RU2791006C1 - System and method for automated electrocardiogram analysis and interpretation - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: method for determining the characteristics of an electrocardiogram (ECG) is carried out using a system containing a processing means and an ECG recording device. In this case, an ECG signal is obtained using a recording device. A smoothed ECG signal is obtained by processing means by filtering the noise of the received ECG signal. The first derivatives of the smoothed ECG signal are calculated using the processing means. The characteristics of the ECG are determined using the processing means on the basis of the first derivatives of the smoothed ECG signal. Based on the processing of the distribution of local extrema of the first derivatives of the smoothed ECG signal, a threshold is determined to highlight the range of change of the extrema corresponding to different types of waves. The characteristics of the peaks of the waves of different types are determined based on the selected ranges of extremes and the relative position of the local maxima and minima of the first derivatives of the smoothed ECG signal.

EFFECT: automatic detection of signs of the presence of pathologies in the ECG signal and the critical condition of the patient is provided with high accuracy, noise immunity, reduction of the ECG signal processing time for making decisions in automatic mode, with the possibility of flexible and quick response to changes occurring in the ECG curve during the observation process in the conditions of high variability of the studied signal to build short-term and long-term automated medical prognosis.

9 cl, 7 dwg

Description

The field of technology to which the invention belongs

The invention relates to a system and method for automated extraction of the morphology of an electrocardiographic signal, presentation of an ECG signal in the form of a sequence of cardiocycles, determination of characteristic teeth for specialized devices for monitoring cardiac activity, namely for stationary cardiac monitors, Holter ECG monitors, wearable ECG recording devices, mobile devices.

State of the art

Diagnosis of the human cardiovascular system is one of the most important tasks of cardiology. The main causes of death among people of working age are related to cardiovascular diseases. This necessitates the development and improvement of monitoring tools for objective assessment and prediction of the state of the cardiovascular system. At the moment, the electrocardiogram (ECG) is the most common method for diagnosing the work of the human cardiovascular system. The ECG signal carries a large amount of information, and the detailed automatic analysis of the ECG signal allows you to generate alarms in a timely manner, as well as make predictive conclusions.

Automatic analysis of an electrocardiosignal is a rather complex theoretical problem. This is primarily due to the physiological origin of the signal, which determines its indeterminacy, diversity, variability, unpredictability and susceptibility to numerous types of interference. One of the main reasons for the ineffective diagnosis of the state of the heart is intense interference of various types - contour drift, motion artifact, muscle tremor and network interference, which distort information about the state of the heart.

The problem of isolating a useful signal against the background of a whole complex of interference and distortion is one of the main problems in modern electrocardiological studies. Noise filtering can distort the ECG and lead to interpretation errors. The presence of artifacts in the electrocardiological signal significantly complicates its analysis and identification of diagnostic features. When solving this problem, the complexity lies in the choice of filtering methods to eliminate a certain type of artifacts. Therefore, not only internal filtering is important, invisible to the user, in which the useful signal is distorted to some extent. The removal of noise from the signal received from the recording device is also of great importance, since it is on this basis that the diagnosis will be made.

At the moment, there are several developments aimed at highlighting the characteristics of the electrocardiogram curve and the subsequent interpretation of the electrocardiogram.

Patent publication US 2013/0190638A1, Motion and nose artifact detection for ECG data, discloses a method aimed at overcoming the main shortcomings of popular ECG signal filtering methods associated with the need to build models, determine thresholds and optimal parameters. The method is based on separating pure parts of the ECG signal from motion and noise artifacts, prevents distortion of the location of the R wave peaks, and, using the R-R intervals selected after artifact separation, performs an analysis for the presence of atrial fibrillation.

Empirical Mode Decomposition (EMD), which decomposes the ECG signal into a sum of mode functions (IMF), is used to detect motion artifacts and noise in real time. Each subsequent IMF has a lower frequency than the previous one. This way artifacts can be isolated as they are mostly concentrated at high frequency. The next step uses statistics that include Shannon's entropy to characterize randomness, the mean of the mode functions, and the rms value of the R-R sequence of differences. Based on the comparison of these three indicators with the given thresholds, a decision is made whether the ECG signal contains artifacts or not.

