KR980008168A - Queue of electrocardiogram signal. Display device and method of ARCS complex characteristic - Google Patents

Abstract

The present invention relates to an apparatus and a method for displaying a QRS complex characteristic of an electrocardiogram signal classifying the shape of a QRS complex generated by an action of the ventricle in an electrocardiogram signal generated by an activity of the kidney, And the number of samples of the processed data is reduced through down-sampling to reduce the amount of calculation for the QRS complex detection to improve the detection speed, thereby real-time processing is enabled, and the QRS complex In the preprocessing process for performing the characteristic display process, a proper reference point is set through the QRS complex adjustment process in the band-pass filtered ECG signal to set a new amplitude set for efficient clustering, thereby improving the accuracy of the overall characteristic display, By drastically reducing the number of nodes, .

In addition, by applying the fuzzy clustering method, we could obtain accurate characteristic display results by improving the classification error even in the poor QRS complex type classification environment which contains a lot of noise components. This result shows that accurate and efficient rhythm analysis It is also expected that it will be possible to extract accurate diagnostic parameters through representative bits. In addition, by determining an accurate dominant type for a QRS complex in a ventricular heterosis, etc., it is possible to solve the problem that a QRS complex having a large occurrence frequency is determined as a dominant QRS complex.

Description

Queue of electrocardiogram signal. ALS complex express display device and method

Since this is a trivial issue, I did not include the contents of the text.

FIGS. 1A to 1G are diagrams illustrating an electrical conduction system and an electrocardiogram generation process of the heart.

FIG. 2 is an exemplary diagram showing representative diagnostic parameters used in the automatic diagnosis of electrocardiogram. FIG.

Figure 3 is an illustration of a QRS complex type template preset for pattern matching;

4A is a waveform diagram showing an input electrocardiogram waveform;

Figure 4b is an exemplary diagram illustrating characteristic points for a single QRS complex.

FIG. 5 is an exemplary view showing typical shapes of characteristic points. FIG.

Figures 6a-6b illustrate the use of a reference amplitude set determined in a QRS complex to perform shape classification for a QRS complex.

Figs. 7a to 7g show an example of a lead configuration of a standard 12-lead electrocardiogram. Fig.

8A to 8B are waveform diagrams showing the frequency and phase characteristics of the band-pass filter used in the lead selection and the QRS characteristic display process.

FIG. 9 is a flow chart showing a method of displaying a QRS complex characteristic of an electrocardiogram signal according to the present invention. FIG.

FIG. 10 is a flow chart showing the morphological classification step of FIG. 9; FIG.

FIG. 11 is a distribution of the training set used to determine the cluster center value in the 2D feature vector space. FIG.

12a-12b illustrate waveforms and distributions used in the morphological classification step of FIG. 9, wherein the first QRS complex is used as the reference QRS complex to perform morphological classification for the other QRS complexes.

Figures 13a-13d show the waveforms and distributions used in the morphological classification step of Figure 9, which performs the morphological classification of the QRS complex using fuzzy clustering for leads I, II, and III in # 32 patients in CSE database set 3.

Figures 14a-14b illustrate the waveforms of the input ECG signals used in the type classification step of Figure 9 to perform the type classification of the QRS complex using fuzzy clustering of leads # V1, V2, V3 of # 32 patients in CSE database set 3 Distribution chart for membership function.

FIG. 15 is an exemplary diagram showing the waveform and shape classification result of an electrocardiogram signal in which ventricular heterosis occurs. FIG.

DESCRIPTION OF THE REFERENCE NUMERALS

10: QRS complex determination step 100: Reference point and amplitude set determination step

110: Bandpass filtering step 130: Lead selection step

140: Reference point determination step 150: Amplitude set determination step

200: type classification step 210: first type classification step

220: second type classification step 240: pending decision type classification step

250: characteristic vector setting and fuzzy clustering step 300: dominant type determination step

The present invention relates to an apparatus and method for automatically diagnosing an electrocardiogram for detecting an electrical signal from a heart in the field of bio-signal processing, and more particularly, to an apparatus and method for automatically diagnosing an electrocardiogram And more particularly, to a QRS complex property display apparatus and method for displaying an electrocardiogram signal characteristic of a generated QRS complex (hereinafter referred to as a QRS complex).

Electrocardiogram (ECG) is a record of the activity of four chambers composed of left and right atria and ventricles. By attaching electrodes at specific locations in the body, electrical activity of the myocardium can be extracted through the body surface It is widely used as a clinical diagnostic tool for heart.

1A shows the electrical conduction system of the heart and the electrocardiographic process. As shown in FIG. 1A, the heart is largely divided into left atrioventricle (LA) and right atrioventricular (RA) The heart's electrical system consists of a sinoatrial node (SA node), which induces an impulse to initiate the contraction of the heart and adjusts the heart rate, and a right ventricle (LV) An atrioventricular node (AV node), a bundle of his bundle and a bundle branch of a right bundle branch and a perkinous bundle branch which provide a path for transferring an electrical impulse generated in the left and right ventricles (LV, RV) Network (Firkinse finger network).

When the atrioventricular node (SA node) generates an electric impulse and depolarizes the atria, the impulse propagates through the atria node (AV node) and the bundle of his bundle of his bundle branch ), And then delivered to the ventricle through the Perkins fiber network to depolarize the ventricles.

The electrical impulses from the sinus node (SA node) flow through the two atria, contracting the atria. At this time, the current generated as the atria are contracted is measured as a potential difference through the electrodes of the leads attached to the surface. As a result, a waveform having a small curve as shown by a solid line in FIG. 1B appears. This waveform is called P-pie.

When the impulse arrives at the atrial node (AV node), a slight delay occurs. The time interval including the delay time of the P wave is referred to as a PR interval, and the PR interval is followed by the impulse Through the heavy bundle between the atrium and the muscular diaphragm, the muscle diaphragm moves right and left.

The activity of the muscle septum as described above causes the first ventricular tortuosity, and it has the same waveform as the first d, which is called the Q wave.

Then, the impulse is rapidly diffused through the conduction structure of the ventricle, causing the largest tortuous wave of the ventricle. The wave generated at this time is the R wave, and the S wave is generated after the conduction process to all portions of the ventricle is finished.

As shown by the solid line in FIG. 1 (e), a waveform is formed in which a group of three waveforms, that is, Q wave, R wave and S wave, constitute a group. A group of these waveforms is referred to as a QRS complex. Pumped.

Thereafter, there is a relatively inactive period for a short period of time. In this case, the recorded ST segment is ST segment 1f. During the recovery period, the heart undergoes repolarization, and the waveform generated at this time is T It is a wave.

Thus, as a series of combined actions is repeatedly performed, abnormal electrocardiogram waveforms occur when a disease occurs in a series of electrical conduction systems. In other words, atrioventricular and third ventricular tachycardia occurs when the sinus node (SA node) does not work, and when the ventricle itself is abnormal, it manifests as ventricular contraction or ventricular fibrillation.

As described above, the QRS complex in the electrocardiogram signal is one of the most important parameters in the automatic diagnosis of electrocardiogram (ECG), which is a waveform generated as a result of the ventricular activity of the market.

Therefore, the result of automatic diagnosis of electrocardiogram using a computer is classified into a group of morphological types of a QRS complex and a series of characteristic display processes such as extracting diagnostic parameters for an electrocardiogram signal therefrom (that is, whether the QRS complexes are the same or different The labeling process).

As shown in FIG. 2, the diagnostic parameters used in the electrocardiogram automatic diagnosis apparatus include the interval (PR _d ) from the start point of the P wave to the start point of the QRS complex, the start point and end point interval (QRS _d ) of the QRS complex, (Q _d ), the Q wave interval (Q _d ), the R wave interval (R _d ), the S wave interval (S _d ), the interval between any R wave and the next R wave immediately adjacent thereto of the parameter related to the time (interval), such as RP interval, such as a Q-wave jinchok (Q _a), R-wave amplitude (R _a), S-wave amplitude (S _a), T-wave amplitude (T1 _a, T2 _a) It can be divided into parameters related to amplitude.

In addition to the above parameters, a QRS complex type, a P wave type, and the like are also used, but it is general to diagnose an abnormal state and a disease type of the heart using parameters related to time and amplitude.

In the conventional electrocardiogram, the diagnostic parameter extraction method used in the automatic diagnosis is to diagnose the electrocardiogram by individually obtaining diagnostic parameters for all bits (waveform including P wave, QRS complex, and T wave) Diagnostic parameters are the mainstream.

Especially, extracting the diagnostic parameters related to the QRS complex among the plurality of diagnostic parameters related to the electrocardiogram is the most basic requirement in the automatic diagnosis of electrocardiogram.

In the process of extracting the diagnostic parameters related to the QRS complex, the good classification characteristics in the QRS complex characteristic display step of classifying the type of the QRS complex have a decisive influence on the diagnosis result.

Generally, the most widely used QRS complex property display methods include a method of classifying the shape of a QRS complex using a pattern matching through a template, and a method of classifying the shape of a QRS complex using an amplitude set of the QRS complex And the like.

Hereinafter, a QRS complex property display method for classifying the shape of a QRS complex using pattern matching through a template according to the related art will be described.

The QRS complex property display method using pattern matching is a method of extracting the characteristic points of each signal using the gradient component of the electrocardiogram signal and then performing pattern matching with the already set QRS type template to determine each QRS complex type , A pattern matching process is performed to find out what type of template each of the input QRS complexes is matched with using a typical QRS complex type template as shown in FIG. 3, and then each QRS complex type Classify.

