Electrocardiosignal analysis system
Through the filter processing and data processing technology of the ECG signal analysis system, various waveforms and deformations in the ECG signal are dynamically processed, solving the problem of misjudgment of noise in the prior art and improving the accuracy of arrhythmia judgment.
Patent Information
- Application Number
- CN202380030313.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-03-29
- Filing Date
- 2023-02-17
- Publication Date
- 2024-11-12
AI Technical Summary
The prior art cannot effectively consider various waveforms and waveform deformations of arrhythmia, which may lead to misjudgment of noise data and affect the accuracy of arrhythmia determination.
An electrocardiogram signal analysis system is designed. Through filter processing, unit waveform data extraction, data processing, reference data selection, similarity calculation and determination components, various waveforms and deformations in the electrocardiogram signal can be dynamically processed, and through interpolation The wave height adjustment generates judgment data, and calculates dynamic similarity to determine noise.
The accuracy of determining various waveforms and deformation of arrhythmia is achieved, and the accuracy and reliability of arrhythmia is improved.
Smart Images

Figure CN118946312A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an electrocardiographic signal analysis system for estimating a measurement state of an electrocardiographic signal based on an electrocardiographic signal (hereinafter referred to as electrocardiographic information), for example. Background Art
[0002] The heart repeatedly contracts and expands, acting as a pump that sends blood throughout the body. This activity is kept regular by weak electrical stimulation in the myocardial cells, but if an abnormality occurs in the electrical stimulation, an abnormality will also occur in the activity of the heart. The abnormality in the activity of the heart is called arrhythmia. There are various types of arrhythmia, for example, ventricular fibrillation, which is highly fatal and causes cardiac arrest, ventricular tachycardia, and atrial fibrillation, which causes cerebral infarction. Therefore, it is very important to determine arrhythmia, and arrhythmia is determined using the electrocardiogram information obtained from the determination object by an electrocardiogram signal measuring device.
[0003] The electrocardiogram information measures the potential of the heart through electrodes installed on the body surface of the subject. At this time, due to the floating of the electrodes and the influence of interference caused by body movement, noise is superimposed on the acquired electrocardiogram information. In particular, in a wearable electrocardiogram such as hitoe (registered trademark) that uses dry electrodes, it is easy to cause floating of the electrodes, so it is easy to superimpose noise. The electrocardiogram information superimposed with noise loses its characteristics as electrocardiogram information, and becomes an obstacle in the implementation of arrhythmia determination, etc. Therefore, it is preferable to exclude abnormal data, that is, noise data, from the determination object in advance.
[0004] For example, Patent Document 1 describes a method for generating data for arrhythmia determination (equivalent to electrocardiographic information) after excluding abnormal data, i.e., data that cannot be used for blood pressure estimation, from measured pulse wave information. Specifically, in Patent Document 1, for each pulse wave waveform, the similarity with the set reference data is used as an index, and thus a waveform that is not similar to the set reference data is determined as abnormal data.
[0005] In addition, a method of estimating the quality of measured electrocardiographic information is described in Patent Document 2. Specifically, in Patent Document 2, for each electrocardiographic waveform, the similarity with an adjacent waveform is used as an index to estimate the quality of the measured electrocardiographic information.
[0006] Prior art literature Patent Literature Patent Document 1: Japanese Patent Application Publication No. 2018-130319 Patent document 2: Japanese Patent Application Publication No. 2019-98183. Summary of the invention
[0007] Problems to be solved by the invention However, the waveforms of arrhythmias vary in shape depending on the type, and are often different from the general ECG waveform, i.e., the shape and timing of occurrence of normal sinus rhythm. In addition, the shape of each ECG waveform may be deformed depending on the state of the ECG signal measuring device or the subject, and the deformation may include changes in wave height. However, unlike the abnormal data described above, the deformed waveform does not lose the characteristics of the ECG waveform, and can therefore be used for arrhythmia determination.
[0008] However, the methods described in Patent Documents 1 and 2 do not take into account various waveforms of arrhythmia and waveform deformation. Therefore, if only Patent Documents 1 and 2 are applied, the waveform of arrhythmia and the deformed waveform in each electrocardiogram waveform may be erroneously determined as abnormal data, that is, noise data.
[0009] The present invention solves the above-mentioned problems of the prior art and provides an electrocardiographic signal analysis system capable of determining noise in consideration of various waveforms and deformed waveforms expressing arrhythmia.