One of the significant disadvantages of EMD, which reduces the quality of decomposition, is the frequent occurrence of mixing of mode functions, which makes frequency separation difficult and reduces its quality (Wu Z., and N.E. Huang (2009), Podladchikova T., R.A.M. Van der Linden, and A.M. Veronig ( 2017), Ensemble empirical mode decomposition: a noise-assisted data analysis method, Advances in Adaptive Data Analysis, Vol. 01, No. 01, pp. 1-41, doi.org/10.1142/S1793536909000047). The components of a multicomponent signal, under a certain influence of destabilizing factors (noise, impulse noise, etc.) and neighboring components with similar frequencies, can “flow” at separate time intervals during decomposition into the mode functions of neighboring IMFs. In addition, to ensure the reliable operation of these algorithms, it is necessary to adjust the thresholds for detecting artifacts in the ECG signal for each patient, which reduces the resistance of the method to noise and ECG variability.

US Patent Publication 2017/0112401 A1, Automatic method to delineate or categorize an electrocardiogram, discloses an ECG signal delineation and classification method based on a convolutional neural network. This method expanded the capabilities of neural networks for interpreting ECGs representing not one, but more anomalies. The method allows you to determine the start and end times of each wave - these are the sections between the characteristic points: the base of the P wave, the QRS complex, the T wave. The neural network displays the list of found anomalies as a vector of scores for anomalies that have exceeded a predetermined threshold.

The disadvantages of convolutional neural networks is a large number of variable network parameters. Variable parameters include: the number of layers, the dimension of the convolution kernel for each of the layers, the number of kernels for each of the layers, the kernel shift step when processing the layer, the activation function, the parameter for highlighting anomalies, etc. All these parameters significantly affect the result, but chosen empirically. The set of training data is also of great importance. In addition, there are difficulties with the accuracy of determining the time characteristic segments. Loss of precision results in systematic partial mixing of the QRS complex with the T wave, erasing the ST segment on the timeline. The limitations of this method should also include the fact that it does not include an assessment of the amplitudes of the waves (ECG voltage) characterizing the excitability of certain areas of the myocardium and is not intended to determine the main R-R interval, the fluctuations in the duration of which are used to judge the correctness of the rhythm of cardiac activity.

To clarify the position of the QRS complex, the derivative of the signal can be used (see patent publication EP 2 676 604 A1, 2013, Real time QRS duration measurement in electrocardiogram). The method involves extracting a portion of the ECG signal around the QRS peak based on the total duration of the cardiocycle pattern and the normal duration of the QRS. After low-frequency interpolation (filtering), the extracted signal is normalized in such a way that its maximum (corresponding to the R wave) is equal to one. The normalized signal is filtered by raising to the N power and differentiated. Two equations of straight lines are formed using the maximum and minimum points of the derivative signal and the R wave. The points of intersection of these straight lines with the baseline approximately determine the boundaries of the QRS complex, i.e. points Q and S. The disadvantage is the high sensitivity of the algorithm to the baseline shift and the decrease in the accuracy of determining the characteristic points Q and S due to baseline drift.

Application of the method to estimation of QRS duration will fail if the R wave amplitude is lower than the T wave amplitude, or if the ECG signal exhibits V pathology, a sign of monomorphic gastric tachycardia. This method does not provide for the selection of the main electrographic sign of the manifestation of atrial fibrillation of the P wave and the T wave, which reflects the recovery phase of the muscular tissue of the heart ventricles between myocardial contractions.

Disclosure of the essence of the invention

A method is proposed for presenting an ECG signal as a sequence of cardiocycles, automatic selection of a sequence of R-R intervals to identify possible heart rhythm disturbances, heterogeneity of contractions in strength and frequency. In particular, methods for diagnosing and analyzing heart diseases and conditions such as atrial fibrillation are presented by determining the presence or absence of the main electrographic sign of the manifestation of atrial fibrillation of the P wave and the isolation of monomorphic gastric tachycardia (V pathology) when premature ventricular contraction is observed.

Methods for detecting a heart rhythm disorder include removing noise from an ECG signal received from a recording device. After receiving the ECG signal using the electrocardiogram recording device, it is smoothed. Smoothing differs in that it does not distort the dynamics of the ECG signal, since it does not require preliminary assumptions about its dynamics, which usually, to one degree or another, can lead to noticeable temporal and amplitude distortions of the characteristic points of the ECG signal. In this case, it is possible to save all the information about the original signal and at the same time filter out noise, stepwise (step) changes that do not contain a useful signal.