4a to 4b and 5, a graphical representation of the input electrocardiogram waveform is shown in FIG. 4a, characteristic points for a single QRS complex are shown in FIG. 4b, Representative examples are shown.

In determining the characteristic point, it is well known that a proper point should be selected for a point that can represent the attribute of the ECG signal most implicitly. Usually, the starting point and the end point of each wave are selected as characteristic points .

For example, the characteristic point 1 of FIG. 4b has a slope having a property of down (d) from horizontal (h), and a characteristic point 2 has a slope of up (u) Attribute slope. In a similar way, characteristic point 3 has a slope with an attribute of rise (u) - fall (d) and characteristic point 5 has an attribute of rise (u) - horizontal (h).

Therefore, a QRS complex having these morphological attributes is compared with a preset template to determine the corresponding QRS complex type, and such a QRS complex type classification is applied to all QRS complexes.

After determining the shape of all the QRS complexes using the pattern matching process, it is determined whether or not the QRS complexes having different shapes are included in the inputted ECG waveform using the determined shape.

However, the primary problem in the QRS complex type classification method using pattern matching is that it is very sensitive to noise components. Since most of the noise components have a peak shape, they are recognized as characteristic points of the QRS complex introduced into the QRS complex, which causes the error in the shape recognition of the QRS complex.

Although the noise component can be removed through the filtering process, only the noise components removed are mostly removed with a large peak component, and a noise component with a small peak component remains. In addition, additional noise removal processes are added to remove the noise components even after the filtering. By adding such a process, improvement in pattern matching can be expected. However, as the amount of calculation to be processed increases, .

In addition, since the pattern matching process is applied independently to each lead, it becomes a great obstacle to the real-time implementation of the automatic diagnosis apparatus in which the 12-lead ECG signal is used as a mode.

Furthermore, since only the information on whether a QRS complex having a different form exists among the QRS complexes of the inputted electrocardiogram signal is required in the process of displaying the QRS complex property, an accurate QRS complex pattern is obtained as described above Using this to classify the type of QRS complex is virtually inapplicable to automatic diagnostics as it involves more computation than is necessary.

6A to 6B illustrate a QRS complex property display method of classifying the shape of a QRS complex using amplitude coupling, which is another example of a conventional QRS complex property display method.

6a shows a reference amplitude amplitude determined in the QRS complex and FIG. 6b shows a cross-correlation between the reference amplitude jop determined in FIG. 6a and the amplitude set of each QRS complex, By way of example, the reference amplitude sets in the QRS complex are treated as a kind of feature vector, and the value of the cross-correlation expression for these feature vectors is used as a measure to determine whether each QRS complex is of the same type Classify it as a different type.

After determining the starting point and ending point of the QRS complex in the circular ECG signal, as shown in FIG. 6A, the amplitude set is determined in these intervals, and when the amplitude set for all the QRS complexes is provided, the reference QRS complex is determined, Obtain the cross-correlation of the reference QRS complex and the other complex to the amplitude set. If the value of the cross correlation is 1 for the two QRS pairs to be compared, the QRS complex is completely matched. If it is 0, the QRS complex having the different form is determined. That is, Classify the QRS complex.

However, although the QRS complex property display method using the amplitude set has an advantage that the influence of the noise component is relatively small, the performance of the partition property is determined by setting the reference point and the amplitude set for obtaining the cross- It is greatly affected.

In other words, increasing the number of amplitude sets is the most effective measure to minimize the influence of noise components, but the amount of computation required to obtain the cross-correlation with the number of increased set of waviness is proportional to the number of amplitude sets Which is a direct obstacle to real-time implementation. On the other hand, when using a relatively small number of sets of axes, the amount of computation can be reduced, but the problem is that it is susceptible to noise components.

In the reference point determination process, it is common to select a point having a maximum absolute value (most of R waves) in the QRS complex as a reference point for applying the cross-correlation equation, and some QRS The precision of the reference point selection is lowered for the complexes, and as a result, it is difficult to expect the accurate QRS complex property display result. In a signal having a small signal-to-noise ratio (SNR), it is difficult to classify the exact QRS complex type only by the correlation set of the amplitude set.

Furthermore, since the process of calculating the cross correlation for the amplitude sets is independently applied to each lead, it takes a lot of time, which is a great obstacle to the real-time implementation of the automatic diagnosis apparatus in which 12-lead ECG signals must be used.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a novel feature vector and a cost function which improve the accuracy of the overall characteristic display and a new feature vector And a cost function, and at the same time, the number of classification nodes is greatly reduced to shorten the processing time, thereby providing an accurate QRS complex property display apparatus and method.

In order to achieve the above object, an apparatus and method for displaying a QRS complex characteristic of an electrocardiogram signal according to the present invention uses only ten frequency sets of the amplitude set of band-pass filtered signals, And endpoints to ensure that some of the amplitude is not determined outside the QRS complex when determining the focus assembly and a new amplitude vector that reflects the correlation values, power ratio, maximum, and minimum amplitude ratios for the three leads In addition, in order to enable simultaneous real-time processing while using the type classification method, it is necessary to apply the fuzzy clustering method to the heart of the 12 electrocardiogram leads. Three leads with similar origins into one lead group For each of these groups, the lead order is determined according to the degree of amplitude, and the QRS complex property display process is applied only to the highest priority lead. In this way, a real-time implementation aspect is provided, and a signal- to-noise ratio (SNR) is similar to that of a small QRS complex, it is necessary to use a fuzzy clustering method to solve the error of the type classification that occurs when there is a slight change in shape, It is characterized in that the performance of the characteristic display is improved by checking whether it is difficult to form.

Hereinafter, an apparatus and method for displaying a QRS complex characteristic of an electrocardiogram signal according to the present invention will be described.

First, the configuration of the multiple leads of the present invention will be described.

Electrocardiograms are spatial representations of various cardiac vecors recorded in the frontal, horizon, and sagittal planes. These planes are generally located at nine lead positions, .

The nine lead positions include a right arm RA, a left arm LA, a left leg LL, a right sternal end V1, a left sternal end V2, The center line V3 of V2 and V4, the front side V5 of the left armpit, and the center line V6 of the left armpit.

In other words, typical electrocardiographic leads have three limb leads (RA, LA, LL) attached to the right arm (RA), left arm (LA) The left chest tip V2 and the left clavicle center line V4 and the center line V3 of V2 and V4 to the right chest lead V1, The front side V5 of the left armpit, and the center line V6 of the left armpit.

In addition, the lead is attached to the right leg (RL) but is used only as the ground. That is, in the measurement of a general electrocardiogram signal, a two-input differential amplifier is used. Since the differential amplifier requires a ground reference, the electrode attached to the right foot (RL) is connected to the internal ground of the automatic amplifier and used as a ground reference.

The limb leads RA, LA and LL are connected to three bipolar leads (leads I, II and III) and three unipolar leads (lead aVR, augmented unipolar left, aVF) And the chest leads are used to obtain six unipolar leads V1, V2, V3, V4, V5 and V6.

Here, bipolar refers to two practical electrodes, one connected to the (+) input of the differential amplifier and the other connected to the negative (-) input, and unipolar is one Electrode means that one electrode is connected to the (+) input terminal and two or three electrodes are connected to the (-) input terminal.

Hereinafter, the configuration of the electrode to which the leads are connected will be described with reference to Figs. 7a to 7g.

As shown in FIG. 7a, the lead I is connected to the (+) input terminal of the differential amplifier and the left hand lead is connected to the (-) input terminal of the differential amplifier to measure the electrocardiogram from the output terminal of the differential amplifier, Lead II is connected to the (+) input terminal of the differential amplifier and the right hand lead is connected to the (-) input terminal of the differential amplifier to measure the electrocardiogram from the output terminal of the differential amplifier as shown in FIG. 7b, , The left foot lead is connected to the (+) input terminal of the differential amplifier, the left hand lead is connected to the (-) input terminal of the differential amplifier, and the electrocardiogram is measured from the output terminal of the differential amplifier, as shown in FIG.

The lead aVR is connected to the (+) input of the differential amplifier through resistor R / 2 and the right hand lead and the left foot lead are connected to the (-) terminal of the differential amplifier through resistor R, respectively, The lead aVR is connected to the (+) input terminal of the differential amplifier through the resistor R / 2, and the left hand lead is connected to the (+) input terminal of the differential amplifier through the resistance R / The left foot lead is connected to the (-) input terminal of the differential amplifier through the resistance R, and the electrocardiogram signal is measured from the output terminal of the differential amplifier. The lead aVF measures the left foot lead by the resistance R / (+) Input terminal of the differential amplifier through the resistor R and the right hand lead and the left hand lead are commonly connected to the (-) input terminal of the differential amplifier through the resistor R and connected to the output terminal of the differential amplifier The emitter ECG signal is measured.

As shown in FIG. 7 (g), the right lead leads, the left hand leads and the left foot leads are commonly connected to the negative input terminal of the differential amplifier through a resistor R, The lead is connected to the position (right sternal end (V1), left sternal end (V2), left clavicle center line (V4), V2 and V4 center line (V3), left armpit front (V5), left armpit center line And connected to the (+) input terminal of the differential amplifier through the resistor R / 3 to measure each electrocardiogram signal from the output terminal of the differential amplifier.

A method of detecting a QRS complex of an electrocardiogram signal using multiple leads according to the present invention sets and uses a lead group composed of a plurality of leads. Since a general electrocardiogram automatic diagnosis device receives and processes 12 lead electrocardiogram signals, As one embodiment of a lead group, a lead group is formed as follows using three leads having the same origin.