[0010] Solutions to Solve Problems The electrocardiographic signal analysis system of the present invention for achieving the above-mentioned object includes the following structure. (1) An electrocardiogram (ECG) signal analysis system for acquiring and processing ECG information, wherein the ECG signal analysis system comprises: An input unit, receiving input of the electrocardiogram information; A filter processing unit, applying a filter that allows characteristic frequencies to pass through the electrocardiographic information, and outputting the electrocardiographic information after the filter processing; A unit waveform data extraction unit, which extracts unit waveform data by segmenting the waveform data of the electrocardiographic information processed by the filter based on the peak value of the R wave or the S wave; A data processing unit that generates processed data by adjusting the wave height of the unit waveform data; A reference data selection unit selects reference data as a reference of a measurement state from the processed data; a similarity calculation unit calculates the similarity between the reference data and the processed data of the determination object; and a determination unit determines whether the electrocardiographic information corresponding to the processed data of the determination object includes noise based on the similarity.
[0011] (2) The electrocardiographic signal analysis system according to (1), wherein the data processing unit determines whether the potential of a minimum point just before or just after the peak of the R wave exceeds a reference, or whether the potential of a maximum point just before or just after the peak of the S wave is lower than a reference, in the unit waveform data, to detect a peak defect, and generates interpolation data in which the peak defect portion is interpolated. The wave height is adjusted for the interpolated data.
[0012] (3) The electrocardiographic signal analysis system according to (1), wherein the data processing unit determines whether the potential of a maximum point just before or just after the peak of the R wave exceeds a reference, or whether the potential of a minimum point just before or just after the peak of the S wave is lower than a reference, in the unit waveform data, to detect a peak defect, and generates interpolation data in which the peak defect portion is interpolated. The wave height is adjusted for the interpolated data.
[0013] (4) The electrocardiographic signal analysis system according to (1), wherein the data processing unit determines whether the potential of the inflection point just before or after the peak of the R wave exceeds a reference, or whether the potential of the inflection point just before or after the peak of the S wave is lower than a reference, in the unit waveform data, to detect a peak defect portion, and generates interpolation data in which the peak defect portion is interpolated. The wave height is adjusted for the interpolated data.
[0014] (5) The electrocardiographic signal analysis system according to (1), wherein the data processing unit: inverts the wave height of the unit waveform data or the processed data; The wave height of the inverted data is adjusted.
[0015] (6) The electrocardiographic signal analysis system according to (1), wherein the data processing unit: performs time reversal on the unit waveform data or the processed data; The wave height of the inverted data is adjusted.
[0016] (7) The electrocardiographic signal analysis system according to any one of (1) to (6), wherein the similarity calculation unit calculates the similarity between the dynamically time-stretched reference data and the processed data to be determined.
[0017] (8) The electrocardiographic signal analysis system according to any one of (1) to (7), wherein the reference data selection unit selects reference data from determination data, the determination data being generated based on electrocardiographic information acquired from an installer who is equipped with a measuring device for acquiring the electrocardiographic information.
[0018] (9) An electrocardiographic signal analysis system according to any one of (1) to (7), wherein the reference data selection unit selects reference data from determination data, the determination data being generated based on electrocardiographic information acquired from a source other than an installer, the installer being equipped with a measuring device for acquiring the electrocardiographic information.
[0019] Effects of the Invention According to the present invention, it is possible to perform noise determination in consideration of various waveforms and deformed waveforms expressing arrhythmia. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 This is a diagram showing a configuration example of an electrocardiographic signal analysis system according to the first embodiment. Figure 2 It is a diagram showing a detailed example of electrocardiographic information used for analysis. Figure 3 : is a flowchart showing the noise determination process according to the first embodiment. Figure 4A This is a diagram (part 1) for explaining a processing example of the filter processing unit according to the first embodiment. Figure 4B This is a diagram (part 2) for explaining a processing example of the filter processing unit according to the first embodiment. Figure 4C This is a diagram (part 3) for explaining a processing example of the filter processing unit according to the first embodiment. Figure 5 This is a diagram showing a processing example of the unit waveform data extraction unit according to the first embodiment. Figure 6 This is a diagram showing a processing example of the data processing unit according to the first embodiment. Figure 7 This is a diagram for explaining the selection process of reference data in the first embodiment. Figure 8 This is a diagram showing a processing example of the similarity calculation unit according to the first embodiment. Fig. 9 It is a diagram for explaining an example of processing a waveform including noise. Fig.10 This is a diagram for explaining the processing of the data processing unit in the second embodiment. Fig.11A This is a diagram (part 1) for explaining a processing example of a data processing unit according to the second embodiment. Fig. 11B This is a diagram (part 2) for explaining a processing example of the data processing unit according to the second embodiment. Fig. 11C This is a diagram (part 3) for explaining a processing example of the data processing unit according to the second embodiment. Fig.12 This is a diagram (part 1) for explaining the processing of the data processing unit in the third embodiment. Fig.13 This is a diagram (part 2) for explaining the processing of the data processing unit in the third embodiment. DETAILED DESCRIPTION
[0021] Hereinafter, an embodiment of the electrocardiogram signal analysis system of the present invention will be described in detail based on the drawings. In addition, the present invention is not limited to this embodiment. In addition, each embodiment of the present invention is not independent, and can be combined and implemented appropriately.
