JP2017192607A - Automatic electrocardiogram analysis apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic electrocardiogram analysis apparatus capable of accommodating to a change in a state such as body motion and posture of a subject by easily measuring an electrocardiographic signal, while avoiding noise and drift, discriminating presence/absence of abnormality and feature for each pulsation, and providing the subject himself/herself or a clinician with information.SOLUTION: The automatic electrocardiogram analysis apparatus performs stability analysis of respective pulsations of an electrocardiographic signal, discriminates and deletes recording failure parts, detects a primary spike point, performs early stage analysis, performs division point recognition and spike wave analysis, analyzes an interval of respective waves and width thereof, forms a standard waveform and analyses a type of QRS composite waveform. In the primary spike point detection, the primary spike point is determined within a section that exceeds a threshold created in a specific regulation, in a movement standard deviation graph calculated within a range of a calculation window width.SELECTED DRAWING: Figure 12

Description

 The present invention relates to an electrocardiogram automatic analyzer for obtaining information useful for diagnosis of heart disease.

Conventionally, an electrocardiogram has been widely used as a diagnostic index for heart disease. The electrocardiogram is an electrical signal waveform obtained by detecting the heart activity on the body surface, and various information related to the heart activity can be obtained by analyzing the electrocardiogram.

A 12-lead electrocardiograph using standard 10 electrodes is often used for the measurement of the electrocardiogram, but the installation is complicated due to the large number of electrodes, and it is necessary to make the subject rest and lay down. Since the apparatus may be large, the subject cannot easily measure. On the other hand, in recent years, a small-sized device with one channel that can easily measure an electrocardiogram at home by a patient has been provided. (See Patent Document 2).

JP-A-5-3862 Japanese Patent Laid-Open No. 10-234689

However, conventionally, an electrocardiogram automatic analyzer that can automatically and in detail evaluate an electrocardiogram simply measured using a portable one-channel electrocardiograph has not been provided. In particular, when the patient's condition is changing, such as after exercise, the waveform determination method using a uniform threshold has a problem in analysis accuracy.

An object of the present invention is to automatically analyze the characteristics of an electrocardiogram waveform and to be able to obtain various information regarding heart activity with high accuracy by the subject himself or her doctor even when the subject's condition is different. Is to provide a device.

The present invention
It consists of an electrocardiograph, a personal computer, and wired signal transmission connecting between them.
An electrocardiograph continuously measures an analog ECG signal for a certain amount of time,
The preprocessed analog ECG signal is converted into a digital signal and recorded as an ECG signal S1 in a storage device built in the personal computer,
The ECG signal S1 is subjected to a “stability analysis” process to determine whether a significant feature can be obtained by subsequent signal processing;
The signal that can be determined is the electrocardiogram signal S2,
With respect to the electrocardiogram signal S2, it is determined whether or not a significant feature can be obtained by signal processing for each beat interval by the “determination and deletion of defective recording portion” process. Mark not to process
For all beats without this mark,
By detecting the main spine point having a spike-like peak for each beat by the “main spine point detection” process, the signal after the process is set as an electrocardiogram signal S3,
With respect to the electrocardiogram signal S3, a beat having a rhythm faster than a normal average beat is detected by an “early analysis” process,
At the same time, for the electrocardiogram signal S3,
Through the “divided point recognition / spinal wave analysis” process, the P wave, QRS synthesized wave, and T wave segment points are recognized, the respective boundary points are clarified, and the signal after the completion of the process is designated as an electrocardiogram signal S4. age,
In the electrocardiogram signal S4, a dominant waveform of one beat that becomes a standard is created by the “standard waveform creation” process,
With respect to the electrocardiogram signal S4, the QRS composite wave type is determined for each pulsation signal by the “QRS composite wave type detection” process, and the type name of the QRS composite wave is determined for each pulsation of the electrocardiogram signal. The signal that has been processed is referred to as an electrocardiogram signal S5.
Processing for the electrocardiogram signals S1, 2, 3, 4, 5 is performed in the order described by the electrocardiogram automatic analysis software installed in the personal computer,
A display device of a personal computer by combining one or more waveforms of one or more of the electrocardiogram signals S1, 2, 3, 4, and 5 with a code indicating a characteristic obtained by the signal processing Displayed by
Furthermore, the electrocardiogram automatic analysis apparatus is characterized in that information relating to an electrocardiogram signal can be transmitted to another computer / storage device through a network.

Electrocardiogram signals can be measured with a small electrocardiograph and analyzed and shown with high accuracy by simple operation of an electrocardiogram automatic analyzer consisting of a combination of a personal computer or portable information device and a calculation processing server and software installed therein. Therefore, heart disease can be detected quickly and easily, and the health of the subject can be protected.

ECG automatic analysis flowchart Stability analysis flowchart Determination of defective recording portion and deletion processing flowchart Main spine point detection processing flowchart Earlyness analysis processing flowchart Classification point recognition and spike wave analysis processing flowchart Analysis processing flowchart of the start point and end point of each wave Analysis flow chart of the interval and width between each wave Standard waveform creation process flowchart Table of rules for detecting QRS composite wave types Diagram of QRS composite waveform pattern Configuration of ECG automatic analyzer Figure of typical waveform of electrocardiogram signal in 1 beat beat section Conceptual diagram of detection of defective recording of spike noise Spike noise detection example Conceptual diagram of detection of defective recording due to drift Conceptual diagram of moving standard deviation calculation Conceptual diagram of main spine detection Conceptual diagram of threshold determination for main spine detection Conceptual diagram of early sex determination Conceptual diagram of beat section determination Conceptual diagram of detection of start and end points of QRS composite wave Conceptual diagram of P wave start and end point detection Conceptual diagram of T wave start and end point detection Illustration of the interval between each wave Illustration 1 of the main spine point interval Illustration 2 of the main spine point interval Table of rules that subdivide QRS synthetic wave types for positive main spikes Table of rules that subdivide QRS synthetic wave types for negative main spike waves Conceptual diagram for creating a standard waveform Illustration before editing the main spine Illustration after editing the main spine Arrhythmia annotation determination flowchart Flowchart 2 for determining arrhythmia annotation Illustration of editing arrhythmia annotation Conceptual diagram of enlarged display of part of ECG signal Illustration of color display according to the characteristics of ECG signal Explanatory drawing to change the order of editing process 2 of the configuration diagram of an electrocardiogram automatic analyzer Sketch drawing showing the state of using an electrocardiograph Diagram showing data structure Temporary main spine point, illustration of detection of main spine point

In the following description, for convenience, “up” or “down” may refer to a plus or minus direction with respect to the base line on the measurement voltage axis, and “before” or “after” may refer to a minus or plus direction on the time axis.
An electrocardiogram signal, which is originally a discrete digital signal, is drawn as a waveform with a continuous line for convenience, with a set of points having coordinates (time, voltage) indicated by positions on the time axis and voltage axis.
The electrocardiogram is obtained by displaying the waveform of an electrocardiogram signal on a display device.

FIG. 12 shows the configuration of a portable 1-channel electrocardiogram automatic analyzer. As shown on the lower left side of FIG. 12, the electrocardiogram automatic analyzer comprises an electrocardiograph 11, a personal computer 12, and a wired signal transmission 15 connecting them. The personal computer 12 is equipped with an electrocardiogram automatic analysis software and performs various processes on the electrocardiogram signal.

The display device of the personal computer 12 simultaneously displays an electrocardiogram signal and a code or mark representing the waveform characteristics of the electrocardiogram signal obtained by analysis. FIG. 39 shows a case where electrocardiographic signal information is transferred from the electrocardiograph 11 to the personal computer 12 using an SD memory card or USB memory instead of the signal transmission device.

The electrocardiogram automatic analysis software installed in the personal computer 12 analyzes the characteristics of the electrocardiogram signal. This software can detect ECGs of any lead and does not select the physique or constitution of the subject. The software can stably analyze ECG waveforms with only one lead (one channel) input. The subject can also be used at home. It has. Although the measurement time of the electrocardiogram signal can be 30 seconds to 24 hours or more, in this embodiment, a case where the measurement time for one measurement is 30 minutes will be described.

As shown in FIG. 12, the electrocardiograph 11 includes at least an electrode 110, a preamplifier 111, a filter 112, an A / D converter 113, a CPU 114, a memory 115, an interface 116, a communication unit 117, an operation switch 118, a power supply 119, and a display lamp. 120, which function as a signal measuring device. The electrocardiograph 11 is connected to the personal computer 12 by a LAN that is a wired signal transmission 15. The personal computer 12 can be connected to another computer 13 or the like via a network.

As shown in FIG. 40 (a), two electrodes 110 are arranged at both ends of a case 121 made of an insulator of a portable one-channel electrocardiograph 11, respectively. When the electrocardiograph 11 is used, the case 121 of the electrocardiograph 11 is held with both hands so that the palms of the right hand 4 and the left hand 5 are in contact with the two electrodes 110 as shown in FIG.

The voltage difference between the two electrodes is measured, and the measured analog electrocardiogram signal is amplified by the preamplifier 111. As a pre-processing, a direct current component, a low frequency drift of 1 Hz or less, a harmful commercial power source or a DC- An induction signal from a DC power source or the like or a myoelectric signal that becomes noise of, for example, 1 kHz or more is removed, and is converted into a digital signal by the A / D converter 113 to be an electrocardiogram signal S1.

The electrocardiogram signal S1 is once recorded in the memory 115, which is a transmission buffer, and immediately transmitted to the personal computer 12 via the LAN by the communication unit 117. The data structure shown in FIG. 41 is recorded in a storage device built in the personal computer 12. In FIG. 41 showing the data structure, the codes corresponding to the fields are attached only to the first row in order to avoid complexity, but the fields in the same column are the same codes.

Information recorded in the memory 115 and the storage device built in the personal computer 12 has a predetermined data structure, and voltage information at one measurement point is arranged as continuous information corresponding to one unit of recorded data. One unit of recording data shown in FIG. 41 includes a plurality of fields, field a61 includes a number indicating the order, field b62 includes a signal value, and fields c63 to field n66 include a plurality of labels.

Since the electrocardiogram signal S1 is obtained by digitizing an analog electrocardiogram signal at a constant sample rate, the number indicating the order of the recorded data recorded in the field a61 indicates a time at a constant interval. In the field b62, the voltage value of the signal at that time is recorded. In fields after the field c63, labels indicating various processing results are recorded in the determined fields.

The memory 115 of the electrocardiograph 11 is a buffer for transmitting measurement data to the personal computer 12, and its storage capacity is 64 ms of measurement. In the specification as an electrocardiogram automatic analyzer, the measurement time is 30 seconds to 24 hours or more. The speed of the sampling circuit corresponds to 125 points / second to 1000 points / second. In the specification as an automatic electrocardiogram analyzer, the sampling period is 125 points / second for the service using the cloud and 256 points / second for the service of the event recorder / holter.