The high noise immunity of the filtering method allows using the first and second derivatives of the smoothed (filtered) signal to delineate the ECG signal and highlight the R, T and P waves without the risk of noise distortion. The effect of noise amplification when calculating derivatives when using this filtering method is neutralized.

The characteristic R, T and P waves of the ECG signal are determined by applying simple logical rules with respect to the first and second derivatives, which are estimated using the first and second differences in the readings of the smoothed ECG signal. The first derivatives eliminate errors due to the shift in the baseline of the ECG signal and approach a stationary process. The use of the first derivatives for finding the R, T, and P waves also eliminates the main disadvantages of the method for detecting local extrema when analyzing the shape of an electrocardiogram, namely, the need to set a sensitivity threshold and skip teeth on inclined signal sections.

Next, a histogram of the distribution of all local maxima and minima of the first derivatives is formed, which allows you to automatically determine the ranges of variation of the extrema corresponding to the R, T and P teeth, and highlight the maxima and minima of the first derivatives for these teeth. Automatic analysis of the distribution histograms of local minima and maxima of the first derivatives allows you to quickly automatically detect signs of a large number of pathologies in the ECG signal and the critical condition of the patient, significantly minimizing the processing time of the ECG signal for making decisions in automatic mode.

To ensure efficiency and ensure that the correct decision is made in automatic mode, the relative position of the local extrema of the second derivatives of the smoothed ECG signal is additionally analyzed.

The criteria for changing the relative position of the local maxima and minima of the first and second derivatives is used to automatically separate the R-wave from the V pathology when premature ventricular contraction is observed. The number of erroneous decisions varies from 3 to 30% depending on the complexity of the ECG signal.

The accuracy of ECG signal detection based on the proposed approach varies from 95% to 99.9% depending on the complexity of the original ECG signal. The limitation of this approach is situations when the same form of the ECG signal is interpreted by a specialist in different ways depending on the additional features of the ECG signal, as well as other leads. There are also a number of unidentified signals that are interpreted individually

An important advantage of the proposed method is the ability to respond flexibly and quickly to changes occurring in the ECG curve during the observation process, which is a necessary requirement for automatic ECG signal processing. These methods underlie the described ECG signal analysis method, which is reliable, flexible, and easy to implement.

In one embodiment of an automated electrocardiogram characterization system, the electrocardiogram recording device is a stationary cardiac monitor, a Holter ECG monitor, a wearable ECG recording device, or a mobile device with an ECG recording function.

Brief description of the drawings

Fig. 1 - Block diagram of the algorithm for automatic analysis of the ECG signal;

Fig. 2 - Fragment of the record 111 of the ECG signal obtained by means of an electrocardiograph from the MIT-BIH Arrhythmia database, and the corresponding smoothed ECG signal;

Fig. 3 - Fragment of recording 111 of the ECG signal obtained by means of an electrocardiograph from the MIT-BIH Arrhythmia database, and the corresponding first differences of the smoothed ECG signal;

Fig. 4 – Histograms of the distribution of local maxima and minima of the first differences and smoothed values of the histograms for record 111 (a, b) and record 203 (c, d) from the MIT-BIH Arrhythmia database;

Fig. 5 - First (a) and second derivatives (b), as well as the sum of the second derivatives of the smoothed ECG signal (c) for the fragment of the ECG signal with a marked R-wave.

Fig. 6 – First (a) and second (b) derivatives, as well as the sum of the second derivatives of the smoothed ECG signal (c) for a fragment of the ECG signal with marked pathology V.

Fig. 7 - First derivatives for the fragment of the ECG signal with marked R, T and P waves.

Implementation of the invention

The algorithm for automatic analysis of the ECG signal consists of a series of steps shown in the block diagram (Fig. 1). Further implementation of the invention is disclosed in detail.

