Methods and systems for the delivery of accurate and precise measurements from the body-surface electrocardiogram
Described here are methods, devices, and systems for characterizing a physiological signal, and more specifically an electrocardiogram (ECG) signal. Generally, the method includes receiving an ECG signal generated by an ECG device coupled to a patient. The ECG signal may comprise a plurality of cardiac beat intervals. A set of evaluable replicates may be identified using a signal-to-noise ratio (SNR) for each cardiac beat, a repolarization signal, and an isoelectric line. Interval measurements may be determined from the set of evaluable multi-beat sequences. An ECG signal characteristic may be determined from the interval measurements.
1. A method of providing interval measurements based on an electrocardiogram (ECG) signal comprising:
receiving an ECG signal generated by an ECG device coupled to a subject, wherein the ECG signal generated by the ECG device comprises a plurality of cardiac beats;
determining, by a computer processor, a signal-to-noise ratio (SNR) for each cardiac beat based on a repolarization signal and an isoelectric line;
identifying a set of evaluable multi-beat sequences using the SNR of each cardiac beat, wherein the sequences have a pre-determined length; and
determining, by a computer processor, the interval measurements from the set of evaluable multi-beat sequences.
2. The method of claim 1 , wherein the repolarization signal corresponds to a T-wave voltage and the isoelectric line corresponds to an ECG voltage level of a PR interval where a sum of voltage gradients is zero.
3. The method of claim 1 , wherein identifying the set of evaluable multi-beat sequences excludes multi-beat sequences that include non-sinus origin.
4. The method of claim 1 , wherein the SNR is based upon a ratio of a root mean square voltage of a T-wave and a root mean square voltage of an isoelectric region prior to Q-onset of a corresponding QRS complex to the T-wave.
5. The method of claim 1 , wherein the pre-determined length is a pre-determined set length or a pre-determined minimum length.
6. The method of claim 5 , wherein the length is a time duration or several cardiac beats.
7. The method of claim 1 , wherein the SNR is
S
N
R
=
20
*
log
(
V
rms
signal
V
rms
noise
)
where
V
rms
=
1
n
*
∑
i
=
1
n
(
v
i
)
2
and n is a V rms interval and v is voltage.
8. The method of claim 7 , wherein the V rms signal is a root mean square voltage in the T-wave from a J-point until a 4/7RR point and V rms noise is a root mean square voltage in an isoelectric region over a 40 ms interval prior to Q-onset.
9. The method of claim 1 , wherein determining the set of evaluable multi-beat sequences is based on beat characteristics comprising at least one of heart rate stability, non-sinus beat morphology and QTcF.
10. The method of claim 9 , wherein the heart rate stability is based on a median RR value for a predetermined duration prior to each cardiac beat, and wherein the beat excluded from the set of evaluable multi-beat sequences when current RR value is greater than 20% of the median RR value.
11. The method of claim 1 , wherein determining the set of evaluable multi-beat sequences comprises determining an order of the multi-beat sequences in the ECG signal based on at least one of a number of beats within the multi-beat sequence, heart rate stability, non-sinus beat morphology, the SNR, and a QTcF.
12. The method of claim 1 , further comprising selecting a processing automation level from a plurality of processing automation levels, wherein determining the interval measurements is based on the selected processing automation level.
13. The method of claim 12 , wherein the plurality of automation levels comprises a highly automatic level, a semi-automatic level and a manual adjudication level, wherein the semi-automatic level includes greater user input than the highly automatic level, and the manual adjudication level requires greater user input than the semi-automatic level.
14. The method of claim 13 , wherein the highly automatic level comprises classifying beats for one of automatic processing and human review.
15. The method of claim 13 , wherein the semi-automatic level is selected when at least three consecutive beats in at least two of three sequences of a pre-determined length in the set of multi-beat sequences are determined as satisfactory.
16. The method of claim 13 , wherein the manual adjudication level comprises determining the interval measurements based on extraction of at least three consecutive beats with full fiducial correction.
17. The method of claim 1 , further comprising removing noise from the ECG signal comprising at least one of baseline removal and 60 Hz interference removal.
18. The method of claim 1 , further comprising performing a consistency check of the interval measurements, wherein the consistency check comprises at least one of a data sequence check, an interval measurement comparison and time point check.
19. The method of claim 1 , further comprising selecting a set of evaluable replicates for a QT evaluation with a new chemical entity.
20. The method of claim 19 , wherein the QT evaluation is performed for a Thorough QT (TQT) study, an exploratory Investigation New Drug study, a First-In-Human study, a microdose study, phase I study, phase II study or a phase III study.
21. A system for providing interval measurements based on an electrocardiogram (ECG) signal, comprising the system comprising:
a receiver to receive an ECG signal generated by an ECG device coupled to a patient, wherein the ECG signal generated by the ECG device comprises a plurality of cardiac beats; and
a processor configured to determine a signal-to-noise ratio (SNR) for each cardiac beat based on a repolarization signal and an isoelectric line;
a processor configured to identify a set of evaluable multi-beat sequences using the SNR of each cardiac beat, wherein the sequences have a pre-determined length; and
a processor configured to determine the interval measurement from the set of evaluable multi-beat sequences.
22. The system of claim 21 , wherein the repolarization signal corresponds to a T-wave voltage and the isoelectric line corresponds to an ECG voltage level of a PR interval where a sum of voltage gradients is zero.
23. The system of claim 21 , wherein identifying the set of evaluable multi-beat sequences excludes multi-beat sequences that include non-sinus origin.
24. The system of claim 21 , wherein the SNR is based upon a ratio of a root mean square voltage of a T-wave and a root mean square voltage of an isoelectric region prior to Q-onset of a corresponding QRS complex to the T-wave.
25. The system of claim 21 , wherein the pre-determined length is a pre-determined set length or a pre-determined minimum length.
26. The system of claim 25 , wherein the length is a time duration or several cardiac beats.
27. The system of claim 21 , wherein the SNR is
S
N
R
=
20
*
log
(
V
rms
signal
V
rms
noise
)
where
V
rms
=
1
n
*
∑
i
=
1
n
(
v
i
)
2
and n is a V rms interval and v is voltage.
28. The system of claim 27 , wherein the V rms signal is a root mean square voltage in the T-wave from a J-point until a 4/7RR point and V rms noise is a root mean square voltage in an isoelectric region over a 40 ms interval prior to Q-onset.