WO2015136567A1 - Method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function - Google Patents

Method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function Download PDF

Info

Publication number
WO2015136567A1
WO2015136567A1 PCT/IT2015/000067 IT2015000067W WO2015136567A1 WO 2015136567 A1 WO2015136567 A1 WO 2015136567A1 IT 2015000067 W IT2015000067 W IT 2015000067W WO 2015136567 A1 WO2015136567 A1 WO 2015136567A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
cardiac cycle
median
mcc
maternal
Prior art date
Application number
PCT/IT2015/000067
Other languages
French (fr)
Inventor
Laura BURATTINI
Angela AGOSTINELLI
Original Assignee
Universita' Politecnica Delle Marche
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Universita' Politecnica Delle Marche filed Critical Universita' Politecnica Delle Marche
Publication of WO2015136567A1 publication Critical patent/WO2015136567A1/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/318Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/318Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A61B5/344Foetal cardiography
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/318Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A61B5/346Analysis of electrocardiograms
    • A61B5/349Detecting specific parameters of the electrocardiograph cycle
    • A61B5/352Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/0002Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
    • A61B5/0004Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by the type of physiological signal transmitted
    • A61B5/0006ECG or EEG signals
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/0002Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
    • A61B5/0004Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by the type of physiological signal transmitted
    • A61B5/0011Foetal or obstetric data
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/389Electromyography [EMG]