The first lead group is composed of I, II, and III leads of bipolar leads and the second lead group is composed of aVR, aVL, and aVF leads that are unipolar leads of a limb lead. V2, and V3 leads that are unipolar leads of the chest lead, and the fourth lead group consists of V4, V5, and V6 leads that are unipolar leads of the chest lead.

Hereinafter, a method of displaying QRS complex characteristics of an electrocardiogram signal according to the present invention using the above-described multiple leads will be described with reference to FIGS. 9 and 10.

As described above, the process of displaying the QRS complex property classifies the type of each QRS complex complex and determines whether there are several QRS complex groups having mutually different forms, thereby enabling accurate cardiac disease classification. Further, It is a preprocessing process for selecting the representative form to extract the parameters.

A method of displaying a QRS complex property of an electrocardiogram signal according to the present invention includes the steps of calculating a spatial velocity using a slope between adjacent samples obtained from a lead of a lead group composed of a plurality of leads, A QRS complex determination step (10) of determining a QRS complex from an electrocardiogram signal; a step (10) of determining a QRS complex from the electrocardiogram signal using information on the detected QRS complexes, and using a difference and a ratio between a maximum extreme value and a second largest extreme value A reference point and an amplitude set determination step (100) for determining an amplitude set such that the positions of the start point and the end point of the QRS complex are symmetrical about the reference point, a correlation value for the amplitude sets and a maximum amplitude A first vector sum (SF1) that reflects a value of the amplitude set and a power ratio (200) for classifying the shape of a QRS complex by performing fuzzy clustering in a feature vector space formed by a second vector sum (SF2) reflecting a width value of the QRS complex; A dominant type that determines the dominant type for the QRS complex by comparing each average PR interval according to the frequency of occurrences between the highest QRS complex type and the second highest QRS complex type among the classified QRS complex types, And a decision step 300.

The QRS complex determination step 10 includes a spatial speed setting step of calculating a spatial speed for an electrocardiogram signal using spatial characteristics of a lead group composed of a plurality of leads, And a QRS complex detection step of comparing the effective pole values detected in the effective pole value detection step with a threshold value and detecting the result with a QRS complex.

Wherein the spatial velocity setting step includes a downsampling step of reducing the number of processed sample data by performing downsampling on the original signal, a downsampling step of reducing the number of processed sample data by performing a downsampling on the three- Wherein the effective extremum detection step comprises a step of detecting a local extremum of the space velocity when the current amplitude value of the spatial velocity is greater than the previous amplitude value and the subsequent amplitude value, And a maximum extreme value detection step of detecting the largest local extrema among the extreme extreme values in which the spike components are removed within a predetermined interval, as the extreme extreme value.

The QRS complex detection step sets a threshold value proportional to an average value of the maximum effective poles determining the maximum effective value among unlabeled effective peaks of the QRS complex and labeling an extremum larger than the threshold value with the QRS complex A QRS complex labeling step and an adjacent extrema removal step for comparing the neighboring validity values existing within a predetermined interval from the labeled QRS complex with a threshold set to be proportional to the currently labeled effective extremum, .

The detailed description of the QRS complex determination step of the present invention as described above is described in detail in Korean Patent Application No. 96-18450, a QRS complex detection apparatus and method using multiple leads, .

The reference point and amplitude set determination step 100 includes a band pass filtering step 110 for performing filtering using a band pass filter for the QRS complexes respectively input from the plurality of leads, There is a lead satisfying the condition that the absolute value of the extreme value includes the largest extremum (maximum extremum) and at the same time, the second extremum of the sign larger than the predetermined ratio to the maximum extremum does not include the second extremum of the sign And if there is no extremum satisfying the above condition, the ratio of the maximum extremum and the second largest pole among the QRS complexes in each single lead (i.e., A reference point determining step (140) of determining a maximum extreme value point of the lead as a reference point; a step of determining a start point and an end point of the QRS complex based on the reference point And an amplitude set determination step 150 for determining the amplitude set to be symmetric with respect to the center.

The shape classification step 200 calculates a first vector sum defined as a vector sum of a product of a correlation value for a pair of QRS complexes and an extreme value of a correlation value in an arbitrary lead group and a relative maximum amplitude value, (SF1), a power ratio of a power ratio to a power ratio of 1 to a power ratio of a QRS complex pair, and sets a second vector sum (SF2) defined by a vector sum of a product of the relative maximum amplitude value and a relative maximum amplitude value, Fuzzy clustering is performed on the feature vector space having the vector sum and the second vector sum as axes so that the reference QRS complex and the rest of the QRS complexes other than the reference QRS complex are subjected to fuzzy clustering on the reference QRS complex Classification of the type of QRS complex into groups of similar morphologies with the same morphology, different morphology groups with different morphologies, groups of similarity to morphology, And a fuzzy clustering step (250) of determining a QRS complex type according to a criterion of a decision hold type classification that classifies a type of a QRS complex belonging to the decision hold type group according to a criterion of QRS complex type classification of other lead groups And a decision hold type classification step 240 for classifying the type of the QRS complex belonging to the group.

Herein, the feature vector setting and fuzzy clustering step 250 may be performed by using the remaining QRS complex excluding the reference QRS complex and the reference QRS complex using the ten of the predefined clusters in the feature vector space in the QRS complexes constituting an arbitrary lead group. A QRS complex having a membership function value included in a predetermined upper and lower intervals around 0.5 is classified into a group of uncertainty types, and a first membership function Classifying a QRS complex having a function value into a group of undetermined types, classifying a QRS complex having a first membership function value of the predetermined interval or more into the similar type group, and classifying the QRS complex having a first membership function value A first type classification step (210) of classifying the first type If there is more than one QRS complex classified as the uncertainty type group after performing the classification step 210, the QRS complexes classified into the uncertainty type group and the similar type group may be classified Calculating a feature vector between the QRS complexes and determining a second membership function reflecting the degree of mutual similarity so that a part of the second membership function values are included in a predetermined section up and down around 0.5, Classifying the preexisting QRS complex into the set of the pending retention types if all of the second membership function values have a value equal to or greater than the predetermined interval; and if the preexisting QRS complex has the second membership function value, And if all of the second membership function values have a value equal to or smaller than the predetermined interval, A second type classification step (220) of classifying the QRS complex into the different morphological types; and, if there are two or more QRS complexes belonging to the morphological type group after performing the second morph classification step (220) Type classification step of performing the first type classification step 210 and the second type classification step 220 between the QRS complexes belonging to the morphological group.

Another embodiment that can replace the type classification step 200 is to define the sum of the maximum and minimum values of the relative amplitudes for each QRS complex constituting the lead group, A selection step of selecting a plurality of lead ECG signals and a correlation, a power ration, a maximum ratio to peak amplitude, a relative amplitude to amplitude ratio, a maximum value of a relative amplitude And a morphological group determination step of classifying the morphology of the QRS complex using the minimum value.

The dominant type determination step 300 includes an occurrence frequency calculation step 310 for calculating the occurrence frequency of the QRS complexes belonging to each type group classified in the type classification step 200 and the most frequently occurring QRS complex A frequency determination step (320) of determining whether a frequency of occurrence of a type (maximum frequency of occurrence) is equal to or more than three times the frequency of occurrence of a second most frequently occurring type of QRS complex; A first dominant type determination step (330) of determining a QRS complex type having the maximum occurrence frequency as a dominant type if the maximum occurrence frequency is three times or more the second highest occurrence frequency, The average RR interval in the form of a QRS complex with the second highest occurrence frequency and the average RR interval in the form of a QRS complex with the highest occurrence frequency Going to the first dominant type determination step if it is less than 150 ms (150 milliseconds) and a second dominant type determination step 360, otherwise determining the QRS complex complex type with the second highest occurrence frequency as the dominant type, .

The first type classification step 210 includes a first feature vector calculation step 212 for calculating a feature vector for a first reference QRS complex for setting the reference QRS complex, A first membership function determination step (213) of determining a first membership function reflecting the degree of similarity between the QRS complexes excluding the QRS complex and the first reference QRS complex; A first membership function decision step (214) for determining whether the first membership function value is between 0.4 and 0.6 or less than 0.4, classifying the QRS complex having the first membership function value of 0.6 or more into a similar type group, And a first classification determination step (215) of classifying the QRS complex having a function value into a different morphology group, and the second morph classification step (220) is a step of determining whether or not the preexisting QRS complex exists And if the QRS complex complex belonging to the uncertainty type group exists as a result of the determination of the uncertainty type presence / absence 221, the first reference QRS complex is changed to the uncertainty QRS complex A second reference vector calculation step 223 for calculating a feature vector between the QRS complexes classified into the similar type group and the second reference QRS complex, A second membership function determining a second membership function reflecting the degree of similarity between the QRS complexes pre-classified into the similar type group and the second reference QRS complex using the cluster center value on the feature vector space, Step 224, determining whether the value of the second membership function is greater than or equal to 0.6, less than or equal to 0.4, some ranging between 0.4 and 0.6, And a second membership function determination step (225) for determining whether or not a preexisting QRS complex having a second membership function value equal to or greater than 0.6 is determined to be in a different group, And a second classification determination step (226) for classifying the microcrystalline QRS complex into a classification suspending morphology group when a part exists between 0.4 and 0.6 and a part exists below 0.4 or 0.6 or more.