1. ECG signal smoothing

The main obstacle to automated ECG analysis is a lot of interference, as well as high signal noise, typical for wearable electronics. The proposed approach to automated ECG analysis is based on finding local amplitude extrema based on the selection of local extrema of the first and second derivatives of the ECG signal. On the one hand, after taking the derivatives, the original non-stationary signal approaches the stationary one, which increases the reliability of identifying its local extrema. However, the first and even more so the second derivatives amplify the interference of the ECG signal received, both from stationary electrocardiographs, ECG holters, and from mobile devices. Therefore, it is necessary to carefully filter the original signal in order to neutralize the effect of noise amplification when taking derivatives and, at the same time, not introduce dynamic distortions into the useful signal. In the work of Podladchikova T., R.A.M. Van der Linden, and A.M. Veronig (2017), Sunspot number second differences as precursor of the following 11-year sunspot cycle, The Astrophysical Journal, 850, 81, doi.org/10.3847/1538-4357/aa93ef an efficient noise filtering method based on smoothness balance was proposed and proximity of the curve of smoothed values to the original curve. This method does not require preliminary assumptions and assumptions about the dynamics of the process and allows you to reproduce the original experimental patterns regardless of the assumptions about the process model, thereby minimizing the risk of distorting important features of the process, erasing important small details and conclusions, which is of great practical importance for building automated ECG processing. curve in conditions of high noise ECG signal received from wearable devices, mobile devices, ECG-holters, stationary cardiographs.

In order for the selected useful signal to be as close as possible to the received ECG signal, but at the same time to be insensitive to random deviations and fluctuations of the measured signal, it is proposed to look for smoothed values that are optimal according to a criterion that includes the requirements for the smoothness of the desired curve and its proximity to the experimental ECG crooked. The level of confidence in the received ECG signal is achieved by requiring the smoothed curve to be close to the original data based on minimizing the sum of squared deviations. The level of smoothness of the smoothed curve is achieved by minimizing the second derivatives of the original ECG signal.

сигнала ЭКГ z_i находится из минимизации функционала J.Thus, the smoothed estimate

ECG signal z _i is found from the minimization of the functional J.

(1)

(1)

характеризует индикатор отклонения, который отвечает за близость сглаженных оценок

к исходному сигналу ЭКГ z_i. Член

показывает индикатор вариабельности, отвечающий за гладкость сглаженной кривой

.Here β is the smoothing coefficient, which determines the proximity of the smoothed ECG signal to the original ECG curve. As β increases, the smoothed signal approaches the original ECG curve, but the degree of smoothing decreases. As β decreases, the degree of smoothing increases, but the deviation from the original ECG curve increases. Member

characterizes the deviation indicator, which is responsible for the proximity of the smoothed estimates

to the original ECG signal z _i . Member

shows the indicator of variability, which is responsible for the smoothness of the smoothed curve

.

Evaluation of the smoothed ECG signal is reduced to solving the following system of equations:

(2)

(2)

Where A =

This approach to filtering noise and optimizing the smoothing algorithm does not require preliminary assumptions and assumptions, often based on the subjective ideas of developers about the ECG signal process under study, which can lead to false conclusions.

In FIG. 2 shows a fragment of the ECG signal obtained by means of an electrocardiograph from the MIT-BIH Arrhythmia database, entry 111 (dashed curve). The solid line shows the smoothed ECG signal obtained from Equation (2).

Efficient noise filtering without distorting the true dynamics of the ECG signal makes it possible to estimate with high accuracy the first and second end differences of the ECG waveform, which by their nature amplify the noise component in the data.

In FIG. 3 shows the first derivatives of the ECG signal (solid curve) obtained from the smoothed ECG curve, which characterize the rate of change of the ECG signal. The corresponding original ECG signal is shown by the dotted curve. As can be seen from FIG. 3, the original ECG signal (dashed curve) is a non-stationary process, while the first derivatives of the smoothed ECG signal approach a quasi-stationary process. While the ECG signal between RR intervals on the dotted curve is characterized by strong variations and shifts, the dynamics of the first derivatives of the smoothed ECG signal (solid curve) changes around a constant level. Eliminating the amplitude trend of the smoothed ECG signal reduces the spread of local maxima of the solid curve corresponding to the R, T, and P waves by a factor of 5 or more compared to the dotted one.

As a rule, the ECG signal received from wearable mobile devices is characterized by a much greater degree of noise and data scatter. Nevertheless, even under conditions of high noise, reliable estimates of the first differences of the process, obtained on the basis of effective noise suppression of the original ECG signal, characterize the rate of change of the ECG signal and represent a quasi-stationary process, the analysis of which greatly simplifies the analysis of the ECG curve, the identification of stable patterns and the detection of pathologies. in automatic mode.