Definitions

  • the present invention relates to a method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function.
  • ECG fetal signal (fetal ECG - fECG) or signal of interest;
  • ECG maternal signal maternal ECG - mECG
  • primary interference
  • the noise comprising measurement interferences and physiological interferences of a type other than the cardiac type.
  • the signal coming from a channel (abdominal RECORDING - aREC) of an abdominal recording can therefore be summed up by means of the following formula:
  • aREC fECG + mECG + NOISE.
  • the fetal signal fECG and the maternal signal mECG are signals that are morphologically the same, made up of the repetition of characteristic waves that represent, as a whole, the entire cardiac cycle (Cardiac Cycle), illustrated and indicated by reference CC in Figure 1, shown by way of example.
  • Such characteristic waves are:
  • the QRS complex the widest in size, which indicates ventricular depolarization and hides atrial repolarization (relaxation);
  • the amplitude of the fetal signal fECG is significantly smaller (at most around 0.06- 0.07 mV) compared to the amplitude of the maternal signal mECG (on average around 0.10-0.15 mV), while the frequency bands are to a great extent superimposed (on average 1.8-45 Hz for the fECG and 1.3-32.5 HZ for the mECG). Consequently, the use of traditional fixed-band linear filters does not appear to be an efficient method for separating the two ECG components.
  • noise the amplitude of which tends to be very variable (from zero to a few tenths of mV) is distinguished by a very wide frequency band with respect to the fetal signal fECG and to the maternal signal mECG (on average from 0 to hundreds of Hz), with components that can however fall within their bands.
  • the fetal signal fECG permits monitoring the electrical functionality of the fetal heart and obtaining considerable information on the fetus, such as heartbeat frequency, the size of the fetal heart and the state of oxygenation. Recording can also be done in an invasive way by applying a spiral electrode on the part presented by the fetus.
  • a big drawback is that such measurements can only be made during childbirth, making the situation somewhat uncomfortable for both the mother and the fetus. Furthermore, the method is not without risks, inasmuch as the fetus is exposed to possible scratches due to the application of the electrode with consequent perinatal infections.
  • One of the solutions practised in the state of the art for the extraction of the fetal signal fECG consists in assessing the maternal signal mECG and then subtracting it from the abdominal signal.
  • the assessment of the maternal signal mECG can be made considering signals ECG coming from the sensors closest to the maternal heart (multi-lead technique) or directly on the abdominal recordings (single-lead technique).
  • the main aim of the present invention is to provide a method for the filtering of the electrocardiographic signals or the like which is effective and non-invasive.
  • Another object of the present invention is to provide a method for the filtering of electrocardiographic signals or the like which can be used for the noninvasive monitoring of the fetal cardiac function and which, in particular, starting from a channel aREC of an abdominal recording (single-lead technique), permits providing the fetal signal fECG at output.
  • Another object of the present invention is to provide a method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, which allows to overcome the mentioned drawbacks of the prior art within the ambit of a simple, rational, easy and effective to use as well as affordable solution.
  • Figure 2 shows an example of input signal and an example of output signal obtained using the method of extraction of the fetal signal fECG according to the invention
  • Figure 3 is a general block diagram of the method according to the invention.
  • Figure 4 shows a block diagram relating to the procedure of construction of the maternal signal mECG;
  • Figure 5 shows the segmentation of a cardiac cycle in the segments QRS and TUP
  • Figure 6 shows the duration of a nth cardiac cycle and its segments QRS and TUP;
  • Figure 7 shows a block diagram relating to the calculation of the median maternal cardiac cycle
  • Figure 8 shows a graph that illustrates an example of calculation of the median maternal cardiac cycle as the median of all the maternal cardiac cycles
  • Figure 9 shows a block diagram relating to the reconstruction of the maternal cardiac cycles for the assessment of the maternal signal mECG
  • Figure 10 shows an example of maximum alignment between the peaks R of the constructed maternal signal mECG and the maternal ones identified in the filtered abdorninal signal faREC;
  • Figure 11 shows an example of extraction of the fetal ECG signal containing noise components
  • Figure 12 shows a block diagram relating to the procedure of construction of the fetal signal fECG
  • Figure 13 shows an example of reconstruction of the fetal signal fECG
  • Figure 14 shows a block diagram relating to the calculation of the median fetal cardiac cycle
  • Figure 15 shows a block diagram relating to the reconstruction of the fetal cardiac cycles for the assessment of the fetal signal fECG
  • Figures 16a and 1 b show an example of a maternal cardiac cycle respectively constructed without and with the use of the segmentation of the median maternal cardiac cycle and consequent modulation and demodulation of the segments TUP.
  • M globally indicated by M is a method for filtering electrocardiographic signals or the like.
  • the proposed method M can be used for monitoring fetal cardiac function and is intended for the category of ECG EMG (Electro Cardiography or Electro Myography) recorders or the like which provide at least a channel aREC.
  • ECG EMG Electro Cardiography or Electro Myography
  • the procedure described below can be repeated for each channel.
  • the proposed method M envisages an aREC signal at input corresponding to an ECG/EMG signal obtained from an abdominal recording and provides the fetal signal fECG at output.
  • the method M permits the extraction of the fetal signal fECG from the signal aREC and involves the following steps:
  • the pre-filtering of the input signal aREC involves the eliniination of the noise components NOISE which fall outside the band of the fetal signals fECG and maternal signals mECG.
  • Such pre-filtering is done by applying to the input signal aREC a bidirectional bandpass filter with cutoff frequency equal to 0.5 Hz and 45 Hz.
  • a filtered signal faREC is obtained at output (filtered abdominal RECORDING) which is mainly, but not exclusively (considering the presence of a number of noise components which can survive pre-filtering) made up of the two ECG components of the fetal signal fECG and of the maternal signal mECG.
  • the filtered signal faREC essentially contains the superimposition of the fetal signal fECG and of the maternal signal mECG, where the latter is distinguished by waves R of much greater amplitude, and which are therefore easier to identify.
  • the identification of the peaks R of the maternal signal mECG (step 2) is done by applying an algorithm deriving, e.g., from the traditional algorithm of Pan- Tompkins, which at output provides the sequence of the samples mRSeq (maternal R-peak Sequence) of the filtered signal faREC in correspondence of which are the peaks R of the maternal signal mECG.
  • an algorithm deriving e.g., from the traditional algorithm of Pan- Tompkins, which at output provides the sequence of the samples mRSeq (maternal R-peak Sequence) of the filtered signal faREC in correspondence of which are the peaks R of the maternal signal mECG.
  • the length of the samples mRSeq corresponds to the number of maternal beats mNB (maternal Number of Beats) included in the filtered signal faREC.