The decision hold type classification step 230 includes a QRS complex existence determination step 231 for determining whether there is a QRS complex belonging to the decision hold type configuration group and a QRS complex existence determination step 231 for determining whether the QRS complex existence determination step 231 determines, A reading step (235) of reading the membership function of the other lead group if the QRS complex belonging to the decision hold type group exists, and if it is classified as a similar QRS complex having a value of 0.9 or more in both of the other lead groups, Classifying a QRS complex belonging to the morphological group into the morphological group and classifying the QRS complex belonging to the morpheme morphological group into the morphological group if all of the QRS complexes are classified into different QRS complexes having a membership function value of 0.1 or less, Having a membership function value between 0.1 and 0.9, the QRS complex belonging to the decision hold type group can be used in the diagnosis of electrocardiogram A decision hold classification step 233 of not using the QRS complex, and a step of repeating the third type classification step when the QRS complex belonging to the different type group is generated again.

Before describing the procedure of the method of displaying the characteristic of the electrocardiogram signal according to the present invention in detail, the necessity of determining the reference point will be briefly described. The reference point determination is a determination of a reference point necessary for obtaining the amplitude set When the reference points of the QRS complexes coincide with each other, the selection of an accurate reference point is the most basic requirement for securing the classification characteristics of the reliable QRS complex as the cross-correlation value becomes the maximum.

Generally, in an automatic electrocardiogram diagnosing apparatus, since a plurality of lead inputs are simultaneously received and signal processing is performed for each lead group constituted by a set of leads, not only a reference point itself but also a lead for which a reference point is determined should be selected.

That is, as in the embodiment of the multiple lead configuration of the present invention, a lead having the most stable reference point is selected instead of obtaining the reference point independently for each lead in the lead group composed of three leads, Use this as a reference point.

Hereinafter, a procedure of a method of displaying a QRS complex characteristic of an electrocardiogram signal according to the present invention will be described in detail.

The spatial speed defined in the QRS complex determination step 10 is described in detail in Korean Patent Application No. 96-18450, a QRS complex detection apparatus using multiple leads, and a method. However, I will quote.

When three leads constituting a single lead group are represented by X, Y and Z, by using each slope for three leads,

The vector sum in the three-dimensional space is defined as Equation (1), and it is used as a detection function for detecting the QRS complex by designating it as a spatial velocity (SV).

Where X _{i + 1} is the i + 1th value of the input lead X, X _i-1 is the i-1 th value, SV _i is the i th spatial velocity, and i is the point it means. It is obvious that the suffixes (i + 1, i-1) assigned to lead Y and lead Z also have the same meaning as those given to lead X.

Therefore, (X _{i + 1} -X _i-1 ) means the velocity component between two points because the sampling rate is constant, and SV _i is determined by the velocity components of the three leads (X, Y, Z) Dimensional vector sum of the three-dimensional vector sum.

In the present invention, formula (1) the sampling frequency of the original signal of an electrocardiogram to the real-time processing before applying the detection method using the same space velocity (SV _i) s × to say 100Hz, a sample point (sample point of an original signal ) Are downsampled to reduce the number of samples by dividing by the value of s, and the final sample data is obtained to reduce the amount of calculation in calculating the communicative rate for the QRS complex detection.

That is, the downsampled data ED is determined by the following equation.

Where E is the value of the original signal and ED _j ^k is the j th downsampled data from lead k, where k is 12 leads (i.e., I, II, III, aVR, aVL, aVf, V1, V2 , V3, V4, V5, and V6), and j represents any one of j = 1, 2, 3, to be.

That is, when the original signal inputted for about 2 seconds at a sampling rate of 500 Hz is downsampled to 100 Hz, n = 1000 and s = 5, and the QRS complex is detected using about 200 data.

Of course, it is a well-known fact that when determining the s-value, which is a down-sampling factor, the actual data for detection is lost and must be appropriately selected so as not to fall into the local minimum at the time of QRS complex detection .

Further, in order to reduce the amount of calculation required to obtain the above-mentioned spatial velocity,

Equation (1) is replaced by the formula of sum of squares, and the equation containing the root of sum of squares is substituted for the sum of absolute values.

The meanings of the signs and suffixes in Equation (3) are the same as in Equations (1) and (2).

On the other hand, in order to roughly determine the positions of poetic points and end points of the QRS complex so that an accurate amplitude set can be determined from the QRS complexes, in the QRS complex determination step 10, After determining the spatial velocity, the QRS complex is determined using the maximum extremes of the spatial velocity, and then the position corresponding to 25% of each maximum extremum is determined as the start point and the end point.

After performing filtering using a band pass filter having frequency characteristics and phase characteristics as shown in FIGS. 8a to 8b for the QRS complexes respectively input from the plurality of leads, the lead selection step 130 Selects a lead satisfying the condition that does not include the second extremum of the same code that includes the maximum extrema among the QRS complexes and is greater than 50% of the maximum extrema, and if the extremum satisfying the above condition If not present, the ratio of the maximum extremum and the second largest extremum among the QRS complexes in each single lead (i.e., ) Is the maximum is selected.

Then, in the reference point determination step 140, the maximum extreme point of the lead is determined as a reference point and stored.

The average duration of each QRS complex is about 90 ms, so that it is possible to fully characterize the QRS complex with 10 amplitude values per lead.

However, these ten amplitude values must be taken within the QRS complex, and the amplitude values contained in the PR interval or the ST segment should not be included in the ten amplitude values.

Accordingly, when the reference point is determined in the reference point determination step 140, the amplitude set determination step 150 determines the amplitude set to be symmetric about the reference point by checking the reference point and the positions of the start and end points of the QRS complex.

When the amplitude set is determined in the reference point and amplitude set determination step 100, the amplitude classification is used to perform a morphological classification of the QRS complex. First, in the preferred embodiment of the present invention, Typical characteristic parameters are correlation, power ration, maximum peak-to-peak ratio, relative amplitude-to-amplitude ratio, maximum and minimum values of relative amplitude.

First, the correlation for the amplitude sets of two different QRS complexes with ten amplitude values as components

Expression is as shown in (4), wherein, Cor _1.m (K) will represent the correlation value between the i th QRS complex, and the complex m-th QRS Complex of lead k, A (k, i, m) is k lead (K, i, l) represents the i-th amplitude value in the i-th QRS complex of the k-lead.

The power ratio for the amplitude set of two different QRS complexes with 10 components is

(5), where Pow _1.m (K) represents the power ratio between the i-th QRS complex of the k-lead and the m-th QRS complex of the k-lead.

The maximum amplitude to amplitude ratio

(6) under the condition that PP (k, m) is greater than or equal to PP (k, l). Here, PP (k, m) is k will showing the maximum amplitude value of the m-th QRS complex in the second lead, PP (k, l) will showing the maximum amplitude of the i th QRS complex of the k-th lead AmP _1, m (k) represents the maximum amplitude-to-amplitude ratio between the I-th QRS complex of the k-lead and the m-th QRS complex complex of the k-lead.

The relative amplitude to amplitude ratio for each QRS complex constituting any lead group is

(7). Here, PP (k, m) represents the maximum amplitude value of the mth QRS complex of the kth lead, Ramp (k) represents an amplitude value having the maximum amplitude among the amplitude sets for the mth QRS complexes of the lead group composed of j leads (j = 3 in this embodiment, that is, three leads) The lead constituting any of the lead groups has a maximum amplitude value of the mth QRS complex of the k leads The relative amplitude ratio between the two.

The maximum and minimum values of the relative amplitudes for each QRS complex constituting an arbitrary lead group are

Ramp (k) = MAX [Ramp (k, m), Ramp (k,

Qamp (k) = MIN [Ramp (k, m), Ramp (k,

(8) and (9), respectively. Here, MAX (Ramp (k, m), Ramp (k, l)) represents the largest value among Ramp (k, m) and Ramp Ramp (k, m) and Ramp (k, l) among the minimum values Ramp (k, m) and Ramp (k, l).

However, in the case of a lead having a QRS complex with a small amplitude, the shape of the waveform is significantly different due to a small change in the position of the heart even when the QRS complex is operated in the same sequence.

According to another embodiment of the present invention, in the type classification step, in order to determine the order of leads using the relative amplitude value as a parameter in the dominant lead selection step

RQ (k) = Ramp (k) + Qamp (k) (10)

The sum of the maximum and minimum values of relative amplitudes as in Eq. (10) is defined and used in the classification of the QRS complex.

That is, after defining the lead having the largest value as L1, the lead having the second largest value as L2, and the lead having the smallest value as L3, five parameters described above only for the lead L1 in the above- (I. E., Correlation, power ratio, maximum amplitude to amplitude ratio, relative amplitude to amplitude ratio, maximum and minimum values of relative amplitude).

However, for an electrocardiogram signal having a low signal-to-noise ratio (SNR) for all leads (three leads) constituting a lead group, or for an electrocardiogram signal having a waveform of a similarity but different in the form of a QRS complex, Results are shown.

Such incorrect classification results are due to the optimization problem of the selected amplitude set and the deterministic classification scheme based on binary logic.

Particularly, although the deterministic classification method has a waveform of the same shape, a QRS complex of an electrocardiogram signal including a slight change in morphology results in a negative morphological classification result, in which a part of the QRS complex is classified into the same waveform and a part of the QRS complex is classified into another waveform .

Accordingly, as shown in FIG. 10, in the type classification step 200 of the present invention, the type of the QRS complex is classified using a newly defined feature vector and a fuzzy clustering method.

In the present invention, the fuzzy clustering method using the FCM algorithm (Fuzzy C-Means Algorithm) is applied, and a detailed description of the FCM algorithm will be given with reference to the fuzzy theory and application of Kim Tae-yoon,

The new feature vector used in the present invention is

The meanings of the signs and subscripts in equations (11) and (12) are the same as in equations (4), (5) and (8) .