[0022] (Implementation Method 1) Figure 1 FIG. 1 is a diagram showing a configuration example of an electrocardiographic signal analysis system according to Embodiment 1 of the present invention. Figure 1 The electrocardiographic information 10 to be analyzed is obtained from the subject 1 via the electrocardiographic signal measuring device 2. The electrocardiographic information 10 here indicates a small amount of current generated with the contraction / dilation of the heart. The subject 1 is not particularly limited to a human or an animal.
[0023] A wearable electrocardiograph is an example of the electrocardiograph signal measuring device 2. Specifically, the electrocardiograph signal measuring device 2 includes a plurality of electrodes provided on an accessory main body mounted on the subject 1, and an electrocardiograph 100 electrically connected to each of the electrodes.
[0024] The electrocardiometer 100 is an example of an electrocardiogram measuring device that measures the electrocardiogram signal of a subject. The electrocardiometer 100 has the function of continuously measuring the electrocardiogram signal (electrocardiogram information 10) of the subject, the function of storing the obtained electrocardiogram information 10, and the function of transferring data by communicating with the electrocardiogram signal analysis system 3. In addition to the above functions, the electrocardiometer 100 preferably has a control mechanism that starts the measurement of the electrocardiogram information 10 at the same time as the assembly to the electrocardiogram signal measuring device 2. As a measurement start method, there is, for example, a method of starting data measurement by connecting to a dedicated connector provided on the electrocardiogram signal measuring device 2, a method of starting data measurement by detecting the continuous electrocardiogram of the electrocardiogram 100 in intermittent measurement, etc., but it is not limited thereto. According to the control mechanism, the measurement of the electrocardiogram information 10 is performed only when the electrocardiogram signal measuring device 2 is assembled to the subject 1, so that useless power consumption and capacity consumption can be prevented. Alternatively, the electrocardiometer 100 may transfer data including the electrocardiographic information 10 to a server device, and the server device may transfer the electrocardiographic information 10 to the electrocardiographic signal analysis system 3 . In the following description, it is assumed that the subject 1 is a person wearing the electrocardiographic signal measurement device 2 .
[0025] The electrocardiographic information 10 acquired from the subject 1 includes information of the subject 1, acquisition date and time, location, and measurement results. The measurement results are waveform data 11 plotted with time on the horizontal axis and voltage on the vertical axis (see Figure 1 ). Figure 2 1 is a diagram showing a detailed example of electrocardiographic information 10 used for analysis. The waveform data 11 may include components called P wave 11a, Q wave 11b, R wave 11c, S wave 11d, and T wave 11e, and the shapes of these components are used as a criterion for determining arrhythmia. In addition, the shape of the waveform data 11 included in the electrocardiographic information 10 may change depending on the subject or the position of the electrocardiograph 100 in the subject.
[0026] Back to Figure 1 The electrocardiographic signal analysis system 3 of the present invention includes an input unit 31, a filter processing unit 32, a unit waveform data extraction unit 33, a data processing unit 34, a reference data selection unit 35, a similarity calculation unit 36, a determination unit 37, an output unit 38, a control unit 39 and a storage unit 40.
[0027] The input unit 31 receives input of the electrocardiographic information 10 output from the electrocardiographic signal measuring device 2. The input unit 31 may be composed of a connector electrically connected to the electrocardiographic signal measuring device 2, a communication device having a communication unit with the electrocardiographic signal measuring device 2, or an input port of a medium for storing data, and may also include a user interface such as a keyboard, a mouse, and a microphone.
[0028] The filter processing unit 32 performs filter processing on the waveform data 11 of the electrocardiographic information 10 .
[0029] The unit waveform data extraction unit 33 extracts the unit waveform data 13 from the waveform data that has been subjected to the filter processing.
[0030] The data processing unit 34 processes the unit waveform data 13 into determination data 14 .
[0031] The reference data selection unit 35 selects data (reference data) serving as a reference for noise determination from the determination data 14 generated by the data processing unit 34 .
[0032] The similarity calculation unit 36 calculates the similarity between the reference data and the determination data 14 acquired from the determination target.