  • the construction of the maternal signal mECG is done in two steps: the calculation of the median maternal cardiac cycle (step 31) and the reconstruction of the maternal cardiac cycles (step 32).
  • the calculation of the median maternal cardiac cycle envisages the calculation of the intervals RR mlntRR (maternal RR Interval; step 311), measured as the temporal distance between the peak R of the heartbeat considered and the peak R of the previous heartbeat.
  • the duration of the characteristic wave complex PQRSTU depends on the RR interval.
  • the duration of the complex QRS is independent of the RR interval, while the duration of all the other characteristic waves depends directly on it (i.e., it grows along with the growth of the RR interval and vice versa).
  • this cycle can be segmented into two parts:
  • the calculation of the median maternal cardiac cycle mCCmdn occurs using all the maternal cardiac cycles mCC (maternal CC), extracted from the filtered signal faREC using the samples mRSeq (step 312) and segmented as described above (step 313).
  • the segment QRS nth (QRSn) is identified within the time interval that starts ⁇ ms before Rn and ends ⁇ ms after Rn;
  • the segment TUP nth (TUP n ) is identified inside the time interval that starts ⁇ ms after Rn and ends ⁇ ms before Rn+1. From what has been said above therefore, each maternal cardiac cycle is distinguished by a duration of its own, but in all cycles, the duration of the segment QRS is the same, i.e., equal to 2 ⁇ .
  • the matemal cardiac cycles mCC must be of the same length.
  • the duration mCCmdnD (maternal CC median Duration) of the median maternal cardiac cycle mCCmdn is calculated as the median of the intervals mlntRR identified by the time difference between two subsequent samples of the sequence mRSeq (step 314).
  • the cardiac cycles must therefore be modulated in length (stretched or compressed) so they all have a length equal to the duration mCCmdnD (step 315).
  • the modulation operation only involves the segments TUP.
  • the calculation of the median matemal cardiac cycle does therefore envisage the reconstruction of modulated cardiac cycles mCC, starting from the segments QRS and with the modulated segments TUP (step 316).
  • the median maternal cardiac cycle mCCmdn can be calculated as the median of same (step 317).
  • figure 8 shows a graph with an example of calculation of the median maternal cardiac cycle as the median of all the maternal cardiac cycles.
  • the median maternal cardiac cycle mCCmdn is cleaned of all interferences, mcluding the fetal signal fECG.
  • the reconstruction of the maternal cardiac cycles (step 32) first of all envisages the segmentation of the median maternal cardiac cycle mCCmdn, in order to obtain segments TUP and segments QRS (step 321).
  • step 323 demodulating (step 322), from time to time, the median maternal cardiac cycle mCCmdn and then concatenating them (step 324).
  • the demodulation (compression or stretch) operation only involves the median segment TUP which is, from rime to time, demodulated, so that its length coincides with that of the segment TUP of the corresponding maternal cardiac cycle mCC of the filtered signal faREC (step 322).
  • correlative optimization processes are used in order to achieve the maximum temporal alignment and the minimum mean amplitude error between the waves of the constructed maternal signal mECG and the maternal ones identified in the filtered signal faREC.
  • an ECG fetal signal nfECG (noisy fetal ECG) is obtained containing the noise components NOISE that have survived pre- filtering and are contained in the filtered signal faREC.
  • FIG 11 shows an example of extraction of the fetal signal with noise components nfECG.
  • the identification of the fetal peaks R occurs by means of the application e.g. of a Pan-Tompkins- inspired algorithm, which provides at output the sequence of the samples fRSeq (fetal R-peak Sequence) of the fetal signal with noise nfECG in correspondence of which are the peaks R of the fetal signal fECG.
  • the length of the samples fRSeq corresponds to the number of fetal beats fNB (fetal Number of Beats) included in the fetal signal with noise nfECG.
  • step 6 the construction of the fetal signal fECG (step 6) is done in two steps.
  • the fetal cardiac cycle fCC nth (fCC n ) and its segments QRS and TUP are identified using an ⁇ generally smaller than that used in the maternal case (e.g., an ⁇ equal to 20 ms).
  • the median fetal cardiac cycle fCCmdn (fetal CC median) is calculated as the median of all the resulting fetal cardiac cycles fCC by modulating in length all the segments TUP (step 61).
  • the reconstruction of the cardiac cycles is made in order to obtain the fetal signal fECG (step 62).
  • the fetal signal fECG is obtained by reconstructing all the fetal cardiac cycles fCC (equal to fNB in number) demodulating, from time to time, the median fetal cardiac cycle fCCmdn and then concatenating them.
  • the fetal signal fECG is shown, by way of example, in figure 13.
  • the figures 14 and 15 show in detail all the phases relating to the calculation of the median fetal cardiac cycle fCCmdn (step 61) and to the reconstruction of the fetal cardiac cycles fCC (step 62).
  • steps 311, 312, 313, 314, 315, 316 and 317 shown in figure 7 are fully identical to steps 611, 612, 613, 614, 615, 616 and 617 of figure 14, while the steps 321, 322, 323 and 324 of figure 9 are fully identical to steps 621 , 622, 623 and 624 of figure 15.
  • the method according to the invention can be implemented by means of a suitable system for the filtering of the electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function.
  • system according to the invention comprises:
  • the construction means of the maternal signal mECG and the construction means of the fetal signal fECG comprise:
  • the construction means of the maternal signal mECG and the construction means of the fetal signal fECG comprise:
  • the figures 16a and 16b show an example of maternal cardiac cycle constructed without and with the use of the segmentation of the median maternal cycle and consequent modulation and demodulation of the segments TUP.
  • Figure 16a shows how the maternal cardiac cycle constructed without the use of the segmentation and consequent modulation and demodulation of the median segments TUP involves, during the subtraction phase, the presence of an artifact of amplitude equal to that of the fetal peak R and therefore the possibility of the wrong identification of the fetal peaks R.
  • figure 16b shows how median maternal cardiac cycle segmentation and consequent modulation and demodulation of the segments TUP makes the entity of the artifact negligible.
  • Another big advantage of the present invention is to be found in the fact that a pure maternal ECG signal is not required. Consequently, irrespective of the number of recording channels and of the position of the electrodes on the body of the expectant mother, the proposed method optimizes the success of the fetal signal fECQ extraction operation.
  • the method according to the invention is easy to implement in the devices already present on the market and can be used as a stand-alone application for any calculator.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Cardiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Pathology (AREA)
  • Biophysics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physiology (AREA)
  • Psychiatry (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The method (M) for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, comprises a step of acquisition of an input signal (aREC) of electrocardiographic type or the like, a step of identification (312, 612) of a cardiac cycle (mCC, fCC) starting from the input signal (aREC), a step of segmentation (313, 613) of said cardiac cycle (mCC, fCC) in two segments of characteristic waves: a QRS segment and a TUP segment.