The rationale for the usefulness of the feature vector is that it can effectively classify similar QRS complex pairs in leads with small correlation values (Cor (k)) and power ratios (Pow (k)).

That is, [1- (Cor (k)] and [1-Pow (k)] have a relatively large value in a lead having a small correlation value and power ratio, but Ramp (k) has a relatively small value As a result, Ramp (k) and the correlation value and the power ratio reflect the compensation relationship for the mutual reduction ratio.

Fuzzy clustering is performed on each feature vector in the two-dimensional feature space having the two feature vectors SF1 and SF2 as axes, and the corresponding QRS complexes are classified into two classes (i.e., similarity or difference) ). At this time, a membership function for the cluster center value of each class is used to check whether or not each feature vector belongs to the class.

Meanwhile, if the clustering process (particularly, the process of obtaining the cluster center value) is performed on the electrocardiogram signal input every time, it takes a lot of time. Therefore, the cluster center value is first determined through a predetermined training set .

In order to determine the center value for each cluster in the feature vector space using the fuzzy clustering, a well-defined training set is required. Therefore, the CSE (Common Standards for Quantitative Electrocardiography) data base set 3 was used, We use a total of 332 pairs of 112 complexes and 112 complexes that differ from similar 220 pairs except for the values of the feature vectors SF1 and SF2 of less than 0.05 and the values of over 2.0, respectively, as a training set to determine the cluster center value.

The reason for excluding the value of less than 0.05 and the value of 2.0 or more is to maximize the distribution of the population of clusters, and the basis for determining these values is 0.1523 and 0.0995 according to the experimental values, respectively, in the similar QRS complex group , And the mean and standard deviation for the different complex groups are 1.249 and 0.550, respectively.

11 shows the distribution of the training set in the feature vector space and the determined center of the cluster, wherein the + markings represent similar QRS complex pairs, the o designation represents a different complex pair, and the center of each QRS complex group The * mark on the center indicates the determined cluster center value.

The center values for similar QRS complex pairs determined from the training set using the CES (Common Standards for Quantitative Electrocardiography) database data set 3 are shown in FIG. 11 as the feature vector space coordinates SF1 , SF2), and the center values for the different QRS complexes are the coordinates (0.9610, 0.9631) on the feature vector space coordinates.

The classification of each QRS complex pair is performed using the determined cluster center value. SF1 and SF2 of each QRS complex pair are obtained, and the degree of assurance (i.e., membership function) for the two groups is obtained. To determine whether these QRS complex pairs are similar or different.

12a to 12c show waveforms and distribution diagrams used in the type classification step 200 in which the first QRS complex complex is used as the reference QRS complex and the other QRS complexes are classified. FIG. 12B is a distribution diagram showing feature vectors SF1 and SF2 determined based on the first QRS complex, and FIG. 12C is a diagram showing a membership function indicating the degree of similarity for each pair of QRS complexes.

In Figure 12b, (1, I) represents the first QRS complex and the first QRS complex, I is I = 2, 3, 4, ... 10 and the mark * (Sim) is a group of similar QRS complexes Center value, and the marker * (Dif) indicates that the cluster center value of different QRS complexes is determined using the training set.

As can be seen in Figure 12c, it can be readily seen that the QRS complex (ie, the QRS complex with a relatively small value of the membership function) with a relatively small QRS complex affects the third, sixth, and ninth QRS complexes have.

In most cases, it is possible to accurately determine whether or not the QRS complexes constituting the QRS complex pair are similar or different using the membership function using SF1 and SF2 as described above. However, the QRS complex pair It is ambiguous to make a judgment.

That is, there may be a case in which there is a slight change in form with respect to the reference QRS complex, and a case in which the form is different from the reference QRS complex and appears almost the same in all the leads. In the case of the former case, distortion occurs in some QRS complex waveforms due to large noise components, and when the signal-to-noise ratio (SNR) in all leads is too small, have.

A morphological classification step 200 for performing accurate morphological classification for a QRS complex pair that can not make such a definite determination will now be described in detail with reference to FIG.

First, after the first QRS complex is set as the reference QRS complex in the reference QRS complex setting step 211, the first feature vector calculating step 212 calculates fuzzy clustering for different QRS complexes in the two- Determines whether the first membership function value is greater than 0.6, 0.4 to 0.6, or less than 0.4, and determines whether the first membership function value is greater than or equal to 0.6, In the first classification decision step 215, the QRS complex having the first membership function value of 0.6 or more is classified into the similar type group, and the QRS complex having the first membership function value of 0.4 to 0.6 is classified into the undecided type group And classifies the QRS complex having the first membership function value of 0.4 or less as a different type group.

If there is a QRS complex belonging to the uncertainty type group as a result of the uncertainty type presence / absence judgment step 221 after determining whether the uncertainty type presence / absence presence step 221 exists, The QRS complex is transformed into the indeterminate QRS complex to be a second reference QRS complex and the second feature vector calculation step 223 calculates a feature vector between the QRS complexes classified into the similar type group and the second reference QRS complex And a second membership function determination step 224 uses the cluster center value on the feature vector space to reflect the degree of similarity between the QRS complexes classified into the similar type group and the second reference QRS complex And determines a second membership function.

In the second membership function decision step 225, it is determined whether or not the second membership function values are all 0.6 or more, 0.4 or less, some are 0.4 to 0.6, and some are 0.4 or less and 0.6 or more The second membership function determination unit 225 classifies the preexisting QRS complex complex having the second membership function value of 0.6 or more into the similar type group and the second membership function value of 0.4 or less If the second membership function value is present between 0.4 and 0.6 and some is less than 0.4 or 0.6 or more, the preexisting QRS complex is classified as the classification pending type group Classify.

Thereafter, as a result of the QRS complex existence determination step 231 for determining whether there is a QRS complex belonging to the decision hold type group, the decision hold classification step 233 determines A QRS complex belonging to the decision hold type group is classified into the similar type group and all QRS complexes having a membership function value of 0.1 or less are classified into different QRS complexes, If the QRS complex belonging to the different morphology group is generated again without using the QRS complex belonging to the decision hold type group in the electrocardiogram diagnosis and if the QRS complex belonging to the morphology group is generated again, And repeats the third type classification step.

Figures 13a-13d show waveforms and distribution diagrams used in the type classification step 200 of the QRS complex using fuzzy clustering for leads I, II, and III in # 32 patients in CSE database set 3, (SNR) of the electrocardiogram signal is smaller than that of the electrocardiogram signal, and the result of the upper part in the classification result of the lower part is a classification result obtained by using NPPA (Nonparametric Partitioning Algorithm) method. And the results classified by the type classification step 200 are shown.

Here, the NPPA method for performing the performance comparison will be briefly described. The NPPA method is a method based on binary logic. If the NPPA method is larger than a predetermined threshold value, the NPPA method is classified into a similar form. Otherwise, the NPPA method is classified into a different form It is one of the representative type classification methods. The threshold is determined statistically using entropy in the feature space.

In fact, all of the QRS complexes of Figure 13a should be classified into a similar morphology group as shown in Figure 1, using the NPPA approach, undecisive as indicated by the indication (u) in the second, fifth and tenth QRS complexes, ) Type group, which can be easily confirmed by the first membership function value for the reference QRS complex and each QRS complex pair of FIG. 13b. That is, since the first membership function for the markers (1, 2), the markers (1, 5), and the markers (1, 10) exists between 0.4 and 0.6 in FIG. 13b, they are classified as QRS complexes belonging to the undetermined form group do.

When the classification of all the QRS complexes for the reference QRS complex of FIG. 13a is over, the QRS complexes belonging to the undetermined form group are compared with the QRS complexes that are pre-classified into the similar type group. If it is determined that the QRS complex is similar to all of the QRS complexes (i.e., the membership function value for the similar function is 0.6 or more), the QRS complex is classified into the similar type group. If the second membership function value is 0.4 or less, The QRS complex belonging to the group of undetermined forms is determined as a group of QRS complexes belonging to the group of non-deterministic form if the second membership function value is present between 0.4 and 0.6, It is classified into the hold type group.

It can be seen in Figure 13c that the second, fifth, and tenth QRS complexes are determined to have the same shape as the reference QRS complexes, but the eighth QRS complex is finally determined as the crystal pending QRS complexes.

As shown in FIG. 13d, the finally sorted QRS complex may be classified into a group of similar morphologies such as a reference QRS complex (i.e., Mark 1), different morphological groups (i.e., values other than Mark 1) (Mark u), as shown in Fig.

Figures 14a-14b illustrate the result of type classification of the QRS complex using the group for lead # V1, V2, V3 of patient # 32 in CSE database set 3. As shown in FIG. 14B, it can be seen that all of the QRS complex pairs have a membership function value of 0.9 or more, so that the preexisting QRS complex belonging to the decision hold type group can be classified into the similar type group.

Hereinafter, the dominant type determination step 300 will be described.

The selection of most dominant types selects the type of QRS complex in which the highest frequency occurs in the type classification step as the dominant type.

However, as shown in FIG. 15, when the abnormal QRS complex is selected as the predominant type of the QRS complex as the abnormal QRS complex appears in the ventricular bigeminy more than or equal to the normal QRS complex number Is generated.

Accordingly, in the dominant type determination step 300 of the present invention, in order to prevent the dominant type selection from being delayed, the type classification result for the QRS complex of the type classification step 200, the occurrence frequency of the most frequently occurring QRS complex type If the frequency of occurrence of the second most frequent QRS complex type is not more than three times, the average value of the RR intervals for the two types of QRS complexes is calculated, When the average value of the RR interval of the type is longer than the average value of the most frequently occurring type by at least 150 ms, the second most frequently occurring type is determined as the dominant type. Otherwise, the dominant type is determined as the dominant type.

