Entropy measures are used to quantify the complexity of a signal, which can be understood as the level of unpredictability of a signal. If you consider that some biological process modify the level of complexity of a signal, you could try to measure such change by means of a entropy measure. Some nonlinear measures are also able to measure similar changes to those described above, but most of the time, pure nonlinear features are based on the assumption that the underlying system producing the signal under study is completely deterministic and such assumptions can be inadequate in several cases.

Dear Reza Saber, Go through the article below. This article will surely answer your querry.

http://www.gao.ece.ufl.edu/my_paper/NonlinearDyn12_Entropy_BioSig_GAO_Final_EB.pdf

Hi,

It is well-accepted today that biological signals are nonlinear in nature and uncertainty and imprecision are the inherent characteristics of them. On the other hand, it is also postulated that a number of pathologies are characterized by a breakdown of the complex non-linear dynamics observed in healthy time series, which results in smaller variability and sometimes the emergence of periodicity. As a result, it is conceivable to obtain useful information even from healthy or diseased time series by using entropy analysis.

Regards,

Thank you all for your great answers.

Dear Fariba,

As you mentioned pathologies could be characterized by the means of nonlinear dynamics and emergence of periodicity is one of the results of illness. I want to know how nonlinear dynamics of system (time series) changes in a non-periodic movements. movements that are not like walking and running. Do you have any idea?

Entropy can be an useful measure, but not in isolation.

Tremors are one example of movements that are periodic, but can be imposed upon non-periodic activities.

Hi,

From my point of view, it strongly depends on your case and study design. F ex, think of a non-periodic movement, which is fatiguing. The fatigue phenomenon can influence the value of entropy. Moreover, different levels of contraction in your desired movement, and the number of recruited motor units can change the entropy, as well. Hope I understood your question, correctly.

Yours,

Dear professor Afaq Ahmad,

Thank you very much for the paper. It was very helpful.

Hello,

Interesting subject! Are different groups of people, all healthy?

According to your statements, I infer that you are investigating the morphology of subjects' jumping with no EMG recording. Am I right?

Bests,

Do you want to compare the difference between two situations for the same subjects group? I researched the statistical stability on entropies, like ApEn, SamEn and MEn.

Dear Reza,

May I ask some questions for more clearly:

1, what signal you use?

2, which groups?

3, the aim of your reseatch is?

Thanks

Using entropy method need to pro-set some paremeters: the sequecne length, space dimention m and threshold paremeter r. Sequence length N is at lesat several hundreds points and it is essential for reliable entropy values. If you want to meassure the relationship between two seuqences, you could use the cross-entropy.

It is possible to measure the Entropy from a non-linear analysis tool which use the chaos theory approach. The technique is called RQA (Recurrence Quantification Analysis) and I'm one of the gratefull users of relative technique for studying HRV (Heart Rate Variability during Tilt-Test) and surface EMG (sEMG). In can send you some details if you send me a request to my E-mail [email protected]