GCFD Coherence Tracker
Generalized Cross-Frequency Decomposition for EEG/MEG Phase Synchronization Analysis
v1.0.0 | Built by TAURUS AI Corp | Global Bio-Foundry Initiative | Apache 2.0 License
Clinical Preset
Pre-configured signal parameters for common clinical states
Signal Parameters
Frequency Bands
Upload EEG Data (optional)
GCFD Coherence Tracker — Technical Documentation
What is GCFD?
Generalized Cross-Frequency Decomposition is a method for quantifying the phase relationship between neural oscillations at different frequencies. The core metric is the Phase Locking Value (PLV), which measures the consistency of the phase difference between two band-filtered signals.
Why Cross-Frequency Coupling Matters
Neural oscillations at different frequencies serve distinct computational roles:
- Theta (4-8 Hz): Memory encoding, spatial navigation, top-down control
- Gamma (30-100 Hz): Local cortical processing, feature binding, attention
The coupling between theta and gamma (theta-gamma PAC/PLV) is a biomarker for:
- Working memory capacity
- Cognitive function in aging
- Severity of neuropsychiatric conditions (MDD, ADHD, Alzheimer's)
- Depth of anesthesia / consciousness level
Signal Processing Pipeline
Raw EEG/MEG Signal
|
v
[Butterworth Bandpass Filter] ─── Order 3, zero-phase (filtfilt)
| Band 1: e.g., Theta (4-8 Hz)
| Band 2: e.g., Gamma (30-100 Hz)
v
[Hilbert Transform] ─────────── Extract analytic signal
|
v
[Phase Extraction] ──────────── np.angle(analytic_signal)
|
v
[Phase Locking Value] ────────── PLV = |mean(exp(j * (phase1 - phase2)))|
|
v
[Sliding Window Analysis] ───── 2s window, 0.5s step for temporal dynamics
|
v
[Global Coherence Score] ────── Mapped to [0.5, 1.0] clinical scale
Clinical Interpretation Scale
| Score Range | Classification | Interpretation |
|---|---|---|
| 0.90 - 1.00 | HEALTHY | Strong cross-frequency coupling |
| 0.70 - 0.89 | MODERATE | Partial synchronization |
| 0.50 - 0.69 | LOW | Weak coupling / potential pathology |
Note: Scores above 0.95 may indicate pathological hypersynchronization (e.g., epileptic activity).
CSV Upload Format
Upload EEG data as CSV with:
- Rows: Time samples
- Columns: Channels (first column used if multiple)
- No time column needed — timing derived from sampling rate parameter
- Units: Microvolts (uV) preferred, but any consistent unit works
Limitations
- This tool uses synthetic EEG by default — clinical use requires real EEG data
- PLV is sensitive to signal length and SNR — minimum 5 seconds recommended
- Single-channel analysis only (multi-channel averaging planned for v2.0)
- Phase-amplitude coupling (PAC) via Modulation Index not yet implemented
References
- Lachaux, J.P. et al. (1999). Measuring phase synchrony in brain signals. Human Brain Mapping, 8(4), 194-208.
- Canolty, R.T. et al. (2006). High gamma power is phase-locked to theta oscillations in human neocortex. Science, 313(5793), 1626-1628.
- Tort, A.B. et al. (2010). Measuring phase-amplitude coupling between neuronal oscillations of different frequencies. Journal of Neurophysiology, 104(2), 1195-1210.
- Goodman, M.S. et al. (2018). Theta-gamma coupling and working memory in Alzheimer's disease and mild cognitive impairment. Frontiers in Aging Neuroscience, 10, 101.
- Sun, L. et al. (2022). Theta-gamma coupling deficit in MDD patients during working memory. Brain Research Bulletin, 189, 49-57.
API Access
This Space exposes a REST API via Gradio's built-in endpoint system.
Python Client
from gradio_client import Client
client = Client("Taurus-Ai-Corp/gcfd-coherence-tracker")
result = client.predict(
preset="Healthy Adult",
duration=10,
fs=250,
theta_amp=1.0,
gamma_amp=1.5,
noise_level=0.8,
seed=42,
theta_low=4.0,
theta_high=8.0,
gamma_low=30.0,
gamma_high=100.0,
csv_file=None,
api_name="/run_analysis"
)
cURL
curl -X POST https://Taurus-Ai-Corp-gcfd-coherence-tracker.hf.space/api/predict \
-H "Content-Type: application/json" \
-d '{"data": ["Healthy Adult", 10, 250, 1.0, 1.5, 0.8, 42, 4.0, 8.0, 30.0, 100.0, null]}'
Rate Limits
| Tier | Rate | Access |
|---|---|---|
| Free (this Space) | 50 requests/day | Open |
| Researcher ($29/mo) | 1,000 requests/day | Contact us |
| Clinical ($149/mo) | 10,000 requests/day + batch | Contact us |
| Enterprise | Unlimited + SLA | Contact us |
Upcoming Features (v2.0)
- Multi-channel analysis with topographic maps
- Phase-Amplitude Coupling (PAC) via Modulation Index
- EDF/BDF file format support
- Batch processing endpoint
- Pre-trained EEG foundation model integration (braindecode/SignalJEPA)
About GCFD Coherence Tracker
Version: 1.0.0 Author: TAURUS AI Corp License: Apache License 2.0 Repository: GitHub
The Global Bio-Foundry Initiative
TAURUS AI Corp's neuroscience research initiative explores the intersection of:
- Neural coherence in biological systems
- Cross-frequency coupling dynamics in EEG/MEG signals
- Open-source neuroscience tooling for clinical research
The GCFD Coherence Tracker is the open-source component of this initiative, providing standard DSP tools for EEG/MEG phase synchronization analysis.
Citation
If you use this tool in your research, please cite:
@software{gcfd_tracker_2026,
title = {GCFD Coherence Tracker: Cross-Frequency Phase Synchronization Analysis},
author = {TAURUS AI Corp},
year = {2026},
version = {1.0.0},
url = {https://huggingface.co/spaces/Taurus-Ai-Corp/gcfd-coherence-tracker},
license = {Apache-2.0}
}
Related Work
- braindecode — Deep learning for EEG/MEG
- MNE-Python — MEG/EEG analysis toolkit
- SignalJEPA — Brain signal foundation model
Contact
- Research inquiries: admin@taurusai.io
- Enterprise licensing: admin@taurusai.io
- Issues: GitHub Issues