Description

METHOD FOR THE FILTERING OF ELECTROCARDIOGRAPHIC SIGNALS OR THE LIKE, USABLE IN PARTICULAR FOR THE MONITORING OF FETAL HEART FUNCTION
Technical Field
The present invention relates to a method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function.
Background Art
It is known that abdominal recordings for monitoring the fetal heart function are obtained using standard electrocardiographic appliances (ECG) or electromyographic appliances (EMG), placing the electrodes on the abdomen of the pregnant woman starting from the thirty-seventh week of gestation.
These recordings, which can be mono or multichannel, provide signals that appear as the sum of three components:
- me ECG fetal signal (fetal ECG - fECG) or signal of interest;
- the ECG maternal signal (maternal ECG - mECG) or primary interference; and
- the noise (NOISE), comprising measurement interferences and physiological interferences of a type other than the cardiac type.
The signal coming from a channel (abdominal RECORDING - aREC) of an abdominal recording can therefore be summed up by means of the following formula:
aREC= fECG + mECG + NOISE.
Because the anatomy and the physiology of the fetal heart and the maternal heart are the same (four chambers: two atriums and two ventricles), the fetal signal fECG and the maternal signal mECG are signals that are morphologically the same, made up of the repetition of characteristic waves that represent, as a whole, the entire cardiac cycle (Cardiac Cycle), illustrated and indicated by reference CC in Figure 1, shown by way of example.
More specifically, such characteristic waves are:
- the wave P representing the atrial depolarization (contraction);
- the QRS complex, the widest in size, which indicates ventricular depolarization and hides atrial repolarization (relaxation);
- the wave T and sometimes the wave U, which represent ventricular repolarization.
Given the different sizes of the maternal heart and the fetal heart, their different distances from the electrodes and the different inter-positioned tissues, the amplitude of the fetal signal fECG is significantly smaller (at most around 0.06- 0.07 mV) compared to the amplitude of the maternal signal mECG (on average around 0.10-0.15 mV), while the frequency bands are to a great extent superimposed (on average 1.8-45 Hz for the fECG and 1.3-32.5 HZ for the mECG). Consequently, the use of traditional fixed-band linear filters does not appear to be an efficient method for separating the two ECG components. Finally, noise, the amplitude of which tends to be very variable (from zero to a few tenths of mV), is distinguished by a very wide frequency band with respect to the fetal signal fECG and to the maternal signal mECG (on average from 0 to hundreds of Hz), with components that can however fall within their bands. The fetal signal fECG permits monitoring the electrical functionality of the fetal heart and obtaining considerable information on the fetus, such as heartbeat frequency, the size of the fetal heart and the state of oxygenation. Recording can also be done in an invasive way by applying a spiral electrode on the part presented by the fetus. In most cases, in the end the fetus assumes a cephalic position (with head turned downwards) and so the electrode is fastened to the scalp. Such method is currently the only one (considered the gold standard of fetal heartbeat frequency measurements) able to provide a pure fetal signal fECG, on which it is easy to filter the low-frequency interferences and perform a morphological analysis.
A big drawback is that such measurements can only be made during childbirth, making the situation somewhat uncomfortable for both the mother and the fetus. Furthermore, the method is not without risks, inasmuch as the fetus is exposed to possible scratches due to the application of the electrode with consequent perinatal infections.
In this perspective, from the clinical viewpoint, very definitely the most promising method is the non-invasive one, distinguished by abdominal recordings obtained by applying electrodes to the pregnant woman's abdomen. This method, nevertheless, presents known problems due to the difficulty in separating the ECG components relating to the maternal heart and the fetal heart (the former very much greater than the latter in terms of amplitude and frequency). ■ ' «
One of the solutions practised in the state of the art for the extraction of the fetal signal fECG, consists in assessing the maternal signal mECG and then subtracting it from the abdominal signal. In particular, the assessment of the maternal signal mECG can be made considering signals ECG coming from the sensors closest to the maternal heart (multi-lead technique) or directly on the abdominal recordings (single-lead technique).
Description of the Invention
The main aim of the present invention is to provide a method for the filtering of the electrocardiographic signals or the like which is effective and non-invasive. Another object of the present invention is to provide a method for the filtering of electrocardiographic signals or the like which can be used for the noninvasive monitoring of the fetal cardiac function and which, in particular, starting from a channel aREC of an abdominal recording (single-lead technique), permits providing the fetal signal fECG at output.
Another object of the present invention is to provide a method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, which allows to overcome the mentioned drawbacks of the prior art within the ambit of a simple, rational, easy and effective to use as well as affordable solution.
The above mentioned objects are achieved by the present method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, according to the characteristics described in claim 1.
The above mentioned objects are achieved by the present system for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, according to the characteristics described in claim 17. Brief Description of the Drawings
Other characteristics and advantages of the present invention will become better evident from the description of a preferred, but not exclusive embodiment of a method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, illustrated by way of an indicative, but not limitative example in the accompanying drawings in which: Figure 1 shows an example of a cardiac cycle;
Figure 2 shows an example of input signal and an example of output signal obtained using the method of extraction of the fetal signal fECG according to the invention;
Figure 3 is a general block diagram of the method according to the invention; Figure 4 shows a block diagram relating to the procedure of construction of the maternal signal mECG;
Figure 5 shows the segmentation of a cardiac cycle in the segments QRS and TUP;
Figure 6 shows the duration of a nth cardiac cycle and its segments QRS and TUP;
Figure 7 shows a block diagram relating to the calculation of the median maternal cardiac cycle;
Figure 8 shows a graph that illustrates an example of calculation of the median maternal cardiac cycle as the median of all the maternal cardiac cycles;
Figure 9 shows a block diagram relating to the reconstruction of the maternal cardiac cycles for the assessment of the maternal signal mECG;
