Actiheart User Manual 5.1.31 (Jun 2022)
32
lost data. In those epochs where recovery fails the AEE is solely based on
Activity.
There are three steps in the cleaning process:
STEP 1:
Cleaning
This involves setting suspect values to zero according to the following criteria:
•
If HR < 30 or
•
If HR > 30 and the rate of change of heart rate is > 100, 132 & 160
BPM/Min for a 1 min, 30 sec & 15 second epoch respectively.
STEP 2:
Removing
Spurious
signals from
the cleaned
data
The cleaned data is analysed minute-by-minute and if the cleaned HR at any
data point is >1.75 * Filtered HR then the cleaned HR is set to zero. Filtered HR
is the average of the HR over the 4 minutes preceding the data point being
analysed. Between 4 and 16 HR values will be averaged depending on the
epoch.
STEP 3:
Replacing
Removed
Data
a)
Recovery
For each minute where the HR has been set to zero each stored IBI minimum
and maximum (there will be minimum of sixteen, if the epoch is 15 sec) is used to
calculate a HR (60,000/ms) which is then compared with the previous valid heart
rate. The calculated HR which is closest to the valid heart rate and within 30
BPM’s of it is used to replace the zero values. This is termed “Recovered Data”.
If the difference is > 30 BPM the calculated HR is not used to replaced the zero
value.
b)
Interpolation
Following the recovery process any remaining zero value gaps of < 5 minutes
are filled by interpolation (straight line join). If the gap is greater than 5 minutes
the value is left at zero.
If there are gaps in the data (see below) due to the unit not being worn the
cleaning is disabled in the gaps.
4.2.8 Gaps in
the Data
Whenever a file is opened from the Advanced EE or Daily EE screen an attempt
is made to identify gaps in the data in each day due to the unit not being worn
and the user is offered the option of filling these gaps with an average of the