[0033] The determination unit 37 determines whether the electrocardiographic information 10 includes noise based on the similarity. The determination unit 37 determines the presence or absence of noise by comparing the similarity with a preset threshold value. The threshold value is set as a boundary value for determining the presence or absence of noise with respect to the similarity.
[0034] The output unit 38 is composed of an output device such as a display device and a printer, and outputs various information under the control of the control unit 39. The output unit 38 has, for example, a display composed of liquid crystal or organic EL (Electro Luminescence), a speaker, or other audio output function.
[0035] The control unit 39 integrally controls the operation of the electrocardiographic signal analysis system 3. The control unit 39 also causes the display to display the determination result of the determination unit 37, the electrocardiographic information 10, and information on normality / arrhythmia determined based on the electrocardiographic information 10, or outputs them to the outside.
[0036] The storage unit 40 stores various programs for operating the electrocardiogram signal analysis system 3, various data including data generated by each unit, etc. The various programs also include a determination program executed using a learning model. The storage unit 40 is composed of a ROM (Read Only Memory) pre-installed with various programs, a RAM (Random Access Memory) storing operation parameters and data of each process, an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc.
[0037] Various programs can also be recorded in a computer-readable recording medium such as an HDD, a flash memory, a CD-ROM, a DVD-ROM, a Blu-ray (registered trademark), and widely circulated. In addition, the input unit 31 can also obtain various programs via a communication network. The communication network referred to here is, for example, a network composed of an existing public line network, a LAN (Local Area Network), a WAN (Wide Area Network), etc., whether wired or wireless.
[0038] The electrocardiographic signal analysis system 3 having the above functional structure is a computer composed of one or more hardware such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), and an FPGA (Field Programmable Gate Array).
[0039] Next, the noise determination process performed based on the electrocardiographic information 10 obtained from the subject 1 will be described. Figure 3 : is a flowchart showing the noise determination process of Embodiment 1. In the noise determination process, first, the input unit 31 acquires the electrocardiographic information 10 obtained from the subject 1 via the electrocardiographic signal measurement device 2 (step S101).
[0040] The filter processing unit 32 performs filter processing on the waveform data 11 obtained in step S101 (step S102). The filter processing unit 32 reduces the difference between normal sinus rhythm and arrhythmia and makes the difference from noise obvious, for example, by allowing the waveform data 11 showing the relationship between time and intensity (here, voltage) to pass only the characteristics of the electrocardiographic signal. Specifically, the filter processing unit 32 applies a frequency filter to the electrocardiographic information 10 input by the input unit 31, and generates electrocardiographic information (waveform data) that has been processed by the filter and only passes the characteristic frequencies of normal sinus rhythm and arrhythmia. As a specific passband of the frequency filter, 5 to 50 Hz is set, but it is not limited to this. The frequency filter is designed using a general IIR (Infinite Impulse Response) filter or a FIR (Finite Impulse Response) filter.
[0041] Figure 4A to Figure 4C is a diagram for explaining an example of processing by the filter processing unit of the first embodiment. Figure 4A to Figure 4C When the waveform data 11A to 11C shown in (a) are subjected to filter processing, filter-processed waveform data 12A to 12C are generated respectively (refer to Figure 4A to Figure 4C The generated filtered waveform data 12 is registered in the storage unit 40 .
[0042] After the filter process, the unit waveform data extraction section 33 generates the unit waveform data 13 by partially extracting the waveform data 12 (step S103 ). Figure 5 1 is a diagram showing an example of processing by the unit waveform data extraction unit according to the first embodiment. Figure 5 The waveform data 12 shown in (a) is generated by extracting unit waveform data 13 for each specified unit (refer to Figure 5 (b)) is stored in the storage unit 40. The unit waveform data 13 may be extracted, for example, by specifying a period based on time axis information or by specifying a number of waveforms based on the peak position of the R wave 12a as an extraction reference position, but is not limited thereto. Figure 5 In the example shown, the peak position 12ap of the R wave 12a is detected, and a specified number of seconds before and after the peak position 12ap of the R wave 12a is set as one unit. Thus, the unit waveform data 13 represents a single waveform based on the waveform data 12 .
[0043] Here, the intervals between the unit waveform data 13 are not fixed due to the fluctuation of the heartbeat in the waveform data 12 (or waveform data 11). However, by segmenting each unit waveform, it is expected that the data processing accuracy of the data processing unit 34 and the accuracy of the determination of arrhythmia can be improved. In addition, since the problem of Figure 5Since there is no section having information useful for determination like the out-of-unit waveform data 12b in (b), the amount of processing can be suppressed. In addition, Figure 5 In the example shown, in order to segment each unit waveform, a specified number of seconds (for example, 0.4 seconds) before and after the peak position 12ap of the R wave 12a detected from the local maximum value of a certain interval is used as one unit waveform, but the present invention is not limited thereto. When the waveform data 11 does not have the R wave 11c and the peak position 12ap of the R wave 12a cannot be detected, the unit waveform data extracting unit 33 obtains the local maximum value of the waveform data 12, uses the time as the reference position, and extracts the specified number of seconds before and after as one unit.