In the following explanation example, the sampling period is 125 points / second. The digitized electrocardiogram signal S1 is temporarily recorded in the memory 115 of the electrocardiograph 11 and is stored in the storage device built in the personal computer 12 via the communication unit 117 for 30 minutes which is one measurement. The length is recorded. Alternatively, a signal of 30 seconds to 24 hours or more is recorded as necessary. The CPU 114 controls the operation of the electrocardiograph 11.

The process according to the electrocardiogram automatic analysis flowchart shown in FIG. 1 is performed by the electrocardiogram automatic analysis software installed in the personal computer 12. The personal computer 12 automatically analyzes the waveform of the electrocardiogram signal S1 collected by the electrocardiograph 11 in accordance with an operation command input from an input device equipped with a keyboard or the like, and obtains information useful for diagnosis. The result is displayed on the display device. In accordance with the input operation command, the personal computer 12 displays the electrocardiogram signal S1 before processing and a signal obtained by processing it on the display device. In that case, all the information for one measurement may be displayed in a lump, or only a part may be enlarged and displayed.

FIG. 13 shows a typical waveform diagram of an electrocardiogram signal in one beat section. FIG. 13A shows a case where the main spike wave is positive, and FIG. 13B shows a case where the main spike wave is negative. Positive main spine wave means that the main spine point is on the plus side above the baseline, and negative main spine wave means that the main spine point is on the minus side below the baseline. The names shown in this figure are used for the names of the electrocardiographic signal waveforms used in the following description. The electrocardiogram automatic analysis apparatus using the personal computer 12 according to the present invention will be described in more detail sequentially in accordance with the electrocardiogram analysis process flowchart shown in FIG.

The stability analysis S200 or the recording failure portion determination and deletion S201 is to detect spike noise, drift, saturation, and a stable QRS composite wave, and is a technique unique to the present invention and is not found elsewhere. .

FIG. 2 shows a detailed processing flowchart of the stability analysis S200 shown in FIG. The electrocardiogram signal S1 recorded in the storage device of the personal computer 12 is first subjected to stability analysis S200 by the electrocardiogram automatic analysis software. If it is determined that the signal is stable, it is recorded as an electrocardiogram signal S2. The As shown in FIG. 2, when it is determined that the signal is not stable even in one processing step, an alarm signal is sent from the personal computer 12 to display an alarm on the display lamp 120 of the electrocardiograph 11 as being abnormal. To prompt the user (subject) to redo the measurement.

Here, a stable signal means that (1) the amount of drift is less than a certain value, and (2) the sum of absolute values of measured values does not exceed a predetermined value because of large spike noises. (3) There is no portion within the maximum positive / negative value (for example, ± 4.95 mV) that is the limit range of the amplitude, and there is no portion exceeding the saturation value that is the limit of hardware related to measurement, and finally, (4) FIG. That is, the output of the QRS synthesized wave peak enhancement filter shown in FIG. 13 is equal to or greater than a certain value, and satisfies the four conditions of being continuously and substantially constant.

The personal computer 12 performs a drift test S301, a spike noise test S302, a saturation test S303, and a QRS synthesized wave test S304 for checking these four conditions according to the flowchart shown in FIG. 2 using the equipped electrocardiogram automatic analysis software.

Next, “determination and deletion of defective recording portion” S201 is performed on the electrocardiogram signal S2 recorded in the storage device built in the personal computer 12. The recording failure portion refers to a portion where spike noise, drift, or saturation is present and the waveform form is not clear. The contents of the processing are performed according to the same rules as those of the stability analysis described above, but further screening is performed by tightening the criterion for acceptance / rejection. This is because a possible signal is widely collected and recorded, and only a signal that can be reliably analyzed is obtained in the “determination and deletion of defective recording portion” S201. Further, when there is a section determined to be defective in this process, only a defective label is attached only to the section, and the entire recording for 30 minutes is cleared and measurement is not performed again.

“Determination and deletion of defective recording portion” S201 will be described in more detail with reference to FIG. First, the drift test S301 shown in FIG. 3 will be described. FIG. 16 shows a conceptual diagram of drift detection. For all measurement points, the standard deviation of the electrocardiogram signal S2 (referred to as moving standard deviation) in a section of a certain length (about 1 beat, for example, 900 ms) starting from the measurement point is obtained. A drift is determined when a certain threshold value (for example, 0.85 mV) is exceeded.

A “defective” label is recorded in a specific field of the measurement data in the section determined to be drift. The measurement data having the “defective” label is not subjected to subsequent analysis processing, and is displayed as a blank when the waveform is displayed on the display device.

FIG. 14 shows a conceptual diagram of spike noise detection. For all measurement points of the electrocardiogram signal S1 for 30 minutes, the absolute value of the difference between all adjacent measurement points in a predetermined interval (analysis window width) of about one to several beats that starts from the measurement point. If the value obtained by taking the sum and dividing the sum by the size of the section is equal to or greater than a predetermined multiple of the value in the case of normal pulsation, the point is determined to be spike noise.

FIG. 15 shows an example of spike noise detection. The square drawn with a broken line is a section detected as spike noise, and it can be seen that the waveform is disturbed unlike the electrocardiogram signal waveform compared with other portions. In a section determined as spike noise, a “defective” label is recorded in a specific field of measurement data. The corresponding section of the electrocardiogram signal S2 having the “defective” label is not subjected to subsequent analysis processing, and is displayed as a blank when the waveform of the electrocardiogram signal S2 is displayed on the display device.

The saturation inspection S303 in FIG. 3 will be described. Saturation is detected by determining saturation when the electrocardiogram digital signal exceeds a certain range (for example, −4.95 mV to +4.95 mV). A “defective” label is recorded in a specific field of measurement data in a section determined to be saturated. The measurement data having the “defective” label is not subjected to subsequent analysis processing, and is displayed as a blank when the waveform is displayed on the display device.

The QSR synthesized wave inspection S304 in FIG. 3 will be described. The test determines that the QSR composite wave is not too small. A moving average length is determined in advance, a moving average of the electrocardiogram signal S2 is calculated, and the result is y (xn). Using y (xn), the QRS synthesized wave peak enhancement filter output of the following equation is calculated.
F (xn) = m1 × [2y (xn) − {y (xn−1) + y (xn + 1)}]
Where m1 is a constant

移動分散Ｖ（ｘｎ）が閾値を超える初めてのピークに続いて、一定時間以内にピークが現れて、Ｖ（ｘｎ）が閾値を越えた場合にＱＳＲ合成波が安定であると判定する。ここで閾値は、計測を開始して一定時間の間に現れたＶ（ｘｎ）の最大値Ｖｍａｘを一定の定数で割った値である。 Further, a moving variance V (xn) that is the square of the moving standard deviation of F (xn) is calculated.

Following the first peak in which the moving variance V (xn) exceeds the threshold, a peak appears within a certain time, and when V (xn) exceeds the threshold, it is determined that the QSR composite wave is stable. Here, the threshold value is a value obtained by dividing the maximum value Vmax of V (xn) appearing during a certain time after the start of measurement by a certain constant.

A “defective” label is recorded in a specific field of measurement data in a section that is not determined to be stable by this inspection. The measurement data having the “defective” label is not subjected to subsequent analysis processing, and is displayed as a blank when the waveform is displayed on the display device. Thus, the “determination and deletion of defective recording portion” process is completed, and a label indicating that “determination and deletion of defective recording portion” is completed is written in a predetermined field of the data structure of the measurement data for 30 minutes.

The “main spine point detection” S202 shown in the flowchart of FIG. 1 will be described. This is the first step in the processing for obtaining the characteristics of the waveform of the electrocardiogram signal in earnest about the electrocardiogram signal S2. “Analysis and deletion of defective recording portion” This analysis is performed on the entire electrocardiographic signal S2 that has not been deleted in S201. The main spine point refers to a measurement point having the largest peak absolute value in a section regarded as one pulsation.

FIG. 4 shows a main spine point detection processing flowchart. The calculation process of “main spine point detection” S202 is shown in three columns in FIG. The main processes are the “moving average” S300, “calculation window (window) condition setting” S305, “section detection / determination” S306, which are in the part surrounded by a broken line written as the whole main spine point detection in the left column. These are “temporary main spine point detection” S307, “main spine point detection” S308, “beat rate inspection” S309, and “pose inspection” S310.

A processing step of the same name and code is also drawn in a portion surrounded by another broken line, which is described as subdivision main spine point detection in the center row described later. The processing content of these steps is the same as each processing step of the whole main spine point detection.

FIG. 17 shows a conceptual diagram such as moving standard deviation calculation for detecting the main spine point. The moving standard deviation calculation means obtaining a standard deviation for a measured value within a certain time range including the relevant point while moving with respect to a measured value along the time axis. The time range at this time is called a calculation window width.

This calculation process for detecting the main spine point is performed on the electrocardiogram signal S2 for 30 minutes, but in FIG. 17, only a waveform of one beat is shown for explanation. First, a moving average of the electrocardiogram signal S2 shown in FIG. 17A using a constant moving length is repeatedly calculated several times to obtain FIG. 17B. The reason for taking the moving average is to improve resistance to noise. However, since various noises are mixed when measuring the electrocardiogram signal S2, the main spine point to be detected and spike noise may be difficult to distinguish in calculation. is there.

I want to improve the noise resistance while maintaining the characteristic waveform form as an ECG. Therefore, a technique is adopted in which the moving average is repeated for the same signal and is calculated a plurality of times to achieve it. Here, the moving length of the moving average was 3 measurement points, and the number of repetitions was 3 times.

In the next step, the number j of measurement points for determining the calculation processing section is set. First, in order to shorten the calculation time, the value of j is set small. If the main spine point is not successfully detected, the value of j is gradually increased as described later. As j increases, it becomes easier to detect the main spine. This is called calculation window (window) condition setting S305.

Determining the value of j corresponds to obtaining the waveform indicated by the solid line in FIG. This is to cut out the electrocardiogram signal S2 after the moving average shown in FIG. That is, with respect to the voltage at the measurement point, the standard deviation σi in the range from the fixed interval i−j to i + j with respect to the index i of xi indicating the position on the time axis is calculated while increasing i by 1 (moving standard deviation). .

The next step is called section detection / determination S306. FIG. 17D shows a graph in which the moving standard deviation σi obtained above is arranged along a time series. The same graph is also shown in FIG. (FIG. 18 (d) follows FIG. 17 (d))

FIG. 19 shows a graph in which the obtained moving standard deviation σi is arranged in the right direction from the smallest value to the largest value. On this graph, from the straight line Ya connecting the point of the median value Hmed of the moving standard deviation σi and the point of the maximum value Hmax, the point where the distance to the graph of the moving half width is maximum is obtained, and the moving standard deviation of that point is obtained. Is the threshold value Hth.