Figure 10 shows an example of maximum alignment between the peaks R of the constructed maternal signal mECG and the maternal ones identified in the filtered abdorninal signal faREC;
Figure 11 shows an example of extraction of the fetal ECG signal containing noise components;
Figure 12 shows a block diagram relating to the procedure of construction of the fetal signal fECG;
Figure 13 shows an example of reconstruction of the fetal signal fECG;
Figure 14 shows a block diagram relating to the calculation of the median fetal cardiac cycle;
Figure 15 shows a block diagram relating to the reconstruction of the fetal cardiac cycles for the assessment of the fetal signal fECG;
Figures 16a and 1 b show an example of a maternal cardiac cycle respectively constructed without and with the use of the segmentation of the median maternal cardiac cycle and consequent modulation and demodulation of the segments TUP.
Embodiments of the Invention
With particular reference to such illustrations, globally indicated by M is a method for filtering electrocardiographic signals or the like.
In particular, the proposed method M can be used for monitoring fetal cardiac function and is intended for the category of ECG EMG (Electro Cardiography or Electro Myography) recorders or the like which provide at least a channel aREC. In case of any multi-channel abdominal recording, the procedure described below can be repeated for each channel.
The use cannot however be ruled out of the method M for different purposes and for filtering electrocardiographic signals of different type.
As schematically shown in figure 2, the proposed method M envisages an aREC signal at input corresponding to an ECG/EMG signal obtained from an abdominal recording and provides the fetal signal fECG at output.
As shown in the general diagram of figure 3, the method M permits the extraction of the fetal signal fECG from the signal aREC and involves the following steps:
- pre-filtering (step 1);
- identification of maternal peaks R (step 2);
- construction of maternal ECG signal mECG (step 3);
- subtraction of maternal ECG signal mECG (step 4);
- identification of fetal peaks R (step 5);
- construction of fetal ECG signal fECG (step 6).
In particular, the pre-filtering of the input signal aREC (step 1) involves the eliniination of the noise components NOISE which fall outside the band of the fetal signals fECG and maternal signals mECG. Such pre-filtering is done by applying to the input signal aREC a bidirectional bandpass filter with cutoff frequency equal to 0.5 Hz and 45 Hz. This way a filtered signal faREC is obtained at output (filtered abdominal RECORDING) which is mainly, but not exclusively (considering the presence of a number of noise components which can survive pre-filtering) made up of the two ECG components of the fetal signal fECG and of the maternal signal mECG.
In particular, the filtered signal faREC essentially contains the superimposition of the fetal signal fECG and of the maternal signal mECG, where the latter is distinguished by waves R of much greater amplitude, and which are therefore easier to identify.
The identification of the peaks R of the maternal signal mECG (step 2) is done by applying an algorithm deriving, e.g., from the traditional algorithm of Pan- Tompkins, which at output provides the sequence of the samples mRSeq (maternal R-peak Sequence) of the filtered signal faREC in correspondence of which are the peaks R of the maternal signal mECG.
The length of the samples mRSeq corresponds to the number of maternal beats mNB (maternal Number of Beats) included in the filtered signal faREC.
As shown by the block diagram of figure 4, the construction of the maternal signal mECG (step 3) is done in two steps: the calculation of the median maternal cardiac cycle (step 31) and the reconstruction of the maternal cardiac cycles (step 32).
More specifically, as schematically shown in figure 7, the calculation of the median maternal cardiac cycle (step 31) envisages the calculation of the intervals RR mlntRR (maternal RR Interval; step 311), measured as the temporal distance between the peak R of the heartbeat considered and the peak R of the previous heartbeat.
In general, in the ECG layout (and therefore in the fECG and mECG layouts) the duration of the characteristic wave complex PQRSTU depends on the RR interval.
More specifically, as a first approximation, the duration of the complex QRS is independent of the RR interval, while the duration of all the other characteristic waves depends directly on it (i.e., it grows along with the growth of the RR interval and vice versa). As shown by way of example in figure 5, by identifying the start of a cardiac cycle in the section PQ (between the end of the wave P and the start of the wave Q) rather than in correspondence of the start of the wave P, as is generally the case, this cycle can be segmented into two parts:
- the segment QRS independent of the R interval;
- the segment TUP proportionate to the RR interval.
Furthermore, the calculation of the median maternal cardiac cycle mCCmdn (maternal CC median) occurs using all the maternal cardiac cycles mCC (maternal CC), extracted from the filtered signal faREC using the samples mRSeq (step 312) and segmented as described above (step 313).
More specifically, as shown by way of example in figure 6, the nth mCC (mCCn) is identified inside the time interval which starts ΔΤ ms before (where for example ΔΤ=40) the nth peak R (Rn) and ends ΔΤ ms before Rn+i; the segment QRS nth (QRSn) is identified within the time interval that starts ΔΤ ms before Rn and ends ΔΤ ms after Rn; the segment TUP nth (TUPn) is identified inside the time interval that starts ΔΤ ms after Rn and ends ΔΤ ms before Rn+1. From what has been said above therefore, each maternal cardiac cycle is distinguished by a duration of its own, but in all cycles, the duration of the segment QRS is the same, i.e., equal to 2ΔΤ.
As again shown in figure 7, to calculate the median maternal cardiac cycle mCCmdn the matemal cardiac cycles mCC must be of the same length.
The duration mCCmdnD (maternal CC median Duration) of the median maternal cardiac cycle mCCmdn is calculated as the median of the intervals mlntRR identified by the time difference between two subsequent samples of the sequence mRSeq (step 314).
The cardiac cycles must therefore be modulated in length (stretched or compressed) so they all have a length equal to the duration mCCmdnD (step 315).
Given the same length of the segments QRS in all maternal cardiac cycles mCC, the modulation operation only involves the segments TUP.
The calculation of the median matemal cardiac cycle (step 31) does therefore envisage the reconstruction of modulated cardiac cycles mCC, starting from the segments QRS and with the modulated segments TUP (step 316).
Once the modulation has been performed of the maternal cardiac cycles mCC, the median maternal cardiac cycle mCCmdn can be calculated as the median of same (step 317).
In this respect, figure 8 shows a graph with an example of calculation of the median maternal cardiac cycle as the median of all the maternal cardiac cycles. For the properties of the median operation, the median maternal cardiac cycle mCCmdn is cleaned of all interferences, mcluding the fetal signal fECG.
