Multifractal Analysis Reveals Decreased Non-linearity and Stronger Anticorrelations in Heart Period Fluctuations of Fibromyalgia Patients

Figures

• 

  Figure 1. Representative RR time series during timeday of a (A) healthy subject and (B) patient with FM. Plot of (C) Fq(s) vs. q of a RR sequence (healthy subject) for several values of q.

• 

  Figure 2. Results of the MFDFA for all healthy and FM subjects. (A) Plot of h(q) vs. q. The values of h(q) are derived from fittings within the interval 6 < s < 100 and a time window of 1, 500 RR values during daytime. The estimated regression models were h(q) = 1.295 − 0.039q + 0.003q2 (for FM patients), and h(q) = 1.261 − 0.042q + 0.002q2 (for healthy group). In both models, β0, β1, and β2 ≠ 0 with p < 0.001 and R2 > 0.55. The confidence intervals of β1 and β2 are overlapped in both groups, while there was no overlap in β0 of the FM group (1.281, 1.309) with the healthy group (1.226, 1.256). (B) Behavior of τ(q) for data showed in (A). (C) Multifractal spectrum F(α) vs α. Error bars represent the standard error of the mean.

• 

  Figure 3. (A) Average behavior of h(q) vs. q for healthy subjects, patients with FM and their corresponding shuffled version. (B) As in (A) but for the phase randomized data. The exponents were calculated for the scaling region 6 < s < 100. (C) Multifractal spectrum F(α) vs α for healthy and FM groups and their corresponding shuffled version. (D) As in (C) but for the phase randomized data, all during timeday. Error bars represent the standard error of the mean.

• 

  Figure 4. Multifractal detrended fluctuation analysis of magnitude increment sequences of healthy and FM patients during timeday. (A) Behavior of h(q) vs. q for scaling exponents obtained from fittings in the scaling region 15 < s < 100. F(α) vs. α for data showed in (B). Error bars represent the standard error of the mean.

• 

  Figure 5. Multifractal detrended fluctuation analysis of sign time series from healthy and patients with FM during timeday.(A) Plot of h(q) vs. q for scaling exponents obtained from fittings in the scaling region 15 < s < 100. (B) F(α) vs. α for data showed in (A). Error bars represent the standard error of the mean.

• 

  Figure 6. (A) Scatter plot of the generalized Hurst exponents hRR(q) vs. hmag(q), for healthy and FM groups during daytime periods. For both groups, symbols with error bars represent the mean value, and the error bars the corresponding standard error of the mean. We observe that both groups are well separated for the different values of q. (B) As in (A) but for the generalized Hurst exponents hRR(q) vs. hsign(q), for healthy and FM groups during daytime periods. In this case, both groups are close each other for negative moments while tend to be more separated as q increases.