Ex-Search TI Optimization Pipeline

The Ex-Search module provides an exaustive search approach for Temporal Interference (TI) simulations. The system features unified logging, multiple EEG net support, flexible leadfield management.

Overview

Ex-Search utilizes leadfield-based optimization to determine optimal electrode configurations for TI stimulation based on:

Multiple EEG Nets: Support for high-density GSN-HydroCel (185, 256), and regular EGI 10-20 nets
ROI Analysis: fixed 3mm sphere sampling with both TImax and TImean calculations
Field Metrics: Volume-weighted analysis with percentiles and focality measures
Unified Logging: Professional pipeline tracking in single log file

User Interface

The interface provides comprehensive controls for:

Subject Selection: Choose from available subjects with automatic leadfield scanning
Leadfield Management: View existing leadfields, create new ones, and show electrode configurations
ROI Configuration: Add target regions with coordinates and manage multiple ROIs
Electrode Setup: Configure E1+, E1-, E2+, E2- with support for both GSN and 10-20 formats
Execution Control: Run optimization with real-time progress tracking

Supported EEG Networks

Ex-Search automatically detects and supports multiple EEG electrode configurations:

Standard 10-20 System

EGI 10-20 Okamoto 2004 electrode configuration - widely used standard with 32 electrodes

High-Density GSN System

GSN-HydroCel 256 electrode configuration - high-density net for precise targeting

Example: TI Field Analysis Pipeline

We demonstrate the Ex-Search pipeline capabilities using a representative TI stimulation analysis targeting the left insula region with F7/T7 and T8/Fz electrode pairs.

Analysis Setup

Subject: 101
EEG Net: EEG10-20_Okamoto_2004
Target ROI: Left insula region
Electrode Configuration: F7_T7 <> T8_Fz
Current: 1mA (default)
Analysis: Enhanced 3mm sphere ROI sampling

Field Visualization

Cortical Surface Distribution: TI field distribution visualized on the cortical surface showing spatial targeting and field intensity

Statistical Analysis

Volume-Weighted Field Distribution: Volume-weighted field distribution histogram with focality thresholds and ROI indicators

Quantitative Results:

Electrode Configuration	TImax ROI	TImean ROI	Peak Field	95th %	99th %	99.9th %	Focality 50%	Focality 75%	Focality 90%	Focality 95%	Peak Location
F7_T7 <> T8_Fz	0.076 V/m	0.062 V/m	0.361 V/m	0.154 V/m	0.200 V/m	0.246 V/m	81.9 mm²	13.4 mm²	2.9 mm²	1.6 mm²	(-45.12, 34.09, -40.45)

Key Features

Enhanced ROI Analysis

3mm Sphere Sampling: 20-point sampling around ROI coordinates for statistical analysis
Metrics: Both TImax (peak) and TImean (average) values in target region

Field Distribution Analysis

Volume-Weighted Statistics: Percentile calculations (95%, 99%, 99.9%) accounting for tissue volume
Focality Metrics: Threshold-based analysis (50%, 75%, 90%, 95% of 99.9 percentile)
Peak Localization: XYZ coordinates of maximum field intensity

Professional Output

Unified Logging: Complete pipeline execution tracking in single timestamped log file
Comprehensive CSV: All metrics combined in final_output.csv for easy analysis
Visualization: Field distribution histograms with ROI indicators and focality cutoffs

Pipeline Workflow

1. EEG Net Selection

[INFO] Scanning available EEG nets for subject 101...
EGI10-10_Cutini_2011.csv
EGI10-10_UI_Jurak_2007.csv  
EGI10-20_Okamoto_2004.csv
GSN-HydroCel-185.csv        # Default selection
GSN-HydroCel-256.csv
easycap_BC_TMS64_X21.csv

2. Leadfield Management

Intelligent Detection: Automatic scanning of existing leadfields
Flexible Creation: Generate leadfields for any supported EEG net
Performance Optimization: Efficient loading of large matrices (2-20GB)

3. Analysis Pipeline

ROI Sampling: Enhanced sphere-based analysis around target coordinates
Field Analysis: Volume-weighted statistical calculations
Visualization: Automatic histogram generation with professional formatting

Performance Characteristics

Processing Times

EEG Network	Leadfield Size	Loading Time	Simulation Time
GSN-HydroCel-185	~16GB	2-3 minutes	3-5 seconds
GSN-HydroCel-256	~25GB	4-6 minutes	5-8 seconds
EGI10-20	~2TGB	30-60 seconds	2-3 seconds

File Structure

Input Organization

derivatives/SimNIBS/sub-{subject}/
├── m2m_{subject}/
│   ├── eeg_positions/           # EEG net configurations
│   │   ├── GSN-HydroCel-185.csv
│   │   ├── GSN-HydroCel-256.csv
│   │   └── EGI10-20_Okamoto_2004.csv
│   └── ROIs/                    # Target regions
│       ├── roi_list.txt
│       └── L_Insula.csv
├── leadfield_vol_GSN-HydroCel-185/   # Leadfield matrices
│   └── leadfield.hdf5
└── leadfield_vol_EGI10-20_Okamoto_2004/
    └── leadfield.hdf5

Output Structure

derivatives/SimNIBS/sub-{subject}/ex-search/
├── L_Insula_GSN-HydroCel-185/      # ROI-specific results
│   ├── analysis/
│   │   ├── final_output.csv        # Complete results
│   │   ├── summary.csv            # Field analysis metrics
│   │   ├── *_histogram.png        # Visualizations
│   │   └── mesh_data.json         # Raw analysis data
│   └── TI_field_*.msh             # Simulation meshes
└── derivatives/logs/sub-{subject}/
    └── ex_search_20250122_143012.log  # Unified pipeline log

Last Updated: June 2025
Version: 2.0.4 (Unified Logging + Enhanced EEG Net Support)

Wiki Guides