As schematically shown in figure 9, the reconstruction of the maternal cardiac cycles (step 32) first of all envisages the segmentation of the median maternal cardiac cycle mCCmdn, in order to obtain segments TUP and segments QRS (step 321).
Furthermore, to obtain the maternal signal mECG, all the maternal cardiac cycles mCC (equal to mNB in number) must be reconstructed (step 323) demodulating (step 322), from time to time, the median maternal cardiac cycle mCCmdn and then concatenating them (step 324).
More specifically, as schematized in figure 9, the demodulation (compression or stretch) operation only involves the median segment TUP which is, from rime to time, demodulated, so that its length coincides with that of the segment TUP of the corresponding maternal cardiac cycle mCC of the filtered signal faREC (step 322).
Finally, as shown by way of example in the graph in figure 10, correlative optimization processes are used in order to achieve the maximum temporal alignment and the minimum mean amplitude error between the waves of the constructed maternal signal mECG and the maternal ones identified in the filtered signal faREC.
By subtracting the constructed maternal signal mECG from the filtered signal faREC (step 4) an ECG fetal signal nfECG (noisy fetal ECG) is obtained containing the noise components NOISE that have survived pre- filtering and are contained in the filtered signal faREC.
Figure 11 shows an example of extraction of the fetal signal with noise components nfECG. In the same way as occurred in the maternal case, the identification of the fetal peaks R (step 5) occurs by means of the application e.g. of a Pan-Tompkins- inspired algorithm, which provides at output the sequence of the samples fRSeq (fetal R-peak Sequence) of the fetal signal with noise nfECG in correspondence of which are the peaks R of the fetal signal fECG.
The length of the samples fRSeq corresponds to the number of fetal beats fNB (fetal Number of Beats) included in the fetal signal with noise nfECG.
As schematically shown in figure 12, in the same way as occurred in the maternal case, the construction of the fetal signal fECG (step 6) is done in two steps.
First of all, the fetal cardiac cycle fCC nth (fCCn) and its segments QRS and TUP are identified using an ΔΤ generally smaller than that used in the maternal case (e.g., an ΔΤ equal to 20 ms).
The median fetal cardiac cycle fCCmdn (fetal CC median) is calculated as the median of all the resulting fetal cardiac cycles fCC by modulating in length all the segments TUP (step 61).
Subsequently, the reconstruction of the cardiac cycles is made in order to obtain the fetal signal fECG (step 62).
The fetal signal fECG is obtained by reconstructing all the fetal cardiac cycles fCC (equal to fNB in number) demodulating, from time to time, the median fetal cardiac cycle fCCmdn and then concatenating them.
The fetal signal fECG is shown, by way of example, in figure 13.
The figures 14 and 15 show in detail all the phases relating to the calculation of the median fetal cardiac cycle fCCmdn (step 61) and to the reconstruction of the fetal cardiac cycles fCC (step 62).
For the description of the single steps, reference should be made to what has been said above with reference to the calculation of the median maternal cardiac cycle mCCmdn (step 31 in figure 4 and figure 7) and to the reconstruction of the maternal cardiac cycles mCC (step 32 in figure 4 and figure 9).
In particular, it is pointed out that the steps 311, 312, 313, 314, 315, 316 and 317 shown in figure 7 are fully identical to steps 611, 612, 613, 614, 615, 616 and 617 of figure 14, while the steps 321, 322, 323 and 324 of figure 9 are fully identical to steps 621 , 622, 623 and 624 of figure 15.
The method according to the invention can be implemented by means of a suitable system for the filtering of the electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function.
In particular, the system according to the invention comprises:
- acquisition means of the input signal aREC;
- construction means of the maternal signal mECG starting from the input signal aREC, suitable for implementing step 3 of method M;
- subtraction means of the maternal signal mECG from the input signal aREC, suitable for implementing step 4 of the method M;
- construction means of a fetal signal fECG, suitable for implementing step 6 of the method M.
Advantageously, the construction means of the maternal signal mECG and the construction means of the fetal signal fECG comprise:
- respective identification means of the cardiac cycles mCC and fCC;
- respective segmentation means of the cardiac cycles mCC and fCC in two segments of characteristic waves: one segment QRS and one segment TUP.
In particular, the construction means of the maternal signal mECG and the construction means of the fetal signal fECG comprise:
- calculation means of the median cardiac cycle mCCmdn and fCCmdn, suitable for implementing the steps 31 and 61 of the method M, respectively;
- reconstruction means of the cardiac cycles mCC and fCC, suitable for implementing the steps 31 and 62 of the method M, respectively.
It is demonstrated how the described invention achieves the proposed objects. In fact, despite various methodologies already present in literature permitting the extraction of the fetal signal fECG from abdominal recordings by subtraction of mECG, in none of such known methodologies however are the maternal signal mECG and the fetal signal fECG constructed by segmenting the maternal cardiac cycles and the fetal cardiac cycles in segments QRS and TUP, in order to maintain in the constructed signal the physiological variability of the interval R and guarantee a perfect alignment between this and the original signal.
By way of example, and to demonstrate such statement, the figures 16a and 16b show an example of maternal cardiac cycle constructed without and with the use of the segmentation of the median maternal cycle and consequent modulation and demodulation of the segments TUP.
Figure 16a shows how the maternal cardiac cycle constructed without the use of the segmentation and consequent modulation and demodulation of the median segments TUP involves, during the subtraction phase, the presence of an artifact of amplitude equal to that of the fetal peak R and therefore the possibility of the wrong identification of the fetal peaks R.
On the contrary, figure 16b shows how median maternal cardiac cycle segmentation and consequent modulation and demodulation of the segments TUP makes the entity of the artifact negligible.
Furthermore the median operation permits obtaining signals as far as possible devoid of the noisy components that have survived pre-filtering.
Another big advantage of the present invention is to be found in the fact that a pure maternal ECG signal is not required. Consequently, irrespective of the number of recording channels and of the position of the electrodes on the body of the expectant mother, the proposed method optimizes the success of the fetal signal fECQ extraction operation.
Furthermore, the method according to the invention is easy to implement in the devices already present on the market and can be used as a stand-alone application for any calculator.