Next, temporary main spine point detection S307 is performed. In FIG. 18D, a range exceeding the threshold value Hth measured from the position corresponding to the base line portion is the temporary main spine point examination section. The temporary main spine point inspection section is a time section in which a main spine point may exist. In FIG. 18 (e), the pulse-shaped examination section indicates the temporary main spine point examination section obtained in this way, in which the main spine point may be detected.

When there is a discontinuous portion as shown in FIG. 18 (e) with respect to the temporary main spine point inspection section in which the main spine point obtained in this way is likely to be detected, a certain rule is used. The gap between the sections where there is a possibility that the adjacent temporary main spine may exist is filled, and the sections where the temporary main spine may exist are combined. As a result, a combined examination section is obtained as shown in FIG. As shown in FIG. 42 (a), one examination section corresponds to one pulsation.

In this way, an examination section in which a temporary main spine is present was obtained. In accordance with the rules shown in the following Table 1, it is determined whether or not a point having a large absolute value in the section is a temporary spine. As a result, as shown in FIG. 42A, temporary main spines are detected. In many cases, two temporary main spines are detected in the same section. When it is determined as a temporary main spine, a label indicating the temporary main spine is recorded in a predetermined field in the data structure.

Therefore, main spine point detection S308, which is the next step, is performed. In this step, two temporary main spines are narrowed down to one by the following method. That is, as shown in FIG. 42B, absolute values v1 and v2 of the difference between the height of the electrocardiogram signal S2 at the head of the section and the height of the two temporary main spines are obtained, and the difference between v1 and v2 is compared. The larger temporary main spine point, the positive peak on the left side in the figure, is taken as the main spine point. In FIG. 42B, only one section is shown and the other sections are omitted, and the main spine mark 31 and the temporary main spine mark 33 are labeled in one section. When the main spine point is determined, a label indicating the main spine point is recorded in a predetermined field in the data.

When the main spine point is detected by the above method, the pulsation rate test S309 shown in FIG. 4 is performed. When the pulsation speed is equal to or higher than a certain value (for example, 40 beats / minute) as an average of all measurement times (that is, in the case of OK in S309), the process proceeds to the pose inspection S310. When the pulsation speed is equal to or lower than a certain value (that is, NG in S309), the process proceeds to subdivided main spine point detection.

In the pose inspection S310, if there is no resting point where the interval between main spine points exceeds a certain value (for example, 3 seconds) in all the measurement times (that is, OK in S310), the label of the temporary main spine point is displayed. The label is corrected, and the main spine point detection S202 ends.

If there is a part where the main spine point interval exceeds a certain value (ie, NG in S310), the detected temporary main spine point is considered to be incompatible, and the value of j in the calculation processing section (analysis window width) condition is Increase the predetermined number and redo the main spine point detection.

When the pulsation speed is equal to or lower than a certain value by the pulsation number test S309 (that is, NG in S309), the subdivided main spine point detection in the center column of the flowchart of FIG. 4 is performed. Each processing step of subdivided main spine point detection is performed in the same procedure as the whole main spine point detection. However, instead of determining a peak with a uniform threshold for the entire measurement time of 30 minutes, measurement is performed. The difference is that the processing is performed by subdividing the time into units of one minute, and the threshold value is set independently in each range.

In some cases, the main spine point to be detected may be missed due to an increase in the average value (that is, the threshold value) of the movement half-value width due to the influence of drift or noise. As a result, since the pulsation speed is determined to be equal to or less than a certain value in the pulsation number test S309, main spine point detection is performed in units of 1 minute in order to save such a case. After that, if there is no pause point where the main spine point interval (for example, 3 seconds) exceeds a certain value in the pose test performed after subdividing the main spine point detection (that is, OK in S310), The label is corrected to the label of the main spine point, and the main spine point detection S202 ends.

If there is a pause portion where the main spine point interval (for example, 3 seconds) exceeds a certain value in the pose test performed after the detection of the subdivided main spine points (that is, NG in S310), the flowchart of FIG. The process proceeds to the missing inspection S311 in the right column. In the missed inspection S311, the inspection is performed by reducing the threshold value so that the peak can be easily found only in the portion having the pause time of 3 seconds or more.

In the subsequent pose inspection S310, if there is no resting point where the interval between the main spine points exceeds a certain value (for example, 3 seconds) (that is, OK in S310), the label of the temporary main spine point is changed to the label of the main spine point. It corrects and ends main spine point detection S202.

In the pause test S310 following the missed test, if there is a rest point where the interval between main spine points exceeds a certain value (for example, 3 seconds) (that is, NG in S310), the value of j is increased by a predetermined number. Redo from the detection of the whole main spine point. Until the pose inspection S310 is passed, the main spine point detection is performed by changing the value of j with three rows of “main spine point detection”, “subdivision inspection”, and “missing inspection” as one set.
The label of the electrocardiogram signal S3 is recorded on the electrocardiogram signal S2 after the detection of the main spine point. This is the end of the description of the main spine point detection S202.

Second, “analysis of earlyness” S203 shown in FIG. 1 will be described. Prematureness means that the electrical excitement that occurs in the sinus node is correctly transmitted to the atrium, atrioventricular node, and ventricle, thereby causing the ECG waveform to appear regularly, which is repeated at a constant rhythm, It is that the heart contracts at an earlier timing (early).

FIG. 5 shows an early analysis processing flowchart. FIG. 20 shows a conceptual diagram of earlyness determination. In the early analysis process, after the detection of the main spine points, the average interval MSIave between the main spine points is obtained over the entire measurement time of 30 minutes, while the MSIgv that is the interval between the continuous main spine points is 1 This is done by determining the ratio of MSIpre, which is the interval between the previous main spines. That is, the following is obtained for all main spines.
K1 = MSIgv / MSIave
K2 = MSIgv / MSIpre

As shown in FIG. 20, the median point and the minimum point of the ratio are connected by a straight line Yb on a graph in which the obtained ratio K1 is drawn as a continuous line in the size order in the right direction on the paper. Then, C1 point at which the distance from the straight line Yb to the ratio graph is maximized is obtained, and the portion on the left side from C1 point on the graph is set as the first condition in the range having earlyness.

Separately, a ratio K2 between the interval MSIgv between two consecutive main spine points and the interval MSIpre between two previous consecutive main spine points is obtained for all the main spine points. The obtained ratios are drawn in a graph as a continuous line arranged in the right direction on the paper in the order of size. Since it is the same figure as FIG. 20, the figure is also used.

As shown in FIG. 20, on the graph in which the obtained ratios are arranged in the right direction on the paper in the order of size, the median point and the minimum value point of the ratio graph are connected by a straight line Yb. C1 point at which the distance from the graph to the ratio graph is maximized is determined, and the portion on the left side from C1 point is set as the second condition in the range having earlyness.

The main spine point where the first condition and the second condition are simultaneously satisfied is determined to have earlyness, and a label indicating the earlyness of the arrhythmia annotation is recorded in the determined field. This is the end of the explanation on the early analysis S203.

Third, the first half of “Division Point Recognition / Spine Wave Analysis” S204 shown in FIG. 1 will be described. The first half of the segment point recognition / spinal wave analysis S204 is to clarify the boundary point Bi for automatic analysis of adjacent pulsation sections. FIG. 6 shows the first half of the flowchart of “Division Point Recognition / Spine Wave Analysis” S204. FIG. 21 shows a conceptual diagram of pulsation section determination. The segment point recognition / spinal wave analysis S204 performs processing for the entire 30-minute electrocardiogram signal, but FIG. 21 shows only four beats.

The ratio between the interval MSIgv between the two consecutive main spine points and the second interval obtained by adding the interval MSIpt between two consecutive main spine points one after the interval is determined as MSIir.
Further, a product B (xi) of two values of the interval MSIgv and the ratio MSIir of the two intervals is obtained, and this is set as the length of time between the start point of the pulsation interval and the main spine point. .
That is,
MSIir = MSIgv / (MSIgv + MSIpt)
B (xi) = MSIgv × MSIir

The starting point of the pulsation section thus obtained is a boundary point with the previous pulsation section. A label indicating the start point of the pulsation is recorded in a predetermined field of the recording area at the start point of the pulsation section.

Fourthly, the second half of “Division Point Recognition / Spine Wave Analysis” S204 shown in FIG. 1 will be described. The interval / width analysis processing between the waves is to obtain a start point and an end point that are division points of the P wave, the QRS composite wave, and the T wave. FIG. 7 shows a flowchart of the start point / end point analysis process of each wave. First, the start point and end point of the QRS composite wave are obtained. FIG. 22 shows a conceptual diagram of detection of the start point and end point of the QRS composite wave.

When the moving average of the electrocardiogram signal S3 shown in FIG. 22A is obtained, FIG. 22B is obtained. Next, j, which is a constant for determining the calculation processing section, is determined. This is called calculation window condition setting S305. This corresponds to cutting off a part of the electrocardiogram signal on the time axis of the electrocardiogram signal waveform, as indicated by a solid line in FIG. The moving standard deviation is the standard deviation in the range from the fixed interval i−j to i + j with respect to the index i of xi indicating the position on the time axis with respect to the voltage at the measurement point over the one beat of the electrocardiogram signal S3. Calculate while increasing i by 1.
FIG. 22D shows a graph in which the obtained moving standard deviation σi is arranged along a time series.

Next, an average value Have (referred to as a first average value) of the obtained moving standard deviation σi is obtained. As indicated by an arrow in the middle of the graph of FIG. 22 (d), with this Have as a threshold value, the moving standard deviation σi in the previous (xi-1) is smaller than Have and continues in front of the main spinal point. The measurement points that satisfy the condition that the moving standard deviation σi in (xi) is larger than Have are searched for sequentially along the positive direction of the time axis (increase i of xi by 1), and the first position that satisfies this is determined as QRS. The starting point of the composite wave.

Similarly, a measurement point that satisfies the condition that the moving standard deviation σi in (xi) is greater than or equal to Have and the immediately following (xi + 1) moving standard deviation σi is smaller than Have is set in the positive direction of the time axis. The first position that satisfies this condition is taken as the end point of the QRS composite wave.

The process described above is similar to the procedure used to determine the main spine point, but the difference is that in the case of the main spine point, one pulsation is undetermined. In the case of the start point and the end point of the composite wave, calculation processing is performed within one pulsation section. Further, when the start point and end point of the QRS composite wave are finally obtained, the average value of the moving standard deviation σi is used as a threshold value. When the start point and end point of the QRS composite wave are detected, a label indicating the start point or end point is recorded in a predetermined field in the recording area of each point.

Next, the start point and end point of the P wave are obtained between the start point of the electrocardiogram signal S3 in the one beat and the start point of the QRS composite wave. FIG. 23 shows a conceptual diagram of detection of the start point and end point of the P wave. The position of the P wave shown in FIG. 23 (a) is obtained in FIG. 23 (d). The moving average of the electrocardiogram signal S3 shown in FIG. 23B is obtained from the electrocardiogram signal S3 in FIG. Next, in front of the main spine point, a line Yp is connected between the start point of one beat of the electrocardiogram signal S3 and the start point of the QRS composite wave. The distance between the moving average of the electrocardiogram signal S3 and the straight line Yp is obtained for each xi and arranged in order, and a graph is drawn in FIG.

