mdnunez/sozhfo

Python, MATLAB, and JAGS code associated with the preprint, "Beyond rates: Time-varying dynamics of high frequency oscillations as a biomarker of the seizure onset zone." (2021)

Citation

Nunez, M. D., Charupanit, K., Sen-Gupta, I., & Lopour, B. A., Lin, J. J. (2022). Beyond rates: Time-varying dynamics of high frequency oscillations as a biomarker of the seizure onset zone. Journal of Neural Engineering. 19(1), 016034.

sozhfo

(Repository version 0.3.1)

Authors: Michael D. Nunez, Krit Charupanit, Indranil Sen-Gupta, Beth A. Lopour, and Jack J. Lin from the University of California, Irvine

Prerequisites

MATLAB

MCMC Sampling Program: JAGS

Scientific Python libraries

Python Repository: pyjags

Downloading

The repository can be cloned with git clone https://github.com/mdnunez/sozhfo.git

The repository can also be may download via the Download zip button above.

Data availability

Automatically identified HFO and qHFO counts per second, standardized delta (1-4 Hz) power, channel localization labels, and samples from posterior distributions for Model 2 and Model 3 are available upon request and on Figshare. Samples from posterior distributions for Model 1 are available upon request.

Processing Steps

1. hfo_extractHFO.m (Extract HFO candidates)
2. hfo_extractqHFO.m (Find likely artifact HFOs)
3. hfo_fitmodel14.py (Model 1 in Paper)
4. hfo_fitmodel12.py (Model 2 in Paper)
5. hfo_fitmodel8.py (Model 3 in Supplementals)
6. hfo_sleepeval.py (Evaluate relationship of model states to sleep stage and/or delta power)
7. hfo_SOZprediction_avg_nomixture.py (Evaluate prediction of SOZ by HFO rate, CV, and clumping parameters from Model 1)
8. hfo_SOZprediction_avg_nomix_boot.py (Generate ''Strong Prediction'' baseline through label mixing with parameters from Model 1)
9. hfo_SOZprediction_avg.py (Evaluate prediction of SOZ by parameters from Model 2)
10. hfo_SOZprediction_avg_boot.py (Generate ''Strong Prediction'' baseline through label mixing with parameters from Model 2)

License

sozhfo is licensed under the GNU General Public License v3.0 and written by Michael D. Nunez, Krit Charupanit, Indranil Sen-Gupta, Beth A. Lopour, and Jack J. Lin from the University of California, Irvine.