MATLAB File Help: msentropy View code for msentropy WFDB Contents
msentropy



function varargout=msentropy(varargin)

 [y,scale,info]=msentropy(x,dn,dm,dr,N,N0,minM,maxM,maxScale,minR,maxR)

    Wrapper to the Multiscale Entropy C code written by Madalena Costa (mcosta@fas.harvard.edu):
         http://physionet.org/physiotools/mse/mse-1.htm

 Calculates the multi scale entropy of a signal 'x'. A tutorial on Mulsticale
 entropy is available at:
 http://www.physionet.org/physiotools/mse/tutorial/


 Please cite these publications when referencing this material:
     Costa M., Goldberger A.L., Peng C.-K. Multiscale entropy analysis of biological signals. Phys Rev E 2005;71:021906.
     Costa M., Goldberger A.L., Peng C.-K. Multiscale entropy analysis of physiologic time series. Phys Rev Lett 2002; 89:062102.

 Also include the standard citation for PhysioNet:
     Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG,
     Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. Circulation 101(23):e215-e220 [Circulation Electronic Pages; http://circ.ahajournals.org/content/101/23/e215.full]; 2000 (June 13)

 Readers of may also wish to read:
     Costa M, Peng C-K, Goldberger AL, Hausdorff JM. Multiscale entropy analysis of human gait dynamics. Physica A 2003;330:53-60.

 Required Parameters:

 x
       Nx1 vector of doubles in which to caculate the multiscale entropy.

 Optional Parameters are:
 dn
       1x1 double. Sets the scale increment to dn (1-40; default: 1).
 dm
       1x1 double. Sets the m increment to dm (1-10; default: 1).
 dr
       1x1 double. Sets the scale increment to dr (>0; default: 0.05).
 N
       1x1 integer. Stop the analysis with row N.
       By default, analysis ends at row 39999, or at the end of the data set if there are fewer rows.
 N0
       1x1 integer. Begin the analysis with row N0.
       By default, analysis begins with row 1.
 minM
       1x1 integer betwee 1-10. Set the minimum pattern length for SampEn to minN (1-10; default: 2).
 maxM
        1x1 integer betwee 1-10. Set the maximum m to maxM (1-10; default: 2).
 maxScale
        1x1 integer betwee 1-40. Set the maximum scale for coarse-graining to maxScale (1-40; default: 20).
 minR
        1x1 double >0. Set the minimum similarity criterion for SampEn to minR (>0; default: 0.15).
 maxR
        1x1 double > 0. Set the maximum m to maxR (>0; default: 0.15).


 Outputs:
 y
       A 1xM vector of doubles corresponding to estimated sample entropies at each scale.
 scale
       A 1xM vector of integers specifying the scales in which 'y' was
       estimated.

 info
       An optional 3x1 cell array of strings providing loggin and verbose information from
       the calculation.

 Wrapper written by Ikaro Silva, 2013
 Last Modified: March 20, 2014
 Version 0.0.1

 Since 0.9.5

 %Example
 N=30000;
 noise=randn(N,1);
 maxScale=10;
[entropyNoise,scale1]=msentropy(noise,[],[],[],[],[],[],[],maxScale);
 %Simulate determistic system with noise-like 2nd order statistics
 nlinear=zeros(N,1);nlinear(1)=0.2;u=4;
 for n=2:N;nlinear(n)=u*nlinear(n-1)*(1-nlinear(n-1));end
[entropyDeterm,scale2]=msentropy(nlinear,[],[],[],[],[],[],[],maxScale);
subplot(2,1,1);
plot(noise(1:1000));hold on;grid on;plot(nlinear(1:1000),'r');legend('Stochastic','Deterministic')
subplot(2,1,2);
plot(scale1,entropyNoise);hold on;grid on;plot(scale2,entropyDeterm,'r');legend('Stochastic','Deterministic')


 See also SURROGATE, DFA, WFDBDESC, PHYSIONETDB, RDANN, ANN2RR, MAPRECORD