2. Histogram of the distribution of local extrema

As can be seen from FIG. 3, the largest values of local minima and maxima of the derivatives of the smoothed ECG signal are observed in the vicinity of the R-wave. The R-wave is characterized by a rapid rise and a rapid fall in the first derivatives of the ECG signal. The rapid growth of the ECG signal to the top of the R-wave is characterized by a large positive local maximum of the first derivatives. The drop following the top of the R-tooth is characterized by a large negative local minimum of the first derivatives. To automatically identify large local maxima and minima of the first derivatives, it is proposed to analyze the histogram of the distribution of all local maxima and minima of the first derivatives of the ECG curve.

In FIG. 4a and 4c show a histogram of the distribution of local minima of the first derivatives of the smoothed ECG signal for records 111 and 203 (MIT-BIH Arrhythmia Database). The X axis reflects the absolute values of the local minima of the first derivatives, and the Y axis reflects the frequency of observing local minima of different magnitudes.

In FIG. 4a, the elevation marked with an arrow characterizes the neighborhood of the tops of the R-teeth, which differ in the largest negative local minima of the first derivatives. To automatically determine the range of local minima corresponding to the R-teeth, the distribution histogram values are further smoothed based on Equation (2), and the results are shown in FIG. 4b (entry 111) and FIG. 4d (entry 203). In FIG. 4b, the local minimum of the smoothed histogram, located between its last local maximum, to which the arrow is directed, and the previous one (second from the end), automatically indicates the threshold that exceeds all absolute values of the local minima of the first derivatives corresponding to the R waves. Similarly, the range of variations is automatically determined local maxima of the first derivatives corresponding to R-teeth. Areas of growth and fall in the rate of change of the ECG signal, slower than the R wave, are also characteristic of other teeth, for example, the T wave. The local maximum (second from the end) reflects the values of the local minima of the first derivatives, corresponding mainly to the vicinity of the T-tooth. To separate uniquely large local maxima and minima of the first derivatives corresponding to the vicinity of the R-tooth, a normalization coefficient is introduced based on the value of the estimated threshold.

As can be seen from FIG. 4b (record 111), the smoothed values of the histogram have at least three pronounced local maxima. The ECG signal (record 203) is characterized by a large number of pathologies and is difficult even for visual diagnosis by doctors. At the same time, FIG. 4d (record 203) is characterized by only two local maxima, which automatically confirms the presence of a large number of pathologies in the ECG signal. Automatic analysis of histograms of the distribution of local minima and maxima of the first derivatives opens up the possibility of rapid automatic detection of signs of a large number of pathologies in the ECG signal and the critical condition of the patient, significantly minimizing the processing time of the ECG signal for making decisions in automatic mode.

3. Determination of the time and amplitude parameters of the tops of the R waves

In FIG. 5a shows a fragment of the ECG signal for record 111 (solid curve) and the first derivatives of the smoothed ECG signal (dotted curve). The local maximum and minimum of the first derivatives, determined automatically based on the analysis of the histogram of their distribution, are marked with a marker and the numbers 1 and 2, respectively. The top of the R-wave will always be between the local maximum (1) and minimum (2) of the first derivatives of the smoothed ECG signal, while the maximum (1) is always to the left of the minimum (2) of the first derivatives, which opens up the possibility of its fast automatic detection.

To ensure a guaranteed result in automatic mode, the second derivatives of the smoothed ECG signal are additionally analyzed (Fig. 5b, dashed curve). These supplements are required mainly either for pathology or to eliminate the effects of residual noise. A large negative local minimum of the second derivatives (marker 2) reflects a sharp drop in the rate of change of the ECG signal in the vicinity of the R-wave.

A large positive local maximum of the second derivatives (marker 1) demonstrates an increase in the rate of change of the ECG signal following the R-wave. To ensure that the R-wave is selected, it is required that the local minimum (2) is always to the left of the local maximum of the second derivatives (1).

For additional guarantees of operation in automatic mode, the sum of the second derivatives of the smoothed ECG signal is also analyzed (Fig. 5c, dotted curve), which represents the integral characteristic of the increment in the rate of change of the ECG signal at a long stage. To guarantee the selection of the R-tooth, it is also required that the local maximum (1), which shows a significant increase in speed before the characteristic point R, is always to the left of the local minimum (2) of the sum of the second derivatives.