[0044] The data processing unit 34 processes the unit waveform data 13 into the determination data 14 (step S104). In the present embodiment 1, as described above, the data processing unit 34 generates the determination data 14 by adjusting the wave height. The data processing unit 34 generates the determination data 14 with the wave height adjusted for the unit waveform data 13. The adjustment of the wave height is a process of correcting the wave heights of the unit waveform data 13 in such a way that the wave heights of the unit waveform data 13 become the same, and when the noise is determined in the determination unit 37 based on the determination data 14, the different effects of the wave heights of the unit waveforms can be ignored. As a method of adjusting the wave height, a method using the R wave peak value and the S wave peak value, a method using the R wave peak value and the offset, etc. are conceivable, but not limited to this. The data processing unit 34 stores the generated determination data 14 in the storage unit 40.
[0045] Figure 6 1 is a diagram showing a processing example of the data processing unit according to Embodiment 1. The data processing unit 34 adjusts the voltage width (wave height) H1 from the peak position 12ap of the R wave 12a to the peak position 12cp of the S wave 12c in the unit waveform data 13 .
[0046] After the determination data 14 is generated, the reference data selection unit 35 selects the determination data as the reference data from the generated determination data 14 (step S105). The reference data has the function of enabling relative evaluation when determining whether the determination data 14 includes noise in the determination unit 37. Specifically, the reference data is assumed to be a normal sinus rhythm or each arrhythmia, i.e., an electrocardiographic waveform, and is used to determine whether the determination data 14 includes noise by comparing with the determination data 14. For example, it is preferable to select determination data obtained based on electrocardiographic information measured in a state where the electrocardiographic signal measuring device 2 is worn in an appropriate assembly order, and determination data obtained based on electrocardiographic information measured in a state where the electrocardiographic signal measuring device 2 is assembled to the subject 1 with the assistance of an operator such as a doctor. The selection may also be electrocardiographic information at a moment after a predetermined time has passed since the start of the measurement, such as electrocardiographic information measured later by the electrocardiograph 100.
[0047] Figure 7 This is a diagram for explaining the selection process of reference data in Embodiment 1. The reference data selection unit 35 selects determination data as reference data B1 from, for example, the determination data 14A to 14C generated by the data processing unit 34 .
[0048] In addition, in the present embodiment 1, it is described that the reference data is selected based on the judgment data obtained from the subject 1 of the judgment object, that is, the judgment data of the same subject, but the reference data may also be selected based on the judgment data obtained from other than the judgment object. Originally, the electrocardiographic information 10 obtained by the electrocardiographic signal measuring device 2 is different for each subject and has individual differences, but by creating data with reduced individual differences through the above-mentioned filter processing and data processing, even data obtained from other than the judgment object can be processed as reference data. In addition, it is preferable to select reference data for a subject with a tendency to ventricular hypertrophy or the like based on the judgment data 14 obtained from the judgment object. Since the electrocardiogram information obtained from such a subject tends to be different from the electrocardiogram information of a healthy subject to a degree exceeding individual differences, the judgment accuracy can be ensured by preparing unique reference data. The method of selecting reference data is not limited to this, and determination data based on electrocardiographic information selected by the operator observing the electrocardiographic information 10 or determination data based on electrocardiographic information obtained by averaging waveforms acquired within a preset period may be selected. The reference data selection process may be executed after a predetermined number of determination data are acquired after the acquisition of the electrocardiographic information 10 is started, or reference data may be selected from a plurality of determination data including determination data acquired in the past based on electrocardiographic information of a subject different from the subject 1 .
[0049] After the reference data is selected, the similarity calculation unit 36 calculates the similarity between the determination data 14 and the reference data (step S106). The similarity calculation unit 36 calculates the dynamic similarity as the similarity. The similarity here refers to the distance between the reference data and the determination data 14, and the greater the distance, the lower the value of the similarity. Moreover, the lower the similarity, the more suspected that the determination data 14 includes noise. In addition, the dynamic similarity refers to the similarity calculated by partially stretching the reference data and the determination data 14 in the time direction. As a method for calculating the dynamic similarity, a well-known method such as the dynamic time warping method can be applied. By using the dynamic similarity, the influence of the deviation in the time direction caused by the arrhythmia or deformed waveform can be ignored, so that a more accurate noise judgment can be made.
