Garner, David M.de Souza, Naiara Maria [UNESP]Valenti, Vitor E. [UNESP]Vanderlei, Luiz Carlos M. [UNESP]2021-06-252021-06-252019-10-24Romanian Journal of Diabetes, Nutrition and Metabolic Diseases, v. 26, n. 3, p. 279-291, 2019.1583-8609http://hdl.handle.net/11449/206267Backgound and aims. In this study we analyzed heart rate variability (HRV) via chaotic global techniques so as to discriminate diabetics from control subjects. Matherial and method. Chaotic global analysis of the RR-intervals from the electrocardiogram and pre-processing adjustments were undertaken. The effect of varying two parameters to adjust the Multi-Taper Method (MTM) power spectrum were evaluated. Then, cubic spline interpolations from 1Hz to 13Hz were applied whilst the spectral parameters were fixed. Precisely 1000 RR-intervals of data were recorded. Results. CFP1 and CFP3 are the only significant combinations of chaotic globals when the default standard conditions are enforced. MTM spectral adjustments and cubic spline interpolation are trivial at effecting the outcome between the two datasets. The most influencial constraint on the outcome is data length. Conclusion. Chaotic global analysis was offered as a reliable, low-cost and robust technique to detect autonomic dysfunction in subjects with diabetes mellitus.279-291engCubic spline interpolationDiabetes mellitusDPSSMulti-taper methodThomson's nonlinear combinationArtigo10.2478/rjdnmd-2019-00292-s2.0-85105030547