FIG. 15 illustrates a case where the occurrence frequencies of the normal QRS complex and the abnormal QRS complex are the same. When the abnormal QRS complex and the normal QRS complex are represented by the same number, the QRS complex belonging to the different type group represented by the mark 2 is the dominant type .

Hereinafter, an electrocardiographic signal characteristic display apparatus according to the present invention will be described.

A method of displaying a QRS complex property of an electrocardiogram signal according to the present invention includes the steps of calculating a spatial velocity using a slope between adjacent samples obtained from a lead of a lead group composed of a plurality of leads, A QRS complex determination means for determining a Q. Als complex from an electrocardiogram signal and a QRS complex determination means for determining a reference point using a difference and a ratio between a maximum extreme value and a second largest extreme value among the QRS complexes, A reference point and an amplitude set determining means for determining a set of amplitudes by selecting a certain sample so that the positions of the start point and the end point are symmetric with each other; a first vector sum reflecting the maximum amplitude value relative to the correlation value for the amplitude sets; By a second vector sum reflecting the power ratio to the sets and the relative maximum amplitude value A QRS complex type having the highest occurrence frequency among the QRS complex types classified by the type classification means and a QRS complex type having the highest occurrence frequency And a dominant type determining means for determining a dominant type for the QRS complex by comparing respective average RR intervals according to occurrence frequencies between QRS complex types.

Herein, the reference point and amplitude set determination means comprises a band pass filtering unit for performing filtering using a band pass filter for the input QRS complexes, and an absolute value of an extreme value among the QRS complexes filtered by the band pass filter Selecting a lead satisfying a condition that includes a largest pole value (maximum peak value) and at the same time does not include a second pole value of the same sign larger than the maximum pole value (50%), If not present, the ratio of the maximum extremum and the second largest extremum among the QRS complexes in each single lead A reference point determining unit for determining a maximum extreme value point of the lead as a reference point; a position corresponding to 25% of the maximum extremum of each QRS complex as a start point and a start point; And an amplitude set determiner for selecting 10 samples to be symmetric about the reference point as an end point to determine an amplitude set.

The shape classification means obtains the difference component between 1, which is the extreme value of the correlation value, and the correlation value with respect to the QRS complex pair, calculates a first vector sum defined by a vector sum of products of the relative maximum amplitude value and an extremum And a second vector sum defined by a vector sum of the product of the relative maximum amplitude value and a relative maximum amplitude value to set a first vector sum and a second vector sum as an axis A fuzzy clustering is performed in a vector space, and a similarity type group having the same morphology with respect to the reference QRS complex according to the degree of similarity between the reference QRS complex and the remainder excluding the reference QRS complex, And a feature vector setting and fuzzy clustering means for classifying the shape of the QRS complex into a decision hold type group in which the difference degree decision is held, and a QRS complex And a decision hold type classifying means for classifying the form of the QRS complex belonging to the decision hold type group according to the criterion of the rex type classification.

Here, the feature vector setting and fuzzy clustering means may include a first membership function that reflects the degree of similarity between the remaining QRS complexes excluding the reference QRS complex and the reference QRS complex using a predetermined cluster center value in the feature vector space Classifying a QRS complex having a membership function value included in a prescribed section up and down with a center of 0.5 into a group of undetermined types, classifying a QRS complex having a first membership function value of the predetermined section or more into the similar type group, A first type classification unit for classifying a QRS complex having a first membership function value equal to or less than a predetermined interval into the different morphological types; and, if there is more than one QRS complex classified into the uncertain morphology group, The QRS complex classified into the microcrystalline morphology group and the similar morphology group Calculating a feature vector between the classified QRS complexes to determine a second membership function reflecting the degree of mutual similarity so that a part of the second membership function values are included in a predetermined section up and down around 0.5, If the second membership function value has a second membership function value or less, the non-definite QRS complex is classified into the decision pending type group, and if all of the second membership function values are less than the predetermined interval, A second type classification unit for classifying the plurality of QRS complexes belonging to the different morphology type group into a plurality of QRS complexes belonging to the different morphology type group, And a third type classification unit for performing a type classification for the morphological type group using a classification unit.

The decision hold type classification unit may further include a decision hold type decision unit for determining whether or not a QRS complex belonging to the decision hold type group exists, and a decision hold type presence decision unit for determining whether the QRS complex belonging to the decision hold type group exists If the membership function is classified into a similar QRS complex having all the membership function values within the interval 0.5 times smaller than the predetermined interval from the maximum value of the membership function in the other lead group, If the QRS complex belonging to the decision hold type group is classified into the similar type group and classified into different QRS complexes having all the membership function values within the interval 0.5 times as large as the predetermined interval from the minimum value of the membership function, Lt; RTI ID = 0.0 > QRS < / RTI & And a membership function value between a maximum value of the membership function and a value that is 0.5 times smaller than the predetermined interval from a minimum value of the membership function to a value 0.5 times larger than the predetermined interval from the minimum value of the membership function, the QRS complex belonging to the decision- And a decision pending classification decision section which does not use this parameter for diagnosis of the electrocardiogram.

The predominant type determining means comprises an occurrence frequency calculator for calculating a frequency of occurrence of a QRS complex belonging to each morphological group classified by using the morphological classification means and an occurrence frequency calculator for calculating a frequency of occurrences of the most frequently occurring QRS complex morphology An occurrence frequency determining unit for determining whether the occurrence frequency (maximum occurrence frequency) of the most frequently occurring QRS complex type is three times or more than the occurrence frequency of the second most frequently occurring QRS complex type; When the maximum occurrence frequency is at least three times higher than the second highest occurrence frequency, and when the maximum occurrence frequency is not at least three times higher than the second highest occurrence frequency, the QRS complex having the second highest occurrence frequency The average RR interval in the form of a QRS complex with the average RR interval of the form and the maximum occurrence frequency A first dominant type determining unit that determines a QRS complex type having the maximum occurrence frequency as a predominant type when the condition of the maximum occurrence frequency is 150 ms (150 milliseconds) or less, If the difference between the average RR interval of the QRS complex type having the second highest occurrence frequency and the average RR interval of the QRS complex type having the maximum occurrence frequency is not less than 150 ms, And a second dominant type determining unit for determining a QRS complex type having a high occurrence frequency as a dominant type.

Hereinafter, the operation of the QRS complex property display apparatus of the electrocardiogram signal according to the present invention will be described.

First, the QRS complex determination means calculates a spatial velocity in a multi-dimensional vector sum using the slope between adjacent samples obtained from a lead of a lead group composed of a plurality of leads, extracts a QRS complex from the electrocardiogram signal using the maximum extremes of the spatial velocity, The bandpass filtering unit performs filtering using a bandpass filter for a QRS complex input from each of the plurality of leads, and the read selector selects the absolute value of the absolute value of the QRS complexes filtered by the bandpass filter, When the lead that satisfies the condition that does not include the second extremum of the same sign having the largest pole value (maximum pole value) and which is greater than the maximum pole value of 50% is present, If no extremes are present, the QRS complexes in each single lead Of the maximum peak to peak and the second largest in the ratio (ie, ) Is the maximum is selected.

Wherein the reference point determining unit determines a maximum extreme value point of the lead as a reference point, and the amplitude set determining unit determines a position corresponding to 25% of the maximum extreme value of each QRS complex as a start point and an end point, A sample is selected to determine the amplitude set.

In this case, the feature vector setting and fuzzy clustering means calculates a difference component between 1, which is the extreme value of the correlation value, and the correlation value for the QRS complex pair, and obtains a first vector sum defined by a vector sum of product And a power ratio for a pair of QRS complexes is obtained, and a second vector sum defined by a vector sum of a product of the relative maximum amplitude value and a relative maximum amplitude value is set, and the first vector sum and the second vector A first membership function that reflects the degree of similarity between the remaining QRS complexes excluding the reference QRS complex and the reference QRS complex is determined using a predetermined cluster center value in a feature vector space having a sum as an axis, Classifying a QRS complex having a membership function value included in the interval into a group of undetermined forms, and assigning a QRS complex having a first membership function value equal to or greater than the predetermined interval Classifying the QRS complex having the first membership function value of the predetermined section or less into the different morphological types, and if there is more than one QRS complex classified into the uncertain morphology group, Calculating a feature vector between the QRS complexes classified into the microcrystalline morphology group and the QRS complexes grouped in the similar morphology group on the feature vector space to determine a second membership function reflecting the degree of mutual similarity, The QRS complexes are classified into the predetermined reserved form group if a part of the predetermined QRS complexes are included in the upper and lower predetermined intervals around 0.5 and some have a second membership function value of more than or equal to the predetermined interval, Classifying the preexisting QRS complex into the decision pending type group if the membership function value is present, Membership function value of all has the above predetermined interval value, and classifying said undecided QRS complex in the different form the group, if the QRS complex, the phase belonging to form groups exist more than one, QRS the phase belonging to form the group Wherein the QRS complex belonging to the decision hold type group is judged to be a QRS complex belonging to the decision hold type group by using the first type classification unit and the second type classification unit among the complexes, The membership function of the other lead group is read and the membership function value of the other lead group is used to set all the membership function values within the interval of 0.5 times smaller than the predetermined interval from the maximum value of the membership function in the other lead group , It is classified as a similar QRS complex. Classifies the QRS complex into the similar type group, classifies the QRS complex belonging to the decision hold type group into different QRS complexes while having all the membership function values within the interval 0.5 times larger than the minimum value of the membership function, If there is a membership function value between a maximum value of an accelerator membership function that is 0.5 times larger than the predetermined interval and a value that is 0.5 times smaller than the predetermined interval from a minimum value of the membership function, Do not use the QRS complex that belongs to the ECG diagnosis.

