Validation of the ultra short term heart rate variability measures under changing psychophysiological state of the healthy participants at rest

Figures

• 

  Fig. 1. The study design: the first ECG measurement was taken before the supplementation (run 1), the second one − 30 min after the supplementation (run 2), and the third one – 90 min after the supplementation. The supplementation was the energy boost dietary supplement in the study group and the placebo in the placebo group. Each resting state measurement was taken twice – with eyes open (EO), and eyes closed (EC).

• 

  Fig. 2. ECG pre-processing steps and HRV analysis.

• 

  Fig. 3. Mean log HRV measures (calculated using natural logarithm with base Euler) (A) HF power, (B) LF power, (C) nHF power, (D) nLF power, (E) LF/HF ratio, (F) pNN50%, (G) RMSSD, (H) SDSD) for all signal lengths (10 s, 30 s, 1 min, 2 min, 3 min, 4 min, 5 min) from run 1 (blue), run 2 (red) and run 3 (green), EC condition. Vertical bars denote 95% confidence intervals. Asterisks show significant differences between the lengths and the reference 5-min length for each run, with the colors representing each run.

• 

  Fig. 4. The changes of mean log HRV measures (calculated using natural logarithm with base Euler) across time (run 1–3) for all HRV measures for all signal lengths (10 s, 30 s, 1 min, 2 min, 3 min, 4 min, 5 min), for the placebo (left column) and study (right column) groups, from the EC condition. The measures from top to bottom: RMSSD, SDSD, pNN50%, LF power, HF power, nLF power, nHF power, and LF/HF ratio. Vertical bars denote 95% confidence intervals.

• 

  Fig. 5. (A) Bland-Altman plots with the difference relative to the benchmark 5-min reference recording, for frequency measures (HF power, LF power, nHF power, nLF power, LF/HF ratio) for the signal lengths of 30 s (1st column), 1 min (2nd column), and 2 min (3rd column), from run 1, EC condition. Blue line: bias, red lines: ±1.96 SD.(B) Bland-Altman plots with the difference relative to the benchmark 5-min reference recording, for statistical measures (RMSSD, SDSD, and pNN50%) for signal lengths of 30 s (1st column), 1 min (2nd column), and 2 min (3rd column), from run 1, EC condition. Blue line: bias, red lines: ±1.96 SD.

• 

  Fig. 6. Mean log HRV measures (calculated using natural logarithm with base Euler) (A) HF power, (B) LF power, (C) nHF power, (D) nLF power, (E) LF/HF ratio, (F) pNN50%, (G) RMSSD, (H) SDSD) for all signal lengths (10 s, 30 s, 1 min, 2 min, 3 min, 4 min, 5 min) from run 1 (blue), run 2 (red) and run 3 (green), EO condition. Vertical bars denote 95% confidence intervals. Asterisks show significant differences between the lengths and the reference 5-min length for each run, with the colors representing each run.

• 

  Fig. 7. The changes of mean log HRV measures (calculated using natural logarithm with base Euler) across time (run 1–3) for all HRV measures for all signal lengths (10 s, 30 s, 1 min, 2 min, 3 min, 4 min, 5 min), for the placebo (left column) and study (right column) groups, from the EO condition. The measures from top to bottom: RMSSD, SDSD, pNN50%, LF power, HF power, nLF power, nHF power, and LF/HF ratio. Vertical bars denote 95% confidence intervals.

• 

  Fig. 8. (A) Bland-Altman plots (calculated as a difference relative to the benchmark 5-min reference recording) for frequency measures (HF power, LF power, nHF power, nLF power, LF/HF ratio) for the signal lengths of 30 s (1st column), 1 min (2nd column), and 2 min (3rd column), from run 1, EO condition. Blue line: bias, red lines: ±1.96 SD. (B) Bland-Altman plots (calculated as a difference relative to the benchmark 5-min reference recording) for statistical measures (RMSSD, SDSD, and pNN50%) for signal lengths of 30 s (1st column), 1 min (2nd column), and 2 min (3rd column), from run 1, EO condition. Blue line: bias, red lines: ±1.96 SD.