As shown in FIG. 23 (c), in this graph, a graph is cut out with a constant calculation processing width, and a movement half-value width is obtained with a constant calculation range width for each xi, and the horizontal axis is shown as shown in FIG. 23 (d). Draw a graph of moving half width as time. Obtaining the movement half-value width means obtaining the half-value width of a measurement value within a certain time range including the corresponding point while moving the measurement value along the time axis.

FIG. 23C illustrates six graphs that change over time. The double-arrow line indicating the calculation range of the movement half-value width is drawn only in the first graph, but it moves to the right as it moves to the lower stage over time. A small double-headed arrow line shown in each graph of FIG. 23C indicates the half-value width. With this half-value width, a graph of the movement half-value width shown in FIG. In the moving half-value width graph, an average value of moving half-value widths indicated by broken lines (referred to as a second average value) is set as a threshold value, and a range in which the moving half-value width exceeds the threshold value is set as a P wave.

At this time, as shown by a broken line graph in FIG. 23 (d), when the threshold value is exceeded and then becomes less than or equal to the threshold value, the threshold value is exceeded again, that is, the interval exceeding the threshold value and the interval not exceeding the threshold value are alternated. In the case where the section exceeding the threshold is not continuous, the section sandwiched between the portions exceeding the threshold is a P wave. When the detection of the start point and the end point of the P wave is completed, a label indicating the start point or the end point is recorded in a predetermined field of the recording area of each point.

Next, the start point and end point of the T wave are obtained between the end point of the QRS composite wave of the electrocardiogram signal S3 and the end point of the one beat. FIG. 24 shows a conceptual diagram of detection of the start point and end point of the T wave. A moving average of the electrocardiogram signal S3 shown in FIG. 24B is obtained from the electrocardiogram signal S3 in FIG. Next, behind the main spine point, the end point of the QRS composite wave and the end point of the one-beat electrocardiogram signal S3 are connected by a straight line Yq.

As shown in FIG. 24C, the distance between the moving average of the electrocardiogram signal S3 and the straight line Yq is obtained in the same manner as in the detection of the start point and the end point of the P wave, and arranged in order and drawn on a graph. In this graph, a moving average graph is cut out with a constant calculation processing width, a moving half width is obtained for each xi, and a moving half width graph is drawn with the horizontal axis as time as shown in FIG.

In this movement half-value width graph, an average value (referred to as a third average value) is used as a threshold value, and a range in which the movement half-value width exceeds this value is designated as a T wave. At this time, as shown by the broken line graph in FIG. 24D, as in the case of detecting the start point and the end point of the P wave, the case where the section exceeding the threshold and the section not exceeding the threshold are not mixed and continuous. The section sandwiched between the excess parts is a T wave. When the detection of the start point and end point of the T wave is completed, a label indicating the start point or end point is recorded in a predetermined field of the recording area of each point.

Here, the main spine point interval will be described. FIG. 25 is an explanatory diagram of the interval between the waves. The interval between the waves refers to the interval between the same waveforms in two consecutive beats. FIG. 25A shows the interval between the P wave start points when the main spike wave is positive, that is, the PP interval, and the interval between the R wave peaks, that is, the RR interval. FIG. 25B shows the interval between the P wave start points in the case of negative main spike waves, that is, the PP interval, and the interval between the S wave peaks, that is, the SS interval.

26 and 27 are explanatory views of the main spine point interval. The interval between two pulsations is preferably determined by the PP interval. However, since it may be difficult to obtain the start point of the P wave, the interval between the main spines that is easy to detect is set as the interval between pulsations. In the present invention, in the electrocardiogram signal waveform, the peak value having the largest fluctuation from the base line in the R wave, QS wave (details will be explained in the section of type detection), S wave, and Q wave of two beats. The interval between the peak points of the spikes is considered as the main spine interval.

FIG. 26 (a) shows a case where there is an R wave peak on the plus side, which is the largest compared with other spike waves, and is the main spike wave, and the RR interval is the main spike interval. ing. FIG. 26B shows a case where the peak of the S wave is on the minus side and is the largest and the main spike wave compared to other spike waves, and the SS interval is the main spike interval. ing.

FIG. 27 (c) (following (a) and (b) in FIG. 26, (c) and (d) are shown). This is the case where the main spike wave is the largest compared to the spike wave, the peak is the S wave on the negative side in the subsequent pulsation, and the largest spike spike wave is compared to the other spike waves, and the RS interval. Indicates the main spine interval.

FIG. 27 (d) shows that the peak of the R wave is on the positive side in the first pulsation, which is the largest compared to the other spikes, and is the main spike, and the peak of the Q wave is on the negative side in subsequent pulsations. This is the case where it is the largest and the main spike wave compared to other spike waves, indicating that the RQ interval is the main spine point interval. This is the end of the description of the segment point recognition / spine wave analysis S204.

Fifth, the “standard waveform creation” S206 shown in FIG. 1 will be described. “Standard waveform creation” refers to creation of a dominant waveform for one beat used to determine whether or not an abnormal waveform is obtained from the waveforms of all measured ECG signals. In the present invention, a standard waveform is created by adding and averaging all ECG signal waveforms in units of one beat excluding a signal with a defective label and a beat that is an arrhythmia. When the waveform is greatly different in an item that is characteristic with respect to the standard waveform, it can be determined that this is an abnormal electrocardiogram signal waveform of the subject.

FIG. 30 (a) shows only 4 beats of the electrocardiographic signal in all beat sections continuously captured within the measurement time of 30 minutes. As shown in FIG. 30 (b), all the measured data are separated in units of one beat section, and each time axis is shifted and overlapped so that the main spines coincide on the waveform graph. Match. Similarly, the respective voltage axes are moved, and the voltages of the respective main spines are matched and overlapped on the waveform graph.

Then, the sum (of the measured values) of all the electrocardiogram signals is obtained, and the average waveform is obtained by dividing by the number of the electrocardiogram signals. FIG. 30 (c) shows the result. The dominant waveform thus obtained is recorded as a standard waveform in a predetermined area in a storage device built in the personal computer 12. This completes the description of the creation of the standard waveform.

Sixth, “type analysis of QRS composite wave” S207 shown in FIG. 1 will be described. The QRS synthetic wave type analysis refers to automatically discriminating a QRS synthetic wave type commonly recognized in the medical community according to a certain rule. Determining the type is very important for understanding heart movement and exchanging information between doctors.

Regarding the expression of the QRS composite wave, basically, large peaks of the Q wave, R wave, and S wave are written in capital letters, and small amplitudes are written in small letters. When two or more waveforms with the same name are recognized in one QRS composite wave, a dash is added to the waveform that appears second. Conventionally, QRS synthesized wave types generally depend on pattern recognition by doctors, and an apparatus that automatically determines the type in detail for long-time data has not been found from the viewpoint of calculation cost.

The present invention automatically determines the type of the QRS composite wave according to a certain rule. FIG. 10 shows a table of rules for detecting the type of QRS composite wave. The pattern of the electrocardiogram signal waveform and the name of the feature point are described at the bottom of the table of FIG.

That is, when the main spine point is positive (a peak that is above the base line and pointed upward), the feature points are Q points and q points from the left (faster in time) to the right. , R point (main spine point), s point, S point. If the main spine point is negative (a peak below the baseline and pointed downward), the feature points are point Q and point r from the left (faster in time) to the right. , S point (main spine point), r ′ point, S point.

In the rule table shown in FIG. 10, when the upper column of each item is selected and followed, for example, the main spine point is positive, and between Q point and main spine point (between QM: Q point and main spine point). If the Q point is the minimum value (between spines) and the s point is the minimum value between the main spines and the S point, a rule that the Rs type is determined is described. This determination is made by the electrocardiogram automatic analysis software. According to this rule table for all other types, all types of requirements defined are determined.

The table shown in FIG. 11 shows the waveform pattern of the QRS composite waveform. The above Rs type is shown in the upper left frame of the eight frames, starting from the Q point, without the q point, the main spine point is positive, the s point, the S point and the feature point are It is the following waveform. Other types can be broadly classified into the eight waveforms shown in the table of FIG.

Next, description will be made of subdividing the QRS composite wave type and determining the detailed type name. The rules of classification differ depending on whether the main spike is positive or negative. First, the rule table for determining the detailed type name of the QRS composite wave when the main spike wave shown in FIG. 28 is positive will be described.

When the main spike is the Rs type shown in FIGS. 10 and 11, the absolute value of the difference between the heights of the Q point and the s point (voltage: the same applies hereinafter) is Hs (s is a lowercase letter), and the Q point and the R point Let HR be the absolute value of the difference in point height.
And D2 which is ratio of both is calculated | required with the following formula | equation.
D2 = Hs ÷ HR × 100 (%)

Here, a, b and c are units in%,
a <b <c
Is a constant that can be determined based on expert experience. Typical numbers are:
a = 5, b = 50, c = 90

And
When D2 ≦ a, R-s type a <D2 ≦ b, Rs type b <D2 <c, RS type c ≦ D2, and R = S type.

When the main spike is the R type shown in FIGS. 10 and 11, all the types of the fine classification are directly used as the R type regardless of other feature points.

When the main spike is of the qRs type shown in FIGS. 10 and 11, the absolute value of the height difference between the Q point and the s point is Hs (s is a small letter), and the height difference between the Q point and the R point is Let HR be the absolute value of H, and let Hq be the absolute value of the difference in height between the Q point and the q point.
And D1 and D2 which are ratio of both are calculated | required with the following formula | equation.
D1 = Hq ÷ HR × 100 (%)
D2 = Hs ÷ HR × 100 (%)

And D1 ≦ a,
When D2 ≦ a, when −qR−s type a <D2 ≦ b, when −qRs type b <D2 <c, when −qRS type c ≦ D2, −qR = S type .

And a <D1 ≦ b,
When D2 ≦ a, qR-s type a <D2 ≦ b, qRs type b <D2 <c, qRS type c ≦ D2, qR = S type.

And b <D1 <c,
When D2 ≦ a, QR-s type a <D2 ≦ b, QRs type b <D2 <c, QRS type c ≦ D2, and QR = S type.

And c ≦ D1, and
When D2 ≦ a, when Q = Rs type a <D2 ≦ b, when Q = Rs type b <D2 <c, when Q = RS type c ≦ D2, when Q = R = S type.

When the main spike is of the qR type shown in FIGS. 10 and 11, the absolute value of the difference in height between the Q point and the q point is Hq, and the absolute value of the difference in height between the Q point and the R point is HR. And
And D1 which is ratio of both is calculated | required with the following formula | equation.
D1 = Hq ÷ HR × 100 (%)

And
When D1 ≦ a, −qR type a <D1 ≦ b, qR type b <D1 <c, QR type c ≦ D1, Q = R type.

