripts/classify_seed_vig.py for vigilance classification.
8. Save outputs into seed_vig_output/.

---

Standard Output Layout

seed_vig_output/
├── bids/                   # BIDS-staged data (or validation report)
├── eeg/                    # Preprocessed EEG derivatives
├── features/               # Extracted features (band power, DE)
├── classification/         # Vigilance classification results
├── qc/                     # QC summaries
└── logs/                   # Processing logs

---

Benchmark Adapter Guidance

For benchmark-style prompts, do not force the full orchestration when the task only asks for local SEED-VIG data validation.

• If the task starts from SEED-VIG data already present on disk and only asks for BIDS validation:
  • Skip the download stage
  • Default to the narrow path local SEED-VIG discovery -> BIDS validation -> report
• In benchmark mode, do not require explicit confirmation before presenting the validation solution.

---

Safety and Execution Policy

• No execution before explicit plan confirmation.
• All execution must be routed via claw-shell.
• Missing dependencies must be resolved by dependency-planner before running.

---

Important Notes and Limitations

• 17-channel EEG provides limited spatial resolution compared to high-density systems.
• Simulated driving may not fully replicate real-world drowsiness.
• Theta/alpha/beta power ratios are commonly used spectral features for vigilance detection.
• Cross-subject calibration is often needed due to individual differences in EEG patterns.
• seed-vig-skill is orchestration-only; detailed preprocessing logic remains in modality skills.

---

When to Call This Skill

• User asks for end-to-end SEED-VIG workflow.
• User asks to process SEED-VIG EEG data.
• User needs BIDS validation for SEED-VIG data.
• User asks for EEG-based vigilance/fatigue detection analysis.
• User asks for drowsiness detection or alertness monitoring.

---

Complementary / Related Skills

• eeg-skill → EEG preprocessing and feature extraction
• bids-organizer → BIDS validation and organization
• brain-visualization → visualization of derivatives
• dependency-planner → dependency resolution
• conda-env-manager → environment management
• claw-shell → command execution

---

Reference

• SEED-VIG: https://bcmi.sjtu.edu.cn/~seed/
• BCMI Lab, Shanghai Jiao Tong University
• Wei et al. (2017): EEG-based vigilance estimation using extreme learning machines. Neurocomputing.

Created At: 2026-05-06 14:21 HKT Last Updated At: 2026-05-06 14:21 HKT Author: chengwang96