4. Detection of premature contraction of the ventricles (V)

Since in some cases, when the heart muscle fails, premature contraction of the ventricles (V) is observed instead of the R-wave, it is necessary to automatically separate the R-wave from the V pathology. In Fig. Figure 6 shows a fragment of the ECG signal for recording 203 (solid curve), first derivatives (a), second derivatives (b), and the sum of second derivatives (c) of the smoothed ECG signal (dashed curve). The local maximum and minimum of the first derivatives, determined automatically based on the analysis of the histogram of their distribution, are marked with a marker and the numbers 1 and 2, respectively.

Thus, in case of pathology V, a sharp drop in the ECG signal precedes its growth, then the local maximum (1) is located to the right of the local minimum (2) of the first derivatives, which is the direct opposite of the situation with the peak R. If this criterion is triggered, the pathology V is detected automatically between the minimum (2) and maximum (1) of the first derivatives. Similarly, the situation opposite to the situation with the top R is observed for the local minimum (2) of the second derivatives (Fig. 6b), which is always to the right of the local maximum (1), and for the local minimum (2) of the sum of the second derivatives (Fig. 6c), which is always to the left of the corresponding local maximum (1).

For automatic selection of the R-wave and V pathology, it is required to identify the above signs of the behavior of local minima and maxima of the first and second derivatives, as well as the sum of the second derivatives of the smoothed ECG signal. In the event of a discrepancy between the indicated features, the algorithm is trained additionally based on the analysis of neighboring local minima and maxima. The number of such cases varies from 3 to 30% depending on the complexity of the ECG signal.

5. Determination of the time and amplitude parameters of the T and P waves

The proposed approach also underlies the automatic extraction of T and P waves. FIG. Figure 7 shows a fragment of the ECG signal for record 111 from the MIT-BIH database (solid curve), as well as the first derivatives of the smoothed ECG signal (dashed curve).

As shown above, the local maxima (1) and minima (2) of the first derivatives corresponding to the vicinity of the R-tooth are the most pronounced among the tops of all the teeth. The next smaller absolute values are the local maxima (1') and minima (2') of the first derivatives in the vicinity of the top of the T-tooth. The third position in absolute value is occupied by the local maxima (1′′) and minima (2′′) of the first derivatives surrounding the top of the P-tooth. The tops of T and P-waves, similarly to the R and T waves, are automatically allocated between the corresponding local minimum and maximum of the first derivatives of the smoothed ECG signal.

The accuracy of ECG signal detection based on the proposed approach varies from 95% to 99.9% depending on the complexity of the original ECG signal. The use of this approach is limited in a situation where the same form of the ECG signal is interpreted by a specialist in different ways, depending on the additional features of the ECG signal, as well as other leads. There are also a number of unidentified signals that are interpreted individually. However, the proposed algorithm automatically identifies difficult fragments, which allows medical personnel to minimize the time spent on manual signal processing for subsequent diagnosis and medical decision making.

An important advantage of the proposed method is the ability to respond flexibly and quickly to possible changes in the shape of the cardiocycle occurring during the observation process, which is a necessary requirement for automatic ECG signal processing. These methods underlie the described ECG signal analysis method, which is reliable, flexible, and easy to implement.

The flexibility and ease of use of the proposed method make it possible to minimize the processing time of the ECG signal for automatic decision making and reliable automated diagnosis of cardiovascular diseases.

A combination of a wearable device with the ability to register an electrocardiogram and software for direct analysis of the ECG waveform morphology is possible.

The task of processing the ECG signal is solved by using both the methods of the phenomenological approach and the methods of statistics. Statistical processing of experimental data, as a rule, is aimed at building a mathematical model of the object under study. However, taking into account all minor aspects and relationships in the model can distort the description of the process, since it requires a lot of provisions and assumptions, as well as lead to a distortion of important features of the process, erasure of important small details, worse results and false conclusions in conditions of insufficient information about the object of study. In conditions where the assumptions when using the model are associated with the risk of distorting the conclusions about the process, methods for approximating experimental data should be developed that do not require assumptions about the processes under study and reproduce the experimental pattern regardless of the assumptions about the model. This approach to filtering noise and optimizing the smoothing algorithm does not require preliminary assumptions and assumptions, often based on the subjective ideas of developers about the ECG signal process under study, which can lead to false conclusions. The proposed solutions are confirmed experimentally and practically confirmed in comparison with known methods.