Next, a table of rules for determining the detailed type name of the QRS composite wave when the main spike wave is negative shown in FIG. 29 will be described. When the main spike is the qrS type shown in FIGS. 10 and 11, the absolute value of the height difference between the Q point and the S point (S is capital letter) is HS (S is a capital letter), and the Q point and the r point are Let Hr be the absolute value of the difference in height.
And D3 which is ratio of both is calculated | required with the following formula | equation.
D3 = Hr ÷ HS × 100 (%)

And
When D3 ≦ a, −rS type a <D3 ≦ b, rS type b <D3 <c, RS type c ≦ D3, R = S type.

When the main spike is of the qrSr ′ type shown in FIG. 10 and FIG. 11, the absolute value of the height difference between the Q point and the S point (S is a capital letter) is HS (S is a capital letter). The absolute value of the difference between the heights of the points is Hr, and the absolute value of the difference between the heights of the S point (S is capital letter) and the r ′ point is Hr ′.
And D3 and D4 which are ratio of both are calculated | required with the following formula | equation.
D3 = Hr ÷ HS × 100 (%)
D4 = Hr ′ ÷ HS × 100 (%)

And D3 ≦ a,
When D4 ≦ a, when −rS−r ′ type a <D4 ≦ b, when −rSr ′ type b <D4 <c, when −rSR ′ type c ≦ D4, −rS = R 'Type.

And a <D3 ≦ b,
When D4 ≦ a, rS−r ′ type a <D4 ≦ b, rSr ′ type b <D4 <c, rSR ′ type c ≦ D4, rS = R ′ type .

And b <D3 <c,
When D4 ≦ a, RS-r ′ type a <D4 ≦ b, when RSr ′ type b <D4 <c, when RSR ′ type c ≦ D4, RS = R ′ type .

And c ≦ D3, and
When D4 ≦ a, R = Sr ′ type a <D4 ≦ b, R = Sr ′ type b <D4 <c, R = SR ′ type c ≦ D4, R = S = R 'type.

In the case where the main spike is the QS type shown in FIGS. 10 and 11, all the types of the fine classification are directly used as the QS type regardless of other feature points.

When the main spike is the QSr type shown in FIGS. 10 and 11, the absolute value of the height difference between the Q point and the S point (S is a capital letter) is HS (S is a capital letter), and the Q point and the r point are Let Hr be the absolute value of the difference in height.
And D3 which is ratio of both is calculated | required with the following formula | equation.
D3 = Hr ÷ HS × 100 (%)

And
When D3 ≦ a, the Qr type a <D3 ≦ b, the Qr type b <D3 <c, the QR type c ≦ D3, and the Q = R type.

As for the detailed type name of the determined QRS composite waveform, a label indicating the type is recorded in a predetermined field of the recording area of the storage device built in the personal computer 12. This completes the description of the analysis of the detailed model name of the QRS composite waveform.

Seventh, the step of editing the result of automatically detecting the characteristics of each pulsation in the electrocardiogram signal obtained by measuring with a constant time using the electrocardiogram automatic analyzer 1 will be described. Editing refers to the characteristics of each automatically analyzed pulsation as a whole, and there are variations, etc., and if it is difficult to make a correct diagnosis as it is or there is a possibility that the diagnosis may be wrong, each qualified person such as a doctor This refers to correcting the characteristics of the beat so that it can be used as a consistent data for correct diagnosis. A doctor makes a comprehensive diagnosis by reviewing the edited data.

The electrocardiogram signal measured over 30 minutes contains about 2000 beats, and it is very time consuming to edit each one. In the present invention, an automatic editing function for collectively editing all measured beats or selected partial beats is provided, and the editing work can be performed simply and efficiently.

FIG. 31 is a diagram illustrating an example of a state before editing the main spine point. FIG. 31 (a) shows a part of the electrocardiogram signal S2 measured for 30 minutes in three rows. The time at the right end of the first row is t1 and continues to the left end of the second row. Further, the time at the right end of the second column is t2, which continues from the left end of the third column. The main spine point of each pulsation is marked with a black circle as the main spine mark 31. As described above, this is the result of the automatic electrocardiogram analysis apparatus according to the present invention automatically performing analysis by the main spine point detection S202 using a certain rule, and marking the main spine point according to the result.

Looking at each beat from the left end of the first row, the main spine mark 31 is attached to the positive peak of the R wave in the 6th and 9th beats from the 1st to the 4th consecutively. In the fifth, seventh and eighth beats, the main spine mark 31 is attached to the negative peak of the Q wave, and it can be seen that a positive peak and a negative peak are mixed. As a result, as shown in FIG. 31 (b), when the main spine point interval is indicated by a black circle, the time graph appears to be divided around three auxiliary lines, and the main spine point interval usable for diagnosis is obtained. Absent.

The reason why such a phenomenon occurs is that the peak values of the Q wave and the R wave are almost the same level, and because of slight changes for each beat, the Q wave is selected as the maximum peak or the R wave is selected. This is because the main spine point moves randomly between the peak of the Q wave and the R wave. This result is inevitable as long as automatic judgment is used. Therefore, it is necessary to edit to obtain the interval between the main spines that can be used for diagnosis.

In this case, the editing may be performed so that either the Q wave or the R wave peak is always the main spine. FIG. 32 is a diagram for explaining a state after editing the main spine point. FIG. 32A shows the result of inputting an editing command to the personal computer 12 so that the main spine point is the R wave peak for the pulsation where the Q wave is the main spine point. As can be seen from the figure, the main spine point of all beats has shifted to the peak of the R wave.

As a result of this editing, the main spine point interval time graph shown in FIG. 32B is distributed near one auxiliary line, and the main spine point interval reflecting the substance can be obtained.

Here, editing of the main spine point will be specifically described. FIG. 31 (c) shows a state in which the electrocardiogram signal S2 measured for 30 minutes shown in FIG. 31 (a) is separated in units of pulsations, and the main spines are overlapped for each pulsation. is there. This method uses the fact that the waveform is divided into an upper side and a lower side depending on whether the main spine point is positive or negative. The overlapping main spine point is in the center of the portion that appears to be X-shaped, and beats with a negative main spine point are displayed on the upper side, and beats with a positive main spine point are displayed on the lower side.

For example, when an instruction to make the polarity of the main spine point positive with respect to the waveform of the upper half waveform is input to the personal computer 12, negative main spine points are collectively changed to positive as shown in FIG. This is called morphological editing. As described above, since this editing operation is not performed for each beat, but is performed for all or a specified portion of the measured electrocardiogram signal S3 designated for the personal computer, it is very easy. It can be executed and editing time can be shortened.

Next, determination of arrhythmia annotation will be described. The arrhythmia annotation is usually a label that is attached to each beat when it is different from the standard waveform and the pulsation of interest is identified as an arrhythmia, assuming that the standard waveform is dominant if it is healthy. . Each beat is labeled N if it is a pulse that approximates a standard waveform, V if it is an arrhythmia of ventricular origin, and S if it is an arrhythmia of supraventricular origin.

This will be described in detail with reference to the drawings. As shown in FIG. 33, the arrhythmia annotation is determined by determining whether the width of the QRS synthesized wave of the standard waveform is smaller or larger than 120 ms. In this description, 120 ms is an example, and can be changed to an arbitrary time. The same applies hereinafter. First, a flowchart of the arrhythmia annotation determination S208 when the width of the QRS composite wave of the standard waveform is less than 120 ms is shown on the left side of FIG.

In the pulsation, whether the width of the QRS composite wave is 120 ms or more,
Or ventricular origin when the width of the QRS composite wave is less than 120 ms, premature contraction, the main spine point is not close, and the polarity of the T wave is opposite to that of the main spine point Arrhythmia and is labeled V.

In this pulsation, the width of the QRS composite wave is less than 120 ms, early premature contraction, the main spine point is not close, and the polarity of the T wave is the same as that of the main spine point When
Or when the width of the QRS composite wave is less than 120 ms, premature extrasystole, and the main spinal point is approaching,
It is determined as an arrhythmia of supraventricular origin and is labeled S.

If the beat is not labeled V or S,
That is, when the width of the QRS composite wave is less than 120 ms in the pulsation and it is not an early premature contraction, it is labeled N.

Next, FIG. 34 shows a flowchart of the arrhythmia annotation determination S209 when the width of the QRS composite wave of the standard waveform is 120 ms or more.
In the beat of the standard waveform, when the width of the QRS composite wave is 120 ms or more,
The waveform type of the QRS composite wave is different from the standard waveform (different type),
When the width of the QRS synthesis is 120 ms or more,
Alternatively, the width of the QRS synthesized wave of the standard waveform is 120 ms or more, the waveform type of the QRS synthesized wave is different from the standard waveform (different type), and the width of the QRS synthesized wave is less than 120 ms, and early early When it is contraction, the main spine is not close, and no P wave is observed,
Alternatively, the width of the QRS synthesized wave of the standard waveform is 120 ms or more, the waveform type of the QRS synthesized wave is the same type as the standard waveform, is an early premature contraction, and the main spine is not approaching, And when P wave is not recognized,
Arrhythmia of ventricular origin is determined and labeled V.

When the width of the QRS composite wave of the standard waveform is 120 ms or more,
The waveform type of the QRS composite wave is different from the standard waveform (different type),
In the pulsation, the width of the QRS composite wave is less than 120 ms,
Alternatively, the waveform type of the QRS composite wave is the same as the standard waveform,

And if it is an early premature contraction,
An arrhythmia of supraventricular origin is determined and labeled S when the main spine point is not close and a P wave is observed or the main spine point is close.

If the beat is not labeled V or S,
That is, the width of the QRS composite wave of the standard waveform is 120 ms or more,
The waveform type of the QRS composite wave is different from the standard waveform, and the QRS composite width is less than 120 ms,
Or when the waveform type of the QRS composite wave is the same as the standard waveform,
And if it is not an early premature contraction, it is labeled N.
This is the end of the description of the arrhythmia annotation determination.

Next, editing of the arrhythmia annotation will be described. FIG. 35 is a diagram for explaining a state before editing of the arrhythmia annotation. When attention is paid to the fourth beat from the left in the electrocardiogram signal S5, it indicates an early extrasystole instead of a beat having a correct period, and it should be detected as an arrhythmia, but the electrocardiogram automatic analyzer 1 of the present invention 1 It is shown that the arrhythmia annotation automatically classified and attached by N is N.

In other words, an arrhythmia annotation that can be used for diagnosis has not been obtained. This is because the width of the QRS composite wave of the standard waveform obtained from the electrocardiogram signal S4 after the “earlyness / partition point recognition / interval and width analysis” processing is as long as about 120 ms. This is because only the length of the dynamic QRS composite wave is not particularly long, and therefore it was not determined as an arrhythmia by the automatic determination rule in the early analysis S203 (see FIG. 20). Therefore, the morphology editing function shown in FIG. 35 (b) is used. Morphological editing is to grasp the characteristics of each waveform on the waveform by displaying the main spines of the waveform of each beat in an overlapping manner.

