It is a MATLAB toolbox for estimating both nonfractal connectivity and fractal connectivity from a set of time series with long-range dependence such as resting state fMRI BOLD signals.
Wonsang You ([email protected])
This toolbox was designed to work from MATLAB version 7.7 (R2008b) to the latest version with the Statistics toolbox. You can install this toolbox as follows.
bfn_demo_nonfractal". This demo shows an example of estimating nonfractal connectivity of resting state fMRI BOLD signals of the rat brain using the maximum likelihood estimator.
Let X be a NxQ matrix of Q time series with length N. It might be a set of BOLD signals corresponding to multiple ROIs of the brain which was extracted from DICOM images. Then, the function "
bfn_mfin_ml" estimates the Hurst exponent, nonfractal connectivity, and fractal connectivity in multivariate time series with long memory using the maximum likelihood method as follows.
[H, nfcon, fcon] = bfn_mfin_ml(X);
H is the Hurst exponent, nfcon and fcon denote the nonfractal connectivity and fractal connectivity respectively. Nonfractal connectivity is identical to the correlation matrix of short memory while fractal connectivity is defined as the asymptotic wavelet correlation of bivariate long memory processes. On the other hand, the function "
bfn_mfin_lms" uses the least-mean squares (LMS) method to estimate all the above parameters.
[H, nfcon, fcon] = bfn_mfin_lms(X);
For more options, run the commands "
help bfn_mfin_ml" or "
Please cite the following paper when using this toolbox.
Wonsang You, Sophie Achard, Joerg Stadler, Bernd Bruekner, and Udo Seiffert, "Fractal analysis of resting state functional connectivity of the brain," in 2012 International Joint Conference on Neural Networks, 2012.