Wherein the predominant type determining means calculates the frequency of occurrence of the QRS complex belonging to each morphologic group classified by using the morphological classification means and calculates the occurrence frequency of the QRS complex type that occurs most frequently using the occurrence frequency determiner ) Is greater than or equal to three times the occurrence frequency of the second most frequently occurring QRS complex type, and the condition that the maximum occurrence frequency is three times or more of the second high occurrence frequency and the condition that the maximum occurrence frequency The difference between the average RR interval of the QRS complex type having the second highest occurrence frequency and the average RR interval of the QRS complex type having the maximum occurrence frequency is 150 milliseconds, Or less, the maximum occurrence frequency is calculated as Determines a QRS complex type as a predominant type, and when the maximum occurrence frequency does not exceed three times the second highest occurrence frequency, the average RR interval of the QRS complex type having the second highest occurrence frequency and the maximum RR interval If the difference in the average RR interval in the form of a QRS complex with a frequency of occurrence is not less than 150 ms, the QRS complex type with the second highest frequency is determined as the dominant type.

As described above in detail, according to the apparatus and method for displaying the QRS complex characteristics of electrocardiogram signals according to the present invention, multiple leads that group three leads into one group are used and the number of samples of processed data is reduced through downsampling , It is possible to perform real-time processing by improving the detection speed by reducing the amount of calculation for the QRS complex detection. In addition, by setting a proper reference point through the QRS complex adjustment in the preprocessing process for performing the QRS complex property display process, As the set is set, the accuracy of the overall characteristic display is improved, and the processing speed can be increased by greatly reducing the number of classification nodes.

In addition, by applying the fuzzy clustering method, we could obtain accurate characteristic display results by improving the classification error even in the poor QRS complex type classification environment which contains a lot of noise components. This result shows that accurate and efficient rhythm analysis It is also expected that it will be possible to extract accurate diagnostic parameters through representative bits.

And, by determining the precise predominant type for the QRS complex in ventricular two-end event, etc., the problem of simply determining the QRS complex with a higher occurrence frequency as the dominant QRS complex can be solved.

Claims (23)