In FIG. 35 (b), not only the waveform of the pulsation but also the pulsation waveform before and after the pulsation waveform are displayed simultaneously. As a result, the base line is divided into two positions and the distance between the front and back waveforms is shifted. This makes it easy to generate a normal heartbeat waveform and an early contraction ECG signal waveform. Can be distinguished. Therefore, an edit command is input to the personal computer 12 so that the arrhythmia annotation of the electrocardiogram signal waveform indicating an abnormality is designated and changed from N to V, and editing can be performed by rewriting the label relating to the arrhythmia annotation.

In FIG. 35, only one abnormal pulsatile signal waveform contracting at an early stage is drawn due to the limitation of the size of the figure, but actually there are many in 30 minutes of electrocardiogram signals. Similar abnormal pulsation signals are included and can be displayed together and edited according to the judgment of a qualified person such as a doctor. Correcting while checking the signal waveform of each pulsation and the arrhythmia annotation attached to it, requires a lot of labor and time, but according to the editing function of the present invention, the waveform is displayed. You can specify multiple beats at once on the screen, which can be executed very easily and the editing time can be shortened.

Next, enlargement and display of the waveform will be described. FIG. 36 is a conceptual diagram of an enlarged display of a part of the electrocardiogram signal S3. FIG. 36 (a) shows a screen in which one minute of the electrocardiogram signal S3 obtained by measurement for 30 minutes is regarded as one line, and 30 lines are arranged vertically and displayed on a personal computer display device (this is compressed). This is a part of the figure. In this display, the detailed part of each beat cannot be seen. In the enlarged display, a part of the screen is enlarged to make it easy to see details of the waveform of the electrocardiogram signal S3.

FIG. 36A shows a situation in which a rectangle surrounding the part selected for enlargement is displayed as a result of selecting a part of the fourth line in the figure using a cursor or the like on the display screen. . When an enlargement command is given to the personal computer 12 at this time, only the measured value of the specific serial number designated in the field a61 in the data structure is selected and displayed. As shown in (b), the selected portion is greatly enlarged and displayed so that the selected portion can be seen. If this function is used, the details of the waveform of the electrocardiogram signal S3 can be confirmed.

Next, an example of a function of displaying a part of an electrocardiogram signal waveform on the display device by color coding will be described. FIG. 37 is a diagram for explaining color display in accordance with the characteristics of an electrocardiogram signal. A main spine point mark 31 or a main spine point mark 32 is attached to each beat of the electrocardiogram signal S4, so that the main spine point can be easily identified. The main spine mark 31 is a black circle, and the main spine mark 32 is a white circle.

At this time, the main spine mark 31 or the circle of the main spine mark 32 on the display device is recorded with a label for designating the color in the data structure by the color designation input to the personal computer 12. By rewriting the label that specifies the color of the field, it can be changed to an appropriate color. Further, by rewriting the label designating the arrhythmia annotation character, the arrhythmia annotation character can be changed corresponding to each main spine.

The color of the character indicating the arrhythmia annotation is, for example, N is a black character surrounded by a white square, and V is a black character surrounded by a red square. Can be changed to any color. These can be realized by selecting a field in which a label designating a color is recorded in the data structure and rewriting the label with another color information.

Next, a function for changing the order of executing various editing functions will be described. FIG. 38 is an explanatory diagram for changing the order of the editing steps. Since the analysis procedure differs depending on the hospital or individual doctor who uses the electrocardiogram automatic analyzer, it is preferable that the order of the editing process is not fixed and has a function that can be arbitrarily changed on the spot where it is used.

FIG. 38 shows a screen in which various editing functions are displayed on the display device of the personal computer 12. The editing process display section 35 displays the names of the editing functions, and has a switch section 34 at the head. The contents of the editing function shown in FIG. 38A will be described. “Automatic detection” means that an arrhythmia annotation is automatically added, and “defect detection unit mask” means an expert who performs editing, The doctor specifies the range to be excluded from the processing target, and “Confirm main spine detection” is the confirmation of the main spine point detected by automatic analysis and if it detects a point that is not the main spine point, If it is deleted and the main spine point is missed, it is added, and “arrhythmic detection” is to set judgment conditions such as bradycardia and pose. ”Is different depending on the doctor where the baseline is, so it is to specify where the baseline is before the main spine point,“ morphology editing ”is a skillful editing by overlaying the main spine point Person watching the ECG waveform pattern “Add important record” is to add comments such as doctor's findings to the ECG measurement result report, and “Edit report” is to select important waveforms, standard waveforms, etc. Combine the data with the report.

When a specific function in the editing process is to be executed, when a selection command is input from the mouse or keyboard of the personal computer 12 after placing the cursor at the head of the function name and selecting it, as shown in FIG. In addition, the color of the switch section 34 is changed to switch to a screen corresponding to the function. Therefore, any processing function can be executed in any order, not in the order from the top in which the processing functions are displayed.

For example, FIG. 38A shows a state where “morphological editing” is selected. In this way, it is possible to change the order of the editing process by selecting the order by the switch unit 34 and starting the “morphology editing” process before the “confirmation of main spine detection” process.

As another method, a selection command is input from the keyboard to the personal computer 12 to change the order of the editing function names displayed in the editing process display unit 35, and the editing functions are displayed in a different order. Each editing function can be automatically executed in order from the top. As shown in FIG. 38B, here, “morphology editing” is moved before “confirmation of main spine detection” with respect to FIG. 38A.

The automatic analysis of the electrocardiogram has been described using an example of using a personal computer connected to the electrocardiograph by wire. However, as shown in FIG. 39, an electrocardiogram signal is stored in a storage device such as a USB memory or an SD card. The information may be transferred from the electrocardiograph to the personal computer by replacing it.

As shown in the lower right of FIG. 12, the second embodiment is configured by an electrocardiograph, a portable information device, and a calculation processing server on the cloud. It is an electrocardiogram automatic analyzer that operates. The communication unit 117 of the electrocardiograph 11 and the portable information device 16 are connected by a short-range wireless information transmission 18. A smartphone or tablet personal computer that is the portable information device 16 and a calculation processing server 17 on the cloud are connected by a long-distance wireless information transmission 19.

The configuration and operation of the electrocardiograph 11 are the same as those in the first embodiment. The portable information device 16 is, for example, a smartphone or a tablet personal computer, and is a portable device having a calculation processing function, an input / output function, and a wireless communication function. The input / output function includes an image display, a speaker, an instruction lamp, and a keyboard, and is used by a subject or user to input an operation to the electrocardiogram automatic analyzer and receive a processing result. The short-range wireless information transmission 18 is, for example, Bluetooth (registered trademark). The long-distance wireless information transmission 19 is, for example, a mobile phone. The calculation processing server has a built-in storage device.

In the present embodiment, the combination of the portable information device 16 and the calculation processing server 17 on the cloud plays the role played by the personal computer 12 in the first embodiment. Moreover, the short-range wireless information transmission 18 and the long-range wireless information transmission 19 each perform the function performed by the wired information transmission 15.

As in the first embodiment, the analog electrocardiogram signal measured by the electrocardiograph 11 and subjected to preprocessing such as filtering is digitized to become an electrocardiogram signal S1, which is recorded in the memory 115 which is a transmission buffer. Immediately, it is wirelessly transmitted to the portable information device 16 via the communication unit 117. The storage capacity assigned to the electrocardiogram signal S1 in the internal memory of a smartphone or tablet computer used as the portable information device 6 is a measurement for 30 seconds, which is one measurement. Or it can also be made into 30 seconds or more as needed.

In the flow chart of the electrocardiogram automatic analysis process shown in FIG. 1, the part for performing the stability analysis S200, which is the first step, is separated from the electrocardiogram automatic analysis software and installed in the portable information device 16. The remaining part of the electrocardiogram automatic analysis software is installed in the calculation processing server 17 on the cloud, and the calculation processing server 17 performs the steps after determination and deletion S201 of the defective recording portion.

When the stability analysis S200 is performed according to the flowchart shown in FIG. 2 and, as a result, it is determined that the signal is stable, the signal is held in the portable information device 16 as an electrocardiogram signal S2, and on the cloud as necessary. It is automatically transmitted to the calculation processing server. If it is determined that the signal is not stable, an alarm signal is sent from the portable information device 16 to the electrocardiograph 11, and an alarm is displayed on the display lamp 120 of the electrocardiograph 11 to indicate that the signal is abnormal. Encourage (subject) to redo the measurement.

After the determination and deletion of the recording failure part S201, the smartphone or tablet personal computer as the portable information device 16 is used as an operation and display terminal, and the calculation processing server 17 on the cloud is transferred from the smartphone or tablet personal computer. With respect to the transmitted electrocardiogram signal S2, the processing of each step after “main spine point detection” S202 is performed along the electrocardiogram automatic analysis flowchart shown in FIG. It changes to ECG signals S3 to S5.

The operations and effects other than those described here are the same if the personal computer 12 in the first embodiment is replaced with the portable information device 16 or the calculation processing server 17.

In this way, an electrocardiogram signal is measured for 30 seconds or more with a one-channel electrocardiograph, and the measured value is digitized and automatically transmitted to a combination of a personal computer or a portable information device and a calculation processing server. , Memorize the measured value of the electrocardiogram signal, automatically analyze various characteristics of the waveform of the electrocardiogram signal according to the operation instruction input to the combination of personal computer or portable information device and calculation processing server, An electrocardiogram automatic analysis device that edits and displays an electrocardiogram, its automatic analysis result, and the edit result on a display device of a combination of a personal computer or a portable information device and a calculation processing server can be provided.