Efficient noise filtering without distorting the true dynamics of the ECG signal makes it possible to estimate with high accuracy the first and second derivatives of the process, which by their nature amplify the noise component in the data. Analysis of derivatives of the ECG signal greatly simplifies the analysis of the ECG signal, the selection of stable patterns and the detection of pathologies in automatic mode. An important property of the proposed method is the ability to respond flexibly and quickly to changes in the shape of the cardiocycle occurring during the observation process, which is a necessary requirement for automatic ECG signal processing. These methods form the basis of the described ECG signal analysis software, which is reliable, flexible, and easy to implement.

Efficient identification of patterns in the ECG signal makes it possible to analyze the choice of the optimal predictive model under conditions of high variability of the signal under study to build a short-term and long-term automated medical prognosis.

Claims (28)

1. A method for determining the characteristics of an electrocardiogram (ECG), comprising the steps of: receiving an ECG signal using an electrocardiogram recording device; obtaining, by means of the processing means, a smoothed ECG signal by filtering the noise of the received ECG signal; calculate, using the processing means, the first derivatives of the smoothed ECG signal; determining, by processing means, ECG characteristics based on first derivatives of the smoothed ECG signal; characterized in that based on the processing of the distribution of local extremes of the first derivatives of the smoothed ECG signal, using processing means, a threshold is determined to highlight the range of variation of extremes corresponding to different types of teeth; And using the processing means, the characteristics of the peaks of the teeth of different types are determined based on the selected ranges of the extremes and the relative position of the local maxima and minima of the first derivatives of the smoothed ECG signal. 2. The method for determining ECG characteristics according to claim 1, further comprising the steps of: calculate, using the processing means, the second derivatives of the smoothed ECG signal; And determine, using the processing means, the characteristics of the tops of the teeth of different types additionally based on the local maxima and minima of the second derivatives of the smoothed ECG signal. 3. The method for determining the characteristics of the ECG according to claim 1 or 2, further comprising the step of detecting, using the processing means, a possible pathology based on certain characteristics of the tops of different types of teeth. 4. A method for determining ECG characteristics according to claim 1 or 2, characterized in that the top of the R-wave is determined between the local maximum and the local minimum of the first derivatives of the smoothed ECG signal, taking into account the relative position of the local maximum and local minimum. 5. The method for determining the ECG characteristics according to claim 4, in which the R-wave is separated from monomorphic ventricular tachycardia, when premature contraction of the ventricles is observed, based on a change in the relative position of the local maximum and local minimum of the first and second derivatives of the smoothed ECG signal. 6. A method for determining ECG characteristics according to claim 1 or 2, characterized in that the tops of T-waves and P-waves are determined between local maxima and local minima of the first derivatives of the smoothed ECG signal, while R-waves, T-waves and P-teeth by the absolute value of the difference between the corresponding local maxima and local minima. 7. An electrocardiogram (ECG) characterization system, comprising: an electrocardiogram recording device configured to receive an ECG signal; processing means configured to: smoothing the ECG signal by filtering the noise of the received ECG signal; calculating the first derivatives of the smoothed ECG signal; And determining ECG characteristics based on first derivatives of the smoothed ECG signal; characterized in that the processing means is additionally configured to: determining, on the basis of processing the distribution of local extrema of the first derivatives of the smoothed ECG signal, a threshold for selecting the range of change of extrema corresponding to different types of teeth; And determining the characteristics of the peaks of different types of teeth based on the selected ranges of extremes and the relative position of the local maxima and minima of the first derivatives of the smoothed ECG signal. 8. The ECG characterization system of claim 7, wherein the processing means is further configured to: determining the second derivatives of the smoothed ECG signal; And determining the characteristics of the peaks of the teeth of different types additionally based on the local maxima and minima of the second derivatives of the smoothed ECG signal. 9. The system for determining the characteristics of the ECG according to any one of paragraphs. 7 or 8, wherein the electrocardiogram recording device is a stationary heart monitor, a Holter ECG monitor, a wearable ECG recording device, or a mobile device with an ECG recording function.

Images