Claims

1) Method (M) for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, comprising:
at least a step of acquisition of at least an input signal (aREC) of electrocardiographic type or the like;
characterized in that it comprises:
at least a step of identification (312, 612) of at least a cardiac cycle (mCC, fCC) starting from said input signal (aREC);
at least a step of segmentation (313, 613) of said cardiac cycle (mCC, fCC) in two segments of characteristic waves: a QRS segment and a TUP segment
2) Method (M) according to claim 1, characterized in that it comprises:
at least a step of construction (step 3) of a maternal signal (mECG) starting from said input signal (aREC), obtained from an abdominal recording;
at least a step of subtraction (step 4) of said maternal signal (mECG) to said input signal (aREC);
at least a step of construction (step 6) of a fetal signal (fECG);
wherein at least one of said step of construction (step 3) of the maternal signal (mECG) and said step of construction (step 6) of the fetal signal (fECG) comprises:
- said at least a step of identification of said at least a cardiac cycle (mCC, fCC);
said at least a step of segmentation of said at least a cardiac cycle (mCC, fCC) in two segments of characteristic waves: a QRS segment and a TUP segment.
3) Method (M) according to one or more of the preceding claims, characterized in that it comprises at least a step of pre-filtering (step 1) of said input signal (aREC), to obtain a filtered signal (faREC).
4) Method (M) according to one or more of the preceding claims, characterized in that it comprises at least a step of identification (step 2) of the peaks R of the maternal signal (mECG).
5) Method (M) according to one or more of the preceding claims, characterized in that it comprises at least a step of identification (step 5) of the peaks R of the fetal signal (nfECG).
6) Method (M) according to one or more of the preceding claims, characterized in that it comprises at least a step of calculation (step 31, step 61) of a median cardiac cycle (mCCmdn, fCCmdn).
7) Method (M) according to claim 7, characterized in that said calculation (step 31, step 61) of a median cardiac cycle (mCCmdn, fCCmdn) comprises the calculation of intervals RR, measured as the temporal distance between the peak R of the heartbeat considered and the peak R of the previous heartbeat.
8) Method (M) according to one or more of the preceding claims, characterized in that said calculation (step 31, step 61) of a median cardiac cycle (mCCmdn, fCCmdn) comprises the calculation of the duration of said median cardiac cycle (mCCmdn, fCCmdn).
9) Method (M) according to one or more of the preceding claims, characterized in that said calculation (step 31, step 61) of a median cardiac cycle (mCCmdn, fCCmdn) comprises a length modulation of said TUP segment.
10) Method (M) according to one or more of the preceding claims, characterized in that said calculation (step 31, step 61) of a median cardiac cycle (mCCmdn, fCCmdn) comprises a reconstruction of modulated cardiac cycles (mCC, fCC).
11) Method (M) according to one or more of the preceding claims, characterized in that said calculation (step 31, step 61) of a median cardiac cycle (mCCmdn, fCCmdn) comprises the calculation of the median of said modulated cardiac cycles (mCC, fCC).
12) Method (M) according to one or more of the preceding claims, characterized in that it comprises at least a step of reconstruction (step 32, step
62) of the cardiac cycles (MCC, FCC).
13) Method (M) according to claim 12, characterized in that said reconstruction (step 32, step 62) comprises the segmentation of said median cardiac cycle (mCCmdn, fCCmdn) in two segments:
- a median QRS segment;
a median TUP segment.
14) Method (M) according to one or more of the preceding claims, characterized in that said reconstruction (step 32, step 62) comprises a length demodulation of said median TUP segment, so that the length of the median TUP segment coincides with that of the TUP segment of the corresponding cardiac cycle (mCC, fCC).
15) Method (M) according to one or more of the preceding claims, characterized in that said reconstruction (step 32, step 62) comprises the reconstruction of demodulated cardiac cycles (mCC), starting from said QRS segments and from said demodulated TUP segments.
16) Method (M) according to one or more of the preceding claims, characterized in that said reconstruction (step 32, step 62) comprises the concatenation of said demodulated cardiac cycles (mCC, fCC).
17) System for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function, comprising:
acquisition means of at least an input signal (aREC) of electrocardiographic type or the like;
characterized in that it comprises:
identification means (312, 612) of at least a cardiac cycle (mCC, fCC) starting from said input signal (aREC);
segmentation means (313, 613) of said cardiac cycle (mCC, fCC) in two segments of characteristic waves: a QRS segment and a TUP segment.
18) System according to claim 17, characterized in that it comprises:
construction means of a maternal signal (mECG) starting from said input signal (aREC), obtained from an abdominal recording;
subtraction means of said maternal signal (mECG) to said input signal (aREC);
construction means of a fetal signal (fECG);
wherein at least one of said construction means of the maternal signal (mECG) and said construction means of the fetal signal (fECG) comprises:
said identification means of said at least a cardiac cycle (mCC, fCC);
- said segmentation means of said at least a cardiac cycle (mCC, fCC) in two segments of characteristic waves: a QRS segment and a TUP segment.
19) System according to one or more of claims 17 and 18, characterized in that it comprises calculation means of a median cardiac cycle (mCCmdn, fCCmdn). 20) System according to one or more of claims from 17 to 19, characterized in that it comprises reconstruction means of the cardiac cycles (mCC, flCC).
PCT/IT2015/000067 2014-03-12 2015-03-11 Method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function WO2015136567A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
ITMO2014A000063 2014-03-12
ITMO20140063 2014-03-12

Publications (1)

Publication Number Publication Date
WO2015136567A1 true WO2015136567A1 (en) 2015-09-17