A diagnostic method using an electrocardiogram, the method comprising: calculating a spatial velocity in a multidimensional vector sum using a slope between adjacent samples obtained from a lead of a lead group composed of a plurality of leads, and extracting a queue from the electrocardiogram signal using the maximum extremes of the spatial velocity; Determining a reference point by using a difference and a ratio between a maximum extreme value and a second largest extreme value among the queue and the algorithm complexes, and determining a QRS complex, A reference point and an amplitude set determination step of determining a set of amplitudes by selecting certain samples so that the positions of the start point and the end point of the queue are symmetric with respect to a reference point, a correlation value for the amplitude sets and a relative maximum amplitude Lt; RTI ID = 0.0 > a < / RTI > second vector sum reflecting the power ratio and relative maximum amplitude value for the amplitude sets, A classifying step of classifying the form of a queue and an alias complex by performing clustering in a feature vector space constituted by a sum of a plurality of types of classes, And the average frequency of occurrence of the second and the most frequently occurring queues and the type of ARS complex. Characterized in that it comprises a predominant type determining step of determining a predominant type for a cue-als complex by comparing intervals (RR intervals). 2. The method of claim 1, wherein the reference point and amplitude set determination step comprises: a band-pass filtering step of performing filtering using a band-pass filter with respect to the input queue IE complexes; A plurality of leads satisfying the condition that the absolute value of the extreme value includes the extreme value having the greatest maximum value (maximum extreme value), and at the same time, not including the second extreme value of the same sign larger than the predetermined ratio with respect to the maximum extreme value, If there is no extremum satisfying the condition, the ratio of the maximum extremum and the second extremum among the Q. Alcoh complexes within each single lead A maximum point of the lead is determined as a reference point; a step of determining a maximum point of the lead as a reference point; and a step of selecting a lead corresponding to 25% of the maximum extremum of each queue, And determining a set of sample stones to be symmetrical with respect to the reference point with respect to a start point and an end point of the electrocardiogram signal. The method of claim 1 or 2, wherein the predetermined sample is one sample for one queue, and the sample is 10 samples for an AL complex. 3. The method of claim 2, wherein the predetermined ratio is 50%. 2. The method according to claim 1, wherein the type classification step comprises the steps of: obtaining a difference component between a peak value of a correlation value, 1, and a correlation value for a queue and an alias complex pair, A second vector sum defined by a vector sum of a product of a product of a vector sum and a power ratio and a relative maximum amplitude value is obtained, A reference queue by performing a fuzzy clustering in a feature vector space in which the first vector sum and the second vector sum are axes, and the other queues except for the reference complex queue and the reference complex queue. A similar type group having the same form, a different type group having a different form, a similar degree to the form, and a decision holding type group in which the degree of difference is reserved for the reference queue and the ALS complex according to the degree. A feature vector setting and fuzzy clustering step for classifying the shape of the rex, and a queue of another lead group, a queue belonging to the decision hold type group in accordance with the ALS complex type classification standard, a decision holding classification of the form of the ALS complex A categorization step of categorizing the electrocardiogram signal, and a shape classification step. 6. The method of claim 5, wherein the feature vector setting and the fuzzy clustering are performed using a predetermined cluster center value in the feature vector space, and the remaining queues except for the reference complex queue and the reference complex queue. A membership function that reflects the degree of similarity between the ALS complexes is determined and a queue having a membership function value included in a predetermined section up and down around 0.5 is classified into a group of unspecified types A queue having a first membership function value equal to or greater than the predetermined score, and a queue having a first membership function value less than or equal to the predetermined interval, , And if there is more than one cue .alpha. Complexes classified into the uncertainty type group, A feature vector is calculated between the Q.RLS complex classified into the non-deterministic form group and the Q.RLS complexes pre-classified into the similar form group on the basis of the similarity type, and a second membership function reflecting the mutual similarity is determined If a part of the second membership function values are included in a predetermined upper and lower section with a center of 0.5 and a part has a second membership function value above or below the predetermined section, And if the second membership function values all have a value equal to or greater than the predetermined interval, classify the undetermined queue ARCS complex into the similar type group, and if all of the second membership function values are A second type classification step of classifying the undecided queue ARCS complex into the different types of groups if the value is less than or equal to the predetermined interval; After performing the classification step, if there are two or more queues belonging to the different morphology group, if there are two or more queues belonging to the morphological group, the first morphological classification step and the second morphological classification step And a third type classification step of repeating the type classification step. A method of displaying a characteristic of an ALS complex. The method according to claim 6, wherein the center value of the cluster type center value is the center value of the similarity type group, the first vector sum is 0.20087, the second vector sum is 0.201, 0.9610, and the second vector sum is 0.9631. 7. The method of claim 6, wherein the first type classification step comprises a first reference setting step of setting the reference queue to the reference complex, A first feature vector calculating step of calculating a feature vector for the first reference queue and the reference complex vector using the cluster center value in the feature vector space constituted by the first feature vector; A first membership function determination step of determining a first membership function reflecting the degree of similarity between the queue and the first and second queues excluding the first and second queues and the first and second queues, Determining whether the first membership function value is 0.6 or more, 0.4 to 0.6, or 0.4 or less; and determining, as a result of the first membership function determination step, Classifying the cf. ARS complex into the similarity type group and classifying the cue.alks complex having the first membership function value between 0.4 and 0.6 into the uncertainty type group, And a first classification determination step of classifying the ARS complex into the different types of groups. 7. The method according to claim 6, wherein the second type classification step comprises: a determination step of determining whether or not there is a queue / AR complex belonging to the uncertainty type group; If there is a queue belonging to the indeterminate form group, the first reference queue, the Al-S-complex, is changed to the undetermined queue and the Al-S-complex, and the second reference queue is changed to the Al- A second feature vector calculation step of calculating a feature vector between the Q.R.A complexes pre-classified into the similar type group and the second reference queue.RES complex, Wherein the feature vector space is classified into the similar type group using the cluster center value in the feature vector space in the feature vector space, and the second reference queue, A second membership function determination step of determining a second membership function reflecting the degree of similarity among the complexes, and a second membership function determination step of determining whether the second membership value is 0.6 or more, 0.4 or less, A second membership function determination step of determining whether the second membership function value is present in a range of 0.4 or less and 0.6 or more; and a queue having a second membership function value of 0.6 or more as a result of the determination of the second membership function, And a second membership function value having a second membership function value of less than or equal to 0.4, some of which are present between 0.4 and 0.6 and some of which are less than 0.4 or 0.6 And a second classification determination step of classifying a queue having the second membership function value existing in the classification category into a classification pending type. A method of displaying the characteristics of an ARCS complex. 6. The method of claim 5, wherein the decision hold type classification step comprises: a decision hold type presence / absence judgment step of judging whether or not there is a Q. ALS complex belonging to the decision hold type form; A readout step of reading the membership function of another lead group if the queues.RES complex exists in the decision hold type group, and a step of reading out the membership function of the other lead group in a range of 0.5 times smaller than the predetermined interval from the maximum value of the membership function in the other lead group Wherein the membership function grouping unit classifies the queues belonging to the decision hold type group into the similar type group and classifies the queue type into the similar type group from the minimum value of the membership function, If all membership functions have values within a 0.5-fold interval and are categorized into different queues, Wherein the membership function is classified into the above-mentioned type group, and from a minimum value of the membership function to a value 0.5 times larger than the predetermined interval, from a maximum value of the membership function to a value that is 0.5 times smaller than the predetermined interval A decision hold classification decision step 233 for not using the queue belonging to the decision hold type group and having the membership function value in the ECG diagnosis step 233 and the queue belonging to the different type group ALS complex And if so, repeating the third type classification step. ≪ RTI ID = 0.0 > 11. < / RTI > 11. The method of claim 6 or claim 10, wherein the predetermined interval is 0.2. 2. The method according to claim 1, wherein the dominant type determination step comprises an occurrence frequency calculation step of calculating an occurrence frequency of a queue ARS complex belonging to each type group classified in the type classification step, An occurrence frequency determining step of determining whether the frequency of occurrence of the S-complex form (the maximum occurrence frequency) is three times or more the frequency of occurrence of the queue, A first dominant type determining step of determining, as a dominant type, a queue having a maximum occurrence frequency if the maximum occurrence frequency is three times or more the second highest occurrence frequency; and (RR interval) in the form of a queue with the second highest occurrence frequency, and the maximum occurrence frequency (150 milliseconds) in the average RR interval (RR interval) in the form of a queue, the RR interval is less than 150 milliseconds, the step of going to the first dominant type determination step and, if not, the second highest occurrence frequency And determining a second dominant type for determining a dominant type of a cue and an escape complex type as a dominant type. A diagnostic method using an electrocardiogram, the method comprising: calculating a spatial velocity in a multidimensional vector sum using a slope between adjacent samples obtained from a lead of a lead group composed of a plurality of leads, and extracting a queue from the electrocardiogram signal using the maximum extremes of the spatial velocity; Determining a reference point by using a difference and a ratio between a maximum extreme value and a second largest extreme value among the queue and the algorithm complexes, and determining a QRS complex, A reference point and an amplitude set determination step of determining a set of amplitudes by selecting a certain sample so that the positions of the start point and the end point of the queue are symmetrical about a reference point, a maximum value and a minimum value of the amplitude relative to the amplitude sets , A dominant lead selecting step of calculating a lead including the largest value by defining a sum of the dominant lead selecting unit The relative amplitude to amplitude ratio, and the maximum and minimum values of the relative amplitudes of the selected leads to the electrocardiogram signal, the power ratio, the maximum amplitude to amplitude ratio, the relative amplitude to amplitude ratio, A form classification step of sorting the form of the ALS complex, and a queue having the highest occurrence frequency among the Q. ALS complex forms classified in the type classification step and the second occurrence frequency of the ALS complex form And a dominant type determining step of determining the dominant type for the queue ARS complex by comparing the average average RR interval (RR interval) according to the occurrence frequency between the high queue and the RR complex forms A cue of an electrocardiogram signal. A diagnostic apparatus using an electrocardiogram, the apparatus comprising: a spatial velocity calculating unit for calculating a spatial velocity using a slope between adjacent samples obtained from a lead of a lead group composed of a plurality of leads; and a queue from the electrocardiogram signal using the maximum extremes of the spatial velocity. Determining a reference point by using a difference between a maximum pole value and a second largest pole value among the queue and algorithm complex determining means and the queue and reference complexes to determine an ALS complex, A reference point and amplitude set determination means for determining a set of amplitudes by selecting a certain sample to be symmetric with respect to a start point and an end point of the reference symbol and a reference value and a maximum amplitude value relative to the correlation value for the amplitude sets; By a second vector sum reflecting the power ratio and relative maximum amplitude value for the amplitude sets A form classification means for classifying the form of a queue, an alleys complex by performing clustering in the feature space, and a queue having the highest occurrence frequency among the queue types classified by the type classification means. The average frequency of occurrence of the S-complex form and the second-most frequent occurrence of the Q-type complex form, And a dominant type determining means for determining a predominant type for the queues and the ALS complex by comparing the intervals (RR intervals) of the ECG signals. 15. The apparatus of claim 14, wherein the reference point and amplitude set determination means comprises: a band-pass filtering unit that performs filtering using a band-pass filter for the input queue IEs complexes; A plurality of leads satisfying the condition that the absolute value of the extreme value includes the extreme value having the greatest maximum value (maximum extreme value), and at the same time, not including the second extreme value of the same sign larger than the predetermined ratio with respect to the maximum extreme value, If there is no extremum satisfying the condition, the ratio of the maximum extremum and the second extremum among the Q. Alcoh complexes within each single lead Of the maximum extremum of each of the queues .alpha. Complexes corresponding to 25% of the maximum extremum of each queue. And an amplitude set determiner for determining an amplitude set by selecting a certain sample to be symmetrical about the reference point as a start point and an end point of the electrocardiogram signal. 16. The method of claim 15, wherein the predetermined ratio is 50%. 16. The method of claim 14 and 15, wherein the predetermined sample is one sample for one queue, 10 samples for an ALS complex. 15. The apparatus according to claim 14, wherein the shape classification means calculates a difference component between a peak value of a correlation value of 1 and a correlation value of a cue and an alias complex pair, and calculates a difference component by a vector sum of a product of the relative maximum amplitude value and a relative maximum amplitude value. A second vector sum defined by a vector sum of a product of a product of a vector sum and a power ratio and a relative maximum amplitude value is obtained, A reference queue by performing a fuzzy clustering in a feature vector space in which the first vector sum and the second vector sum are axes, and the other queues except for the reference complex queue and the reference complex queue. , The similarity type with the same shape for the reference queue, the ALS complex, the image type with different form, the similarity degree with respect to the form, and the decision holding type group in which the difference degree determination is suspended. A feature vector setting and fuzzy clustering means for classifying the shape of the flex, and a queue of other lead groups, a queue belonging to the decision hold type group in accordance with the criteria of the ALS complex type classification, a decision And a pending type classification means. The method of claim 1, 19. The method of claim 18, wherein the feature vector setting and fuzzy clustering means uses a predetermined cluster center value in the feature vector space to remove the reference queue, the ALS complex, and the remaining queue and reference complexes. A first membership function that reflects the similarity degree between the ALS complexes is determined and the queue having the membership function values included in the upper and lower predetermined sections around 0.5 is classified into the unfixed form group, A queue having the first membership function value and a queue having the first membership function value of the predetermined interval or more classified into the similar type group and the queue having the first membership function value of the predetermined interval or more, If there is more than one cue .alpha. Complex that is classified into the uncorrected morphological group and the feature vector space, A feature vector is calculated between a queue and an alias complex classified as a positive definite group, and a queue and an alias complex classified as the similar type group, and a second membership function reflecting mutual similarity is determined, 2 membership function values are included in predetermined upper and lower intervals around 0.5 and some have a second membership function value that is greater than or equal to the predetermined interval, Classifying the undetermined queue ACLS into the similar type group if all of the second membership function values have a value equal to or greater than the predetermined interval, A second type classification unit for classifying the undecided queue ARCS complex into the different types of groups if the value is less than or equal to the interval, and a queue belonging to the different type group. If there are two or more S-complexes, the morphological classification of the morphological group is performed using the first morphological classification unit and the second morphological classification unit among queues and AR complexes belonging to the morphological group And a third type classification unit. 20. The method of claim 19, wherein the center value of the similarity group is 0.20087 of the first vector sum, the second vector sum is 0.201, and the center value of the different group is 0.9610 And the second vector sum is 0.9631. A method of displaying an AR complex property, comprising: 19. The apparatus of claim 18, wherein the decision hold type classification unit comprises: a decision hold type presence / absence determination unit for determining whether or not there is a Q.RES complex belonging to the decision hold type group; and a decision hold type presence / A read unit for reading a membership function of another lead group if it is judged that a queue belonging to the group belongs to the same group; And a similar queue type belonging to the decision hold type group and having a function value and being classified into a similar queue and an ALS complex, If all the membership function values are categorized into different queues and ARs complexes within the interval, And the membership function is classified into the above-mentioned type group, and from the minimum value of the membership function to a value that is 0.5 times larger than the predetermined interval, from a maximum value of the membership function to a value that is 0.5 times smaller than the predetermined interval And a decision hold classification decision unit which does not use a queue belonging to the decision hold type group if the membership function value has a membership function value, and a decision hold classification decision unit which does not use an ALS complex for diagnosis of an electrocardiogram. The method of claim 19 or claim 21, wherein the predetermined interval is 0.2. 15. The method according to claim 14, wherein the dominant type determining means comprises an occurrence frequency estimator for calculating a frequency of occurrences of a queue / AR complex belonging to each type group classified using the type classification means, An occurrence frequency determining unit for determining whether or not the frequency of occurrence of the ARS complex form is three times or more than the frequency of occurrence of the ARS complex form, the second occurrence of which is the second largest occurrence frequency (maximum occurrence frequency) Wherein the second highest occurrence frequency is three times higher than the second highest occurrence frequency and the second highest occurrence frequency when the maximum occurrence frequency is not more than three times the second highest occurrence frequency. (RR interval) and the maximum occurrence frequency, the difference between the average RR interval (RR interval) in the form of an RR complex is 150 ms (1 RR interval) A first dominant type determining unit for determining a queuing type ARCS complex type having the maximum occurrence frequency as a predominant type when the maximum occurrence frequency is equal to or less than the second highest type frequency, (RR interval) in the form of an ALS complex and a queue having the maximum occurrence frequency in the form of an ALS complex form And a second dominant type determining unit for determining a dominant type of a queue having a second highest occurrence frequency as the dominant type if the difference between the average RR interval (RR interval) is not less than 150 ms. A method of displaying the characteristics of an ales complex. ※ Note: It is disclosed by the contents of the first application.