1 ECG automatic analysis device 11 ECG 110 Electrode 111 Preamplifier 112 Filter 113 A / D converter 114 CPU
115 Memory 116 Interface 117 Communication Unit 118 Operation Switch 119 Power Supply 120 Display Lamp 121 Case 12 Personal Computer 13 Computer 15 Wired Information Transmission 16 Portable Information Device 17 Calculation Processing Server 18 Short-Distance Wireless Information Transmission 19 Long-Distance Wireless Information Transmission 31 Mark (black circle)
32 Main spine mark (white circle)
33 Temporary spine mark 34 Switch part 35 Editing process display part 4 Right hand 5 Left hand 61 Field a
62 field b
63 Field c
64 fields d
65 field e
66 field n
S1 ECG signal (end of digitalization)
S2 ECG signal (Stability analysis finished)
S3 ECG signal (ends main spike detection)
S4 ECG signal (earlyness, classification point recognition finished)
S5 ECG signal (termination of QRS composite wave type detection)
S200 Stability analysis S201 Determination and deletion of defective recording S202 Main spine point detection S203 Earlyness analysis S204 Segment point recognition / spinal wave analysis S205 Spacing / width analysis between each wave S206 Standard waveform creation S207 QRS composite waveform Type analysis S208 Arrhythmia annotation determination S209 Arrhythmia annotation determination S300 Moving average S301 Drift test S302 Spike noise test S303 Saturation test S304 QRS composite wave test S305 Calculation window condition setting S306 Interval detection / determination S307 Temporary main spine point detection S308 Main spine point detection S309 Beat number inspection S310 Pose inspection S311 Missing inspection

Claims (24)