Family

ID=50733202

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IT2015/000067 WO2015136567A1 (en) 2014-03-12 2015-03-11 Method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function

Country Status (1)

Country Link
WO (1) WO2015136567A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108652613A (en) * 2017-03-30 2018-10-16 深圳市理邦精密仪器股份有限公司 The method and device that signal time-frequency figure generates
CN109157212A (en) * 2018-08-30 2019-01-08 武汉吉星医疗科技有限公司 The compound filter computing system and its method of electrocardiograph based on android system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4211237A (en) * 1977-04-14 1980-07-08 Biotronik Mess- Und Therapiegerate Gmbh & Co. Method and apparatus for identifying recurring signal patterns
US20050267376A1 (en) * 2004-05-28 2005-12-01 Dorothee Marossero Maternal-fetal monitoring system
CN102764117A (en) * 2012-04-25 2012-11-07 重庆大学 Portable fetal ECG monitor
WO2014030162A1 (en) * 2012-08-22 2014-02-27 Ben-Gurion University Of The Negev Research & Development Authority Separating clinically relevant sources of electrical activity in ecg signals

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4211237A (en) * 1977-04-14 1980-07-08 Biotronik Mess- Und Therapiegerate Gmbh & Co. Method and apparatus for identifying recurring signal patterns
US20050267376A1 (en) * 2004-05-28 2005-12-01 Dorothee Marossero Maternal-fetal monitoring system
CN102764117A (en) * 2012-04-25 2012-11-07 重庆大学 Portable fetal ECG monitor
WO2014030162A1 (en) * 2012-08-22 2014-02-27 Ben-Gurion University Of The Negev Research & Development Authority Separating clinically relevant sources of electrical activity in ecg signals

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ANGELA AGOSTINELLI ET AL: "Extracting a Clean ECG from a Noisy Recording: a New Method based on Segmented-Beat Modulation", 1 September 2014 (2014-09-01), XP055158517, Retrieved from the Internet <URL:http://www.cinc.org/archives/2014/pdf/0049.pdf> [retrieved on 20141215] *
LIPPONEN JUKKA A ET AL: "Advanced maternal ECG removal and noise reduction for application of fetal QRS detection", COMPUTING IN CARDIOLOGY 2013, N/A, 22 September 2013 (2013-09-22), pages 161 - 164, XP032550587, ISSN: 2325-8861, ISBN: 978-1-4799-0884-4, [retrieved on 20140114] *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108652613A (en) * 2017-03-30 2018-10-16 深圳市理邦精密仪器股份有限公司 The method and device that signal time-frequency figure generates
CN108652613B (en) * 2017-03-30 2020-11-03 深圳市理邦精密仪器股份有限公司 Method and device for generating signal time-frequency diagram
CN109157212A (en) * 2018-08-30 2019-01-08 武汉吉星医疗科技有限公司 The compound filter computing system and its method of electrocardiograph based on android system

Similar Documents

Publication Publication Date Title
US7809433B2 (en) Method and system for limiting interference in electroencephalographic signals
US7747316B2 (en) Blind adaptive filter extraction of fetal electrocardiogram signal estimate
Assaleh et al. A novel technique for the extraction of fetal ECG using polynomial networks
Martens et al. A robust fetal ECG detection method for abdominal recordings
US5372139A (en) Method for suppressing a maternal electrocardiogram signal from a fetal electrocardiogram signal obtained with invasive and non-invasive techniques using an almost pure maternal electrocardiogram signal as a trigger
Agostinelli et al. Extracting a clean ECG from a noisy recording: a new method based on segmented-beat modulation
WO2009013246A1 (en) Fetal monitoring
WO2001067950A1 (en) Qrst subtraction using an adaptive template for analysis of tu wave obscured atrial activity
WO2008014895A1 (en) Method for processing cardioelectric signals and corresponding device
Jeyarani et al. Analysis of noise reduction techniques on QRS ECG waveform-by applying different filters
Fanelli et al. Telefetalcare: A first prototype of a wearable fetal electrocardiograph
Lipponen et al. Advanced maternal ECG removal and noise reduction for application of fetal QRS detection
Lamesgin et al. Extraction of fetal ECG from abdominal ECG and heart rate variability analysis
Christov et al. Cancellation of the maternal and extraction of the fetal ECG in noninvasive recordings
Viunytskyi et al. Signal processing techniques for fetal electrocardiogram extraction and analysis
CN112839584A (en) Systems and methods for maternal uterine activity detection
Illanes-Manriquez et al. An algorithm for robust detection of QRS onset and offset in ECG signals
Di Maria et al. An algorithm for the analysis of fetal ECGs from 4-channel non-invasive abdominal recordings
Ruffo et al. Non-invasive foetal monitoring with combined ECG-PCG system
WO2015136567A1 (en) Method for the filtering of electrocardiographic signals or the like, usable in particular for the monitoring of fetal heart function
Hansen et al. Comparing twelve-lead electrocardiography with close-to-heart patch based electrocardiography
Viunytskyi et al. Fetal ECG and heart rhythm analyzing using BabyCard
Niida et al. Fetal Heart Rate Detection Using First Derivative of ECG Waveform and Multiple Weighting Functions
Agostinelli et al. Robustness of the segmented-beat modulation method to noise
Janusek et al. Application of wavelet based denoising for T-wave alternans analysis in high resolution ECG maps

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15720811

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase
122 Ep: pct application non-entry in european phase

Ref document number: 15720811

Country of ref document: EP

Kind code of ref document: A1