[0050] Figure 8 1 is a diagram showing an example of processing by the similarity calculation unit of the first embodiment. The similarity calculation unit 36 generates, for example, a waveform L2 by scaling the waveform L1 of the reference data B1 in the time axis direction. In addition, the points between the waveforms L1 and L2 show the trajectory until the waveform L1 becomes the waveform L2. The similarity calculation unit 36 calculates the distance using the scaling waveform closest to the determination data 14 for the scaling processed waveform such as the waveform L2.
[0051] Fig. 9 It is a diagram for explaining an example of processing a waveform including noise. Fig. 9 The waveform data 11D shown is data including noise, and the determination data 14D is generated by data processing. The distance of the determination data 14D is far and the similarity is small. For example, in the case of determination data generated based on electrocardiographic information not including noise, the similarity is greater than 0.8, and in the case of determination data generated based on electrocardiographic information including noise, the similarity is less than 0.8. In this case, for example, the threshold value for determination is set to 0.8.
[0052] The determination unit 37 determines whether the electrocardiographic information 10 corresponding to the determination data 14 includes noise based on the similarity calculated in step S106 (step S107). The determination unit 37 compares the similarity with a threshold value to determine whether noise is included, and outputs the determination result. The determination result of whether each determination data output is noise is simply associated as bool-type noise index data or stored in the storage unit 40 as replaced electrocardiographic information after replacing the noise part with other values for the corresponding electrocardiographic information, but the present invention is not limited thereto.
[0053] After the noise is determined, the control unit 39 outputs the determination result (step S108). When the control unit 39 determines that the noise is not included in step S107, it outputs information indicating that the noise is not included. On the other hand, when the control unit 39 determines that the noise is included in step S107, it outputs information indicating that the noise is included. The output unit 38 outputs sound or light to notify the determination result, or displays the determination result in a terminal owned by the subject 1 or the operator. In addition, the control unit 39 may also store the determination result in the storage unit 40. The above-described determination process is assumed to be executed every time new electrocardiographic information 10 is input, or every time a certain amount of electrocardiographic information 10 is accumulated, but the present invention is not limited thereto.
[0054] According to the present embodiment 1 described above, the waveform data 11 is subjected to filter processing that only passes the characteristic frequencies of normal sinus rhythm and arrhythmia, and the wave height is adjusted, thereby generating the judgment data 14, and the dynamic similarity is calculated based on the reference data to judge the presence or absence of noise. Therefore, it is possible to judge the noise after taking into account various waveforms and deformed waveforms that express arrhythmia.
[0055] (Implementation Method 2) Next, the second embodiment of the present invention will be described. The structure of the electrocardiogram signal analysis system of the second embodiment is the same as that of the electrocardiogram signal analysis system 3 of the first embodiment, and therefore the description thereof will be omitted. The following describes the parts that are different from the first embodiment. In the second embodiment, the processing of the data processing unit 34 is different from that of the first embodiment. In addition, the processing flow of the noise determination is different from that of the first embodiment. Figure 3 The flowchart shown is the same.
[0056] In the second embodiment, the data processing unit 34 detects the peak missing portion of the R wave or S wave peak value, and interpolates data for the detected peak missing portion. The peak missing portion here refers to the missing of the peak value of the R wave or S wave of the obtained waveform due to poor measurement of the electrocardiogram signal measuring device 2 or the influence of interference. The peak missing portion becomes a cause of reducing the similarity in the similarity calculation unit 36.
[0057] Fig.10 This is a diagram for explaining the processing of the data processing unit in the second embodiment. The data processing unit 34 detects the unit waveform data 13A (see Fig.10 (a)) in region R D The peak missing portion of the R wave shown in FIG. 1 is detected, and interpolation data 15 (see FIG. 14 ) is generated by interpolating data for the detected peak missing portion. Fig.10 (b)). Fig.10 In (b), the interpolated waveform is shown by a dotted line.
[0058] The data processing unit 34 first detects the peak defect portion of the determination data 14. As a method for detecting the peak notch, for example, the peak defect can be detected by using the extreme value just before or just after the R wave / S wave peak and comparing the extreme value with a preset reference. Specifically, for example, it is determined whether the potential of the minimum point just before or just after the R wave peak exceeds the reference to detect the peak defect, or it is determined whether the potential of the maximum point just before or just after the R wave peak exceeds the reference to detect the peak defect. In addition, for the S wave, it is also determined whether the potential of the maximum point just before or just after the S wave peak is lower than the reference to detect the peak defect, or it is determined whether the potential of the minimum point just before or just after the S wave peak is lower than the reference to detect the peak defect. In this way, it is possible to detect a spike-shaped peak defect. As a method for setting a benchmark for detecting peak deficiency, for example, it is conceivable to use the baseline (zero point) of the electrocardiographic information 10, use a value of 50% of the potential from the baseline to the peak of the R wave, etc., but regardless of the actual method, it is preferable to make appropriate adjustments with reference to the value obtained, for example, by the above method.
[0059] In addition, as another method for detecting peak deficiency, it is possible to detect peak deficiency by using an inflection point just before or just after the R wave / S wave peak and comparing the potential of the inflection point with a preset reference. Specifically, for example, it is determined whether the potential of an inflection point just before or just after the R wave peak exceeds the reference to detect peak deficiency, or it is determined whether the potential of an inflection point just before or just after the S wave peak is lower than the reference to detect peak deficiency. In this way, it is possible to detect a peak deficiency site caused by deformation.
[0060] Next, the data processing unit 34 interpolates data for the detected peak missing portion. As a method for interpolating the peak missing portion, a known method such as general spline interpolation can be adopted.
[0061] Figures 11A to 11C 1 is a diagram for explaining an example of processing by a data processing unit in Embodiment 2. Figures 11A to 11C The interpolated waveform is shown by the dotted line in . Through the peak defect detection and interpolation described above, interpolation data 15 in which the defect portion is interpolated is obtained.
[0062] [When using a minimum point slightly before or after the peak] The data processing unit 34 detects, for example, a local minimum point T immediately before the peak value of the R wave with respect to the unit waveform data 13B. 11 or the later minimum point T 12 , determine whether the potential of the minimum point exceeds the reference, and detect the peak defect (refer to Fig.11A Then, the data processing unit 34 interpolates the interpolation waveform L3 for the peak missing part (refer to Fig.11A (b)) to generate interpolation data 15A (refer to Fig.11A (c)). As a result, the peak position of the R wave changes from the peak position T 21 becomes the peak position T 31 .
[0063] (When using the maximum point slightly before or after the peak) The data processing unit 34 detects, for example, a local maximum point T immediately before the peak value of the R wave peak value with respect to the unit waveform data 13C. 13 or the later maximum point T 14 , determine whether the potential of the maximum point exceeds the reference, and detect the peak defect (refer to Fig. 11B Then, the data processing unit 34 interpolates the interpolation waveform L4 for the peak missing part (refer to Fig. 11B (b)) to generate interpolation data 15B (refer to Fig. 11B (c)). As a result, the peak position of the R wave changes from the peak position T 22 becomes the peak position T 32 .
[0064] (When using an inflection point slightly before or after the peak) The data processing unit 34 detects, for example, an inflection point T immediately before the peak of the R wave with respect to the unit waveform data 13D. 15 or later inflection point T 16 , determine whether the potential of the inflection point exceeds the reference, and detect the peak loss (refer to Fig. 11C (a)). Then, the data processing unit 34 interpolates the interpolation waveform L5 for the peak missing part (refer to Fig. 11C (b)) to generate interpolation data 15C (refer to Fig. 11C (c)). As a result, the peak position of the R wave changes from the peak position T 23 becomes the peak position T 33 .
[0065] After the interpolation process, the data processing unit 34 performs a wave height adjustment on the interpolation data 15 .
[0066] In the second embodiment described above, a filter process is performed on the waveform data 11 to pass only the characteristic frequencies of normal sinus rhythm and arrhythmia, and the peak missing part is interpolated to generate the determination data 14 after adjusting the wave height of the interpolation data 15, and the dynamic similarity is calculated based on the reference data to determine the presence or absence of noise. According to the second embodiment, by performing the filter process and data processing, it is possible to determine the noise after taking into account various waveforms and deformed waveforms that express arrhythmia.
[0067] (Implementation 3) Next, the third embodiment of the present invention will be described. The structure of the electrocardiogram signal analysis system of the third embodiment is the same as that of the electrocardiogram signal analysis system 3 of the first embodiment, and therefore the description thereof will be omitted. The following will describe the parts that are different from the first embodiment. In the third embodiment, the processing of the data processing unit 34 is different from that of the first embodiment. In addition, the processing flow of the noise determination is different from that of the first embodiment. Figure 3 The flowchart shown is the same.
[0068] Here, in the similarity calculation unit 36, when calculating the similarity between data with a stronger influence of the R wave (hereinafter referred to as R-type data) and data with a stronger influence of the S wave (hereinafter referred to as S-type data), the accuracy of the similarity calculation is sometimes lower than the similarity calculated between the R-type data or the S-type data.
[0069] Therefore, in the third embodiment, the data processing unit 34 generates the inversion data 16 in which at least one of the wave height and the time is inverted with respect to the unit waveform data 13 or the determination data 14 . Fig.12 and Fig.13 This is a diagram for explaining the processing of the data processing unit in the third embodiment.
[0070] When the wave height is inverted, the data processing unit 34 generates, for example, inverted data 16A (see Fig.12 Here, the inversion of the wave height is synonymous with the process of inverting the sign of the unit waveform data 13 or the determination data 14 in the amplitude direction (vertical axis). Therefore, the inverted data 16A after the wave height inversion can also be regarded as the original inverted data.
[0071] When time-reversing, the data processing unit 34 generates, for example, the inversion data 16B (see Fig.13 Here, time reversal is synonymous with the process of reversing the sign of the unit waveform data 13 or the determination data 14 in the time direction (horizontal axis). Therefore, the time-reversed inversion data 16B can also be regarded as the inversion (inverse reproduction) data of the original data.
[0072] In addition, the data processing unit 34 may also generate the inversion data 16C in which the wave height and time are inverted. In addition, the inversion object can be determined by setting, and the determination data 14 may include data that has been inverted and data that has not been inverted. In this case, the determination unit 37 may also set different thresholds for the various data. In addition, the plurality of determination results outputted for the various data are synthesized by logical OR, logical AND, or a combination thereof, and finally output a determination result.
[0073] After the inversion process, the data processing unit 34 performs a wave height adjustment on the inversion data 16 .
[0074] In the third embodiment described above, a filter process is performed on the waveform data 11 that passes only the characteristic frequencies of normal sinus rhythm and arrhythmia, and at least one of the wave height and time is inverted to generate the determination data 14 after the wave height is adjusted. The dynamic similarity is calculated based on the reference data to determine the presence or absence of noise. According to the third embodiment, by performing the filter process and data processing, it is possible to determine the noise after taking into account various waveforms and deformed waveforms that express arrhythmia.
[0075] According to the third embodiment, by generating the determination data 14 in which the wave height and time are reversed, it is possible to calculate the similarity between the R-type data and the S-type data with high accuracy.
[0076] (Implementation 4) Next, a fourth embodiment of the present invention will be described. The configuration of the electrocardiographic signal analysis system of the fourth embodiment is the same as that of the electrocardiographic signal analysis system 3 of the first embodiment, and thus the description thereof will be omitted. The following describes the parts different from the first embodiment.
[0077] In the present fourth embodiment, in addition to the above-mentioned processing, the determination unit 37 calculates the ECG acquisition rate based on the determination result, and the output unit 38 outputs the ECG acquisition rate calculated as the determination result to the subject 1 or the operator. The ECG acquisition rate refers to the proportion of the non-noise interval in the measurement time, that is, the interval effective for arrhythmia determination. At this time, data with a low ECG acquisition rate needs to be re-measured according to the situation because the accuracy of arrhythmia determination or signal acquisition efficiency is low.
[0078] According to the fourth embodiment, since the electrocardiogram acquisition rate is calculated and output by the control unit 39 , the subject 1 or the operator can immediately determine whether to re-measure. In addition, all or part of the functions of the electrocardiogram signal analysis system 3 may be included in the electrocardiograph 100 .
[0079] (Other embodiments) So far, the method for implementing the present invention has been described, but the present invention should not be limited to the above-mentioned embodiment. For example, it is assumed that the electrocardiogram signal analysis system 3 and the electrocardiogram signal measuring device 2 are separately provided for description, but the electrocardiogram signal analysis system 3 can also be integrally configured with the electrocardiogram signal measuring device 2. In addition, at least a part of the functions of the electrocardiogram signal analysis system 3 can also be provided on the electrocardiograph 100. For example, the electrocardiograph 100 can also include an input unit 31, a filter processing unit 32, a unit waveform data extraction unit 33, a data processing unit 34, a reference data selection unit 35, a similarity calculation unit 36, a determination unit 37, an output unit 38, and a control unit 39.
[0080] Industrial Applicability By using the electrocardiographic signal analysis system of the present invention, it is possible to perform noise determination in consideration of various waveforms and deformed waveforms expressing arrhythmia.
[0081] Description of Reference Numerals 1: Subject 2: ECG signal detector 3: ECG signal analysis system 10: ECG information 11, 11A~11D, 12, 12A~12C: Waveform data 13, 13A~13E: Unit waveform data 14, 14A~14D: Judgment data 15, 15A~15C: interpolation data 16, 16A~16C: Reverse data 31: Input section 32: Filter processing unit 33: Unit waveform data extraction unit 34: Data Processing Department 35: Benchmark data selection