Supplemental Figure 3: Extraction of resting ECG. 11 am resting ECG Supplemental Figure 3:...
1
09:00 12:00 0 2 4 0 0.5 0.5 1 1.5 2 2.5 0 2 4 6 8 10 −20 0 20 40 60 11:00 10:00 Freq (Hz) 09:50 10:00 10:10 1 1.5 2 Freq (Hz) 0 0.5 1 SD Freq 1 Freq (Hz) SD Frequency Time of day Time of day A B C Time (s) Voltage (μV) D E 100 ms 20 μV Optimised resting ECG Typical 11 am resting ECG Supplemental Figure 3: Extraction of resting ECG. A) R peak frequency in the period between 8am and12pm for a typical participant. Beats with a frequency above 2Hz (120bpm, red dashed line) were considered non-physiological and excluded from further analysis. Within the entire patient dataset, the region between 9.50am and 10.10am showed the lowest SD of RR interval on average. This region was therefore selected for extraction of a typical ‘resting ECG’ from each participant. B-C) The 10s with the lowest frequency and standard deviation (red asterisk) between 9:50am and 10:10am was selected for each participant, ensuring we extracted the ‘best’ resting ECG in terms of stability and resting heart rate (average heart rate was 67 ± 0.6 bpm or 1.12 ± 0.01 Hz (SEM; n=385)). D) 10s of Holter ECG activity selected according to the criteria identified in B-C. E) An averaged beat from the 10s of data presented in D (optimised resting ECG, Blue) compared to an averaged beat taken at random from 10s of data at 11am.
Transcript of Supplemental Figure 3: Extraction of resting ECG. 11 am resting ECG Supplemental Figure 3:...
09:00 12:000
2
4
0 0.50.5
1
1.5
2
2.5
0 2 4 6 8 10
−200
204060
11:0010:00
Freq
(Hz)
09:50 10:00 10:101
1.5
2
Freq
(Hz) 0
0.5
1
SD F
req
1
Freq
(Hz)
SD Frequency
Time of day
Time of day
A
B
C
Time (s)
Vol
tage
(μV
)
D
E
100 ms20 μV
Optimised resting ECGTypical 11 am resting ECG
Supplemental Figure 3: Extraction of resting ECG. A) R peak frequency in the period between 8am and12pm for a typical participant. Beats with a frequency above 2Hz (120bpm, red dashed line) were considered non-physiological and excluded from further analysis. Within the entire patient dataset, the region between 9.50am and 10.10am showed the lowest SD of RR interval on average. This region was therefore selected for extraction of a typical ‘resting ECG’ from each participant. B-C) The 10s with the lowest frequency and standard deviation (red asterisk) between 9:50am and 10:10am was selected for each participant, ensuring we extracted the ‘best’ resting ECG in terms of stability and resting heart rate (average heart rate was 67 ± 0.6 bpm or 1.12 ± 0.01 Hz (SEM; n=385)). D) 10s of Holter ECG activity selected according to the criteria identi�ed in B-C. E) An averaged beat from the 10s of data presented in D (optimised resting ECG, Blue) compared to an averaged beat taken at random from 10s of data at 11am.