It consists of an electrocardiograph, a personal computer, and wired signal transmission connecting between them.
An electrocardiograph continuously measures an analog ECG signal for a certain amount of time,
The preprocessed analog ECG signal is converted into a digital signal and recorded as an ECG signal S1 in a storage device built in the personal computer,
The ECG signal S1 is subjected to a “stability analysis” process to determine whether a significant feature can be obtained by subsequent signal processing;
The signal that can be determined is the electrocardiogram signal S2,
The electrocardiogram signal S2 is determined by the “determination and deletion of defective recording portion” process to determine whether or not a significant feature can be obtained by signal processing for each beat section. Mark not to process
For all beats without this mark,
By detecting the main spine point having a spike-like peak for each beat by the “main spine point detection” process, the signal after the process is set as an electrocardiogram signal S3,
With respect to the electrocardiogram signal S3, a beat having a rhythm faster than a normal average beat is detected by an “early analysis” process,
At the same time, for the electrocardiogram signal S3,
Through the “divided point recognition / spinal wave analysis” process, the P wave, QRS synthesized wave, and T wave segment points are recognized, the respective boundary points are clarified, and the signal after the completion of the process is designated as an electrocardiogram signal S4. age,
In the electrocardiogram signal S4, a standard waveform of one beat that becomes a standard in the electrocardiogram signal S4 is created by “standard waveform creation” processing,
With respect to the electrocardiogram signal S4, the QRS composite wave type is determined for each pulsation signal by the “QRS composite wave type detection” process, and the type name of the QRS composite wave is determined for each pulsation of the electrocardiogram signal. The signal that has been processed is referred to as an electrocardiogram signal S5.
Processing for the electrocardiogram signals S1, 2, 3, 4, 5 is performed in the order described by the electrocardiogram automatic analysis software installed in the personal computer,
By combining the waveform of one or more of the electrocardiogram signals S1, 2, 3, 4, and 5 or a part of the section with a code indicating the characteristic obtained by the processing, the display device of the personal computer Display
Further, an electrocardiogram automatic analysis apparatus characterized in that information relating to an electrocardiogram signal can be transmitted to another computer / storage device through a network. It consists of an electrocardiograph, a portable information device, a calculation processing server on the cloud, and wireless signal transmission connecting them,
The electrocardiograph continuously measures an analog electrocardiogram signal for a certain period of time, converts the preprocessed analog electrocardiogram signal into a digital electrocardiogram signal, and forms an electrocardiogram signal S1;
The combination of the portable information device and the calculation processing server sends and receives the electrocardiogram signal S1 or a signal obtained by processing it by wireless transmission, and performs a plurality of processes on the digital electrocardiogram signal by the equipped electrocardiogram automatic analysis software,
The process includes
Determining whether a significant feature can be obtained by subsequent signal processing by performing a “stability analysis” process on the electrocardiogram signal S1;
Recording the determined signal as an electrocardiographic signal S2 in a storage device built in the portable information device and / or a storage device built in the calculation processing server;
With respect to the electrocardiogram signal S2, it is determined whether or not a significant feature can be obtained by signal processing for each beat interval by the “determination and deletion of defective recording portion” process. Mark it not to process
For all beats without this mark,
By detecting a main spine point having a spike-like peak for each beat by the “main spine point detection” process, this is used as an electrocardiogram signal S3;
For the electrocardiogram signal S3, detecting a pulsation having a rhythm faster than a normal average pulsation by an “early analysis” process;
At the same time, for the electrocardiogram signal S3,
Recognizing the division points of the P wave, QRS composite wave, and T wave by the “division point recognition / spinal wave analysis” process, clarifying the respective boundary points, and setting this as the electrocardiogram signal S4;
In the electrocardiogram signal S4, a “standard waveform creation” process is performed to create a standard waveform with a single beat,
With respect to the electrocardiogram signal S4, the QRS composite wave type is determined for each pulsation signal by the “QRS composite wave type detection” process, and the type name of the QRS composite wave is determined for each pulsation of the electrocardiogram signal. And including this as an electrocardiogram signal S5,
The processing is performed in the order described,
The waveform of one or more of the electrocardiogram signals S1, 2, 3, 4, 5 is combined with a waveform indicating a characteristic obtained by the processing in combination with a waveform of the whole or a part of the section. Display by display device,
An electrocardiogram automatic analyzer characterized in that it can also transmit information about an electrocardiogram signal to another computer / storage device through a wired or wireless network. An electrocardiogram automatic analyzer according to claim 2,
Perform stability analysis processing on a portable information device,
An electrocardiogram automatic analyzer characterized in that all processing other than the stability analysis processing is performed by a calculation processing server. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
About the electrocardiogram signal S1 obtained by converting an analog electrocardiogram signal measured by an electrocardiograph and removing a direct-current component and a low-frequency component into a digital signal,
Through stability analysis processing,
The amount of drift of the electrocardiogram signal S1 is below a certain value for a certain time;
The sum of absolute values of the fluctuation value of the electrocardiogram signal S1, that is, the value from peak to peak, is not more than a certain value for a certain time,
The value of the electrocardiogram signal S1 is within the maximum value and the minimum value, which are the limits of amplitude,
The QRS composite wave is a main spike wave with continuous amplitude of a certain level,
When all the above are established, it is determined that the electrocardiogram signal S1 is stable,
An electrocardiogram automatic analyzer characterized in that the electrocardiogram signal S1 is continuously recorded in a storage device built in a personal computer for a predetermined specific time. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
About the electrocardiogram signal S1 obtained by converting an analog electrocardiogram signal measured by an electrocardiograph and removing a direct-current component and a low-frequency component into a digital signal,
Taking the sum of absolute values of differences between all measurement points in a certain section of the electrocardiogram signal S1, and determining that the value obtained by dividing the sum by the size of the section is greater than or equal to a certain value, is spike noise,
When the standard deviation of all measurement points exceeds a certain threshold in a certain section of the electrocardiogram signal S1, it is determined as a drift,
When the absolute value of the electrocardiogram signal S1 is equal to or greater than a certain value in a certain interval, it is determined as saturated,
An electrocardiogram automatic analysis apparatus characterized in that the section of the electrocardiogram signal S1 for which any of the above is determined is determined as a recording failure portion, and the electrocardiogram signal S1 in the section is deleted. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
An electrocardiogram signal S2 measured by an electrocardiograph, converted from an analog electrocardiogram signal obtained by removing a DC component and a low-frequency component by a filter to a digital signal, and subjected to stability analysis processing, recording defect determination and deletion processing ,
By the main spine point detection process,
Calculate the moving standard deviation of the measured value within a certain calculation window width of the electrocardiogram signal S2,
A threshold is obtained from the moving standard deviation graph,
In the graph of the moving standard deviation, obtain a continuous section exceeding the threshold,
If there are multiple sections, combine the sections according to certain rules.
An electrocardiogram automatic analyzer characterized in that a peak having the largest absolute value in the combined section is determined as the main spine. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
An analog electrocardiogram signal measured by an electrocardiograph, with the DC component and low frequency component removed by a filter, is converted into a digital signal, and stability analysis processing, recording defect determination and deletion processing, and main spine point detection processing are completed. About the electrocardiogram signal S3
The early analysis process
After the detection of the main spine point, the average interval between the main spine points is obtained for the entire measured value for a certain period of time.
Determining the ratio of the mean interval to the interval between two consecutive main spines for all the main spines;
In the graph in which the ratios are arranged in order of magnitude, the median point and the minimum value point of the ratio are connected with a straight line on the graph,
The measurement point on the left side from the point where the distance from the straight line to the graph is the maximum is the first condition of the point where earlyness exists,
Determining the ratio of the interval between the two consecutive main spine points to the interval between the two previous consecutive main spine points for all the main spine points;
In the graph in which the ratios are arranged in order of size, the point of the median value and the point of the minimum value of the graph of the ratio are connected by a straight line,
The point on the left side from the point where the distance from the straight line to the graph is the maximum is the second condition of the point having earlyness,
An electrocardiogram automatic analyzer characterized by determining that there is earlyness at a main spine point where the first condition and the second condition are simultaneously satisfied. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
An analog electrocardiogram signal measured by an electrocardiograph, with the DC component and low frequency component removed by a filter, is converted into a digital signal, and stability analysis processing, recording defect determination and deletion processing, and main spine point detection processing are completed. About the electrocardiogram signal S3
In segment point recognition and spike wave analysis processing,
The interval between the two consecutive main spine points is defined as a first interval, and the ratio between the interval and the second interval obtained by adding the interval between the two previous consecutive main spine points is determined.
For automatic analysis of adjacent pulsation sections by setting the position going back from the main spine point as the starting point of the pulsation section for the time of the product of the ratio of the first interval and the second interval An electrocardiogram automatic analyzer characterized by clarifying boundary points. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
An analog electrocardiogram signal measured by an electrocardiograph, with the DC component and low frequency component removed by a filter, is converted into a digital signal, and stability analysis processing, recording defect determination and deletion processing, and main spine point detection processing are completed. About the electrocardiogram signal S3
Obtain an average value of the moving standard deviation of the electrocardiogram signal cut out with a certain calculation window width for all beat intervals within a specified specific time,
The position of the measurement point that is in front of the main spine point in the pulsation interval, is the earliest in time, the moving standard deviation is smaller than the average value, and the moving standard deviation after one is equal to or larger than the average value Is the starting point of the QRS composite wave,
In the pulsation section, the position of the measurement point that is after the main spine point, is earliest in time, the moving standard deviation is equal to or greater than the average value, and the moving standard deviation after one is smaller than the average value is QRS. An electrocardiogram automatic analyzer characterized by being the end point of a synthesized wave. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
An analog electrocardiogram signal measured by an electrocardiograph, with the DC component and low frequency component removed by a filter, is converted into a digital signal, and stability analysis processing, recording defect determination and deletion processing, and main spine point detection processing are completed. About the electrocardiogram signal S3
On the ECG signal waveform, draw a straight line connecting the starting point of the pulsation interval and the starting point of the QRS composite wave,
Within the interval between the start point of the pulsation interval and the start point of the QRS composite wave, an average value of the movement half-value width of the distance between the electrocardiogram signal waveform and the straight line is obtained,
An electrocardiogram automatic analyzer characterized in that a range of measurement points where the electrocardiogram signal waveform is larger than the average value is a P wave. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
An analog electrocardiogram signal measured by an electrocardiograph, with the DC component and low frequency component removed by a filter, is converted into a digital signal, and stability analysis processing, recording defect determination and deletion processing, and main spine point detection processing are completed. About the electrocardiogram signal S3
On the ECG signal waveform, draw a straight line connecting the end point of the beat section and the end point of the QRS composite wave,
Within the interval between the end point of the pulsation interval and the end point of the QRS composite wave, the average value of the movement half-value width of the distance between the measurement point and the straight line is obtained,
An electrocardiogram automatic analyzer characterized in that a range of measurement points larger than the average value is a T wave. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early About electrocardiogram signal S4 which finished the analysis processing of sex, recognition of division point / revision processing of interval and width between each wave,
Standard waveform creation process
Divide into all the beats that exist within a specific time,
Move each time axis for the divided electrocardiogram signals, and superimpose the positions of each main spine point on the time axis on the graph,
Similarly, move each voltage axis and superimpose the position of each main spine point on the voltage axis on the graph.
An electrocardiogram automatic analyzer characterized by obtaining a sum of measured values of all the divided electrocardiogram signals and dividing an average waveform by dividing the sum by the number of the divided electrocardiogram signals. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early About the electrocardiogram signal S4 after the analysis processing of the sex, the recognition of the division point / the analysis processing of the interval / width between each wave,
In classifying the waveform of the QRS composite wave,
In the feature point pattern of the waveform of the one-beat electrocardiogram signal S4, the first feature point is Q point, the end feature point is S point, and the downward peak between the main spine point and Q point existing between them Is named q point, and the downward peak between the main spine point and S point is named s point,
The case where the Q point is the minimum value point between the Q point and the main spine point, and the s point is the minimum value point between the main spine point and the S point is Rs type,
The case where the Q point is the minimum point between the Q point and the main spine point and the S point is the minimum value point between the main spine point and the S point is R-type,
When the q point is the minimum value point between the Q point and the main spine point, and the s point is the minimum value point between the main spine point and the S point, the qRS type,
When the q point is the minimum value point between the Q point and the main spine point, and the S point is the minimum value point between the main spine point and the S point, the main spine point is changed to the qR type. An electrocardiogram automatic analyzer characterized by performing type detection processing of the QRS synthesized wave that is positive. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early About the electrocardiogram signal S4 after the analysis processing of the sex, the recognition of the division point / the analysis processing of the interval / width between each wave,
In classifying the waveform type of the QRS composite wave,
In the feature point pattern of the waveform of the electrocardiogram signal S4 of one beat, the first feature point is Q point, the end feature point is S point, and the upward peak between the main spine point and Q point existing between them And the downward peak between the main spine point and the S point is named r ′ point,
The case where the r point is the maximum value point between the Q point and the main spine point, and the S point is the maximum value point between the main spine point and the S point is defined as qrS type,
The case where the r point is the maximum value point between the Q point and the main spine point, and the r ′ point is the maximum value point between the main spine point and the S point is defined as qrSr ′ type,
The case where the Q point is the maximum value point between the Q point and the main spine point, and the S point is the maximum value point between the main spine point and the S point is defined as a QS type,
When the Q point is the maximum value point between the Q point and the main spine point, and the r 'point is the minimum value point between the main spine point and the S point, the main spine point is obtained by using the QSr type. An electrocardiogram automatic analyzer characterized by performing a type detection process of the QRS composite wave in which is negative. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early About the electrocardiogram signal S4 after the analysis processing of the sex, the recognition of the division point / the analysis processing of the interval / width between each wave,
In determining the detailed model name of the QRS composite wave,
If the main spike is positive and a, b and c are constants,
When the type classification is Rs type, the absolute value of the difference between the Q point and the s point is set as Hs, the absolute value of the difference between the Q point and the R point is set as HR, and a ratio D2 = Hs / HR × 100 of both is obtained.
Rs type when D2 ≦ a,
Rs type when a <D2 ≦ b,
RS type when b <D2 <c,
When c ≦ D2, R = S type,
When the type classification is R type, the absolute value of the difference is not calculated and all are set to R type.
When the type classification is qRs type, the absolute value of the difference between the Q point and the s point is Hs, the absolute value of the difference between the Q point and the R point is HR, and the ratio D2 = (Hs / HR) × 100 of the two is obtained. Seeking
The absolute value of the difference between the Q point and the q point is set as Hq, the absolute value of the difference between the Q point and the R point is set as HR, and a ratio D1 = (Hq / HR) × 100 between the two is obtained.
D1 ≦ a,
When D2 ≦ a, −qRs type is assumed,
When a <D2 ≦ b, −qRs type,
When b <D2 <c, −qRS type,
When c ≦ D2, −qR = S type,
a <D1 ≦ b,
QR-s type when D2 ≦ a,
When a <D2 ≦ b, the qRs type is assumed.
When b <D2 <c, it is qRS type,
When c ≦ D2, qR = S type,
b <D1 <c,
QR-s type when D2 ≦ a,
When a <D2 ≦ b, the QRs type is assumed.
QR type when b <D2 <c,
When c ≦ D2, QR = S type,
c ≦ D1, and
When D2 ≦ a, Q = Rs type,
When a <D2 ≦ b, Q = Rs type,
When b <D2 <c, Q = RS type,
When c ≦ D2, Q = R = S type,
When the type classification is qR type, the absolute value of the difference between the Q point and the q point is Hq, the absolute value of the difference between the Q point and the R point is HR, and the ratio D1 = (Hq / HR) × 100 is obtained. Seeking
When D1 ≦ a, −qR type is assumed,
qR type when a <D1 ≦ b,
QR type when b <D1 <c,
An electrocardiogram automatic analyzer characterized by Q = R when c ≦ D1. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early About the electrocardiogram signal S3 after the analysis processing of the sex, the recognition of the division point / the analysis processing of the interval / width between each wave,
In determining the detailed model name of the QRS composite wave,
If the main spike is negative and a, b and c are constants,
When the type classification is qrS type, the absolute value of the difference between the Q point and the S point is HS, the absolute value of the difference between the Q point and the r point is Hr, and the ratio D3 = (Hr / HS) × 100 between the two Seeking
When D3 ≦ a, −rS type;
rS type when a <D3 ≦ b,
RS type when b <D3 <c,
R = S type when c ≦ D3,
When the type classification is qrSr ′ type, the absolute value of the difference between the Q point and the S point is HS, the absolute value of the difference between the Q point and the r point is Hr, and the ratio D3 = (Hr / HS) × Seeking 100,
The absolute value of the difference between the S point and the r ′ point is set to Hr ′, and a ratio D4 = (Hr ′ / HS) × 100 between the two is obtained.
D3 ≦ a,
When D4 ≦ a, the type is −rS−r ′,
When a <D4 ≦ b, −rSr ′ type,
When b <D4 <c, −rSR ′ type,
When c ≦ D4, −rS = R ′ type,
a <D3 ≦ b,
When D4 ≦ a, the rSr ′ type is assumed.
When a <D4 ≦ b, the rSr ′ type is assumed.
When b <D4 <c, rSR ′ type
When c ≦ D4, rS = R ′ type
b <D3 <c,
When D4 ≦ a, the RS-r ′ type is assumed.
When a <D4 ≦ b, the RSr ′ type is assumed.
When b <D4 <c, RSR ′ type
When c ≦ D4, RS = R ′ type,
c ≦ D3,
When D4 ≦ a, R = Sr ′ type,
R = Sr ′ type when a <D4 ≦ b,
When b <D4 <c, R = SR ′ type,
When c ≦ D4, R = S = R ′ type,
If the type classification is QS type, without calculating the absolute value of the difference, it is all QS type as it is,
When the type classification is the QSr type, the absolute value of the difference between the Q point and the S point is HS, the absolute value of the difference between the Q point and the r point is Hr, and the ratio D1 = (Hr / HS) × 100 between the two is obtained. Seeking
Qr type when D3 ≦ a,
Qr type when a <D3 ≦ b,
QR type when b <D3 <c,
An electrocardiogram automatic analyzer characterized by Q = R when c ≦ D3. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early In the electrocardiogram signal S5 which has finished the analysis processing of the sex, the segment point recognition / analysis of the interval / width between each wave, and the type detection processing of the QRS composite wave,
An electrocardiogram characterized by determining an arrhythmia annotation from the characteristics of a plurality of waveforms including the width of the QRS synthesized wave of the standard waveform, the width of the QRS synthesized wave of the waveform of the pulsation, the presence or absence of the P wave, and the polarity of the T wave. Automatic analyzer. It consists of a combination of an electrocardiograph and a personal computer or portable information device and a calculation processing server.
Converts an analog ECG signal measured by an electrocardiograph and removes DC and low-frequency components with a filter into a digital signal. Stability analysis processing, recording defect determination and deletion processing, main spine detection processing, early In the waveform of the electrocardiogram signal S4 after the analysis processing of the sex, the recognition of the division point / the analysis of the interval / width between each wave, the base line among the R wave, QS wave, S wave, and Q wave of two consecutive beats An electrocardiogram automatic analyzer characterized in that the interval between the peak points of the spikes having the peak value with the largest fluctuation is regarded as the main spine interval. In the electrocardiogram automatic analyzer shown in claim 5,
If you automatically detect multiple main spines and you have a mix of positive and negative main spines,
An electrocardiogram automatic analyzer characterized by having a morphological editing function for collectively changing a main spine point to one of the positive main spine point or the negative main spine point, and shortening the editing time. The electrocardiogram automatic analyzer according to claim 16,
About the result of automatically determining arrhythmia annotation,
It has an editing function to change the arrhythmia annotation by specifying the range and changing it at once by morphological editing,
An electrocardiogram automatic analyzer characterized in that editing time can be greatly shortened. An electrocardiogram automatic analyzer according to any one of claims 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18.
In the display screen that displays all the ECG signals obtained by measuring for a certain time on the display device at once,
An electrocardiogram automatic analyzer characterized by displaying a mark that can be distinguished from other parts at an arbitrary designated part of the waveform of the electrocardiogram signal in the display screen. An electrocardiogram automatic analyzer according to any one of claims 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18.
In the display screen that displays all the ECG signals obtained by measuring for a certain time on the display device at once,
An electrocardiogram automatic analyzer characterized by transitioning to a screen that displays an enlarged portion of an arbitrary part. An electrocardiogram automatic analyzer according to any one of claims 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18.
In the display screen that displays all or part of the ECG signals obtained by measuring for a certain time on the display device,
An electrocardiogram automatic analysis device characterized in that an electrocardiogram signal for each beat having a specific feature can be displayed in a color designated for each feature or a mark can be attached. An electrocardiogram automatic analyzer according to any one of claims 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18. A plurality of editing steps for editing the electrocardiogram signal or the characteristics of the electrocardiogram signal obtained by measuring the time of
An electrocardiogram automatic analyzer characterized in that the order of the editing steps can be arbitrarily selected.

Images