1
The AMPS 12-lead ECG Extractor
Antares
2
Antares
The AMPS 12-lead ECG Extractor
Antares is an Holter-based tool for
the extraction of 12 lead ECG strips.
Interfaced with several commercial
systems
Usable as a standalone tool or
embedded into CRO software workflow,
thanks to its rich API library
3
A Smart 12-lead ECG Extractor
The ECGs snapshots extracted for each
time-point are chosen among the best
within an optimum window.
Best means the lowest-noise strips among
those with only normal beats and
preceded by stable heart rate.
If (within each time-point) more than 1
snapshot is extracted, the second best
(strictly non-overlapping) is extracted,
then the third, and so on.
4
t
Optimum window
1° extraction (best ECG)
2° extraction (2° best ECG)
(n-1)
th
timepoint n
th
timepoint (n+1)
th
timepoint
0
200
400
600
800
1000
1200
1400
08.27.01
08.27.12
08.27.21
08.27.32
08.27.40
08.27.48
08.27.57
08.28.06
08.28.16
08.28.26
08.28.37
08.28.47
RRs
RRo
Heart Rate
noise
Among stable strips the one
with minimal noise is chosen!!
Antares Philosophy
5
Each individual beat determines a
potential extraction
n
th
timepoint
Number of potential candidate extractions depends on heart
rate (# of beats available in the optimum window).
If multiple extractions are requested, the selected extractions
will NOT overlap (no single beat will be used more than once).
A single abnormal beat will determine the exclusion of all the
potential extractions including that beat.
t
Extraction criteria
6
Parametric extraction features
Length of extracted ECGs
Scalable output (sampling rate and
amplitude resolution)
Number of ECGs to be extracted for each
selected time-point
Duration of optimum window (to be used
for best selection)
Time-point rule of selection (fixed distance
or from external file)
7
Flexible duration of ECGs snapshots
Time-point resolution
Optimum window size
# of ECGs to be extracted per each time-point
Sampling rate of output ECG is a variable!
External file with extraction times
Extraction settings
8
0
200
400
600
800
1000
1200
1400
1600
1800
12.42.01
12.42.10
12.42.19
12.42.28
12.42.37
12.42.45
12.42.54
12.42.59
12.43.06
12.43.11
12.43.18
12.43.25
12.43.33
12.43.43
12.43.52
RRs
RRo
Expected timepoint: 12:43:00
Extracted timepoint: 12:42:28
Example of
Heart Rate selection
9
RR = 850
QT = 382
QTcB = 414
QTcF = 403
changes on selection
ECGs 2 minutes apart!!
RR = 672
QT = 350
QTcB = 427
QTcF = 400
Effect of HR
10
0
200
400
600
800
1000
1200
1400
1600
1800
09.37.01
09.37.08
09.37.16
09.37.24
09.37.32
09.37.40
09.37.47
09.37.56
09.38.04
09.38.12
09.38.20
09.38.28
09.38.36
09.38.43
09.38.51
09.38.59
RRs
RRo
Expected timepoint: 09:38:00 Extracted timepoint: 09:38:32
on selection
Effect of noise
11
Flexible Time-point selection
Time-points can be chosen at fixed
distance (time-point resolution)
or
A table of extraction points can be
provided in an external file
12
Example of beat-to-beat Antares
log report
sample time N/A RRs RRo Noise
156906 10:29:01 N 16/0 635 737 0.107
157039 10:29:02 N 16/0 665 736 0.106
157177 10:29:02 N 16/0 690 734 0.106
157320 10:29:03 N 16/0 715 731 0.106
157477 10:29:04 N 15/0 785 728 0.103
157646 10:29:05 N 15/0 845 731 0.103
157820 10:29:06 N 15/0 870 731 0.102
157995 10:29:06 N 15/0 875 731 0.099
158165 10:29:07 N 14/0 850 733 0.097
158335 10:29:08 N 14/0 850 736 0.097
...........
175353 10:30:33 N 12/0 800 816 0.082
175513 10:30:34 N 12/0 800 815 0.082
175668 10:30:35 N 13/0 775 816 0.084
Choosen ECG at 175513 10:30:34
RR interval in
the strip
RR interval in
observation
period (30” or 1
minute)
Each row
represents
the strip of N
seconds
starting at a
given beat……
Each normal beat
determines a
separate
candidate strip
“Unstable” strip
(RRs RRo is
large)
13
Unique Antares features that you won’t
find in other extractors (1)
Works seamlessly across different
platforms.
Performs an optimum selection taking in
account:
noise
heart rate stability
hysteresis
Powerful interface allowing an easy
integration into existing systems.
14
Unique Antares features that you won’t
find in other extractors (2)
Possibility to program extraction times.
Scalable outputs (i.e. usable by any
measurement tool).
Full audit trail (which strips were selected,
and why).
15
AMPS internal usage of Antares
- example
Two separate studies with two
pharmaceutical companies.
Both studies with more than 5000 ECGs.
Antares-based ECG analysis compared with
that from standard 12-lead ECGs.
Great control of patient-individual QTc
i
variability.
16
SD of QTc
i
(power-law) after Antares
selection of stable 20-seconds ECGs
32 subjects, 70 ECGs per subjects
SD < 6 ms: n = 7
6 SD < 8 ms: n = 19
8 SD < 10 ms : n = 6
17
For questions, additional information, or a price proposition
please contact:
E-mail: Marketing@amps-llc.com
Web: www.amps-llc.com
418 Central Park West suite 21 - 10025 - New York - New York
Via Paolo VI, 34 - 25018 - Montichiari (BS) - Italy
127 rue de la Glaciere - 75013 - Paris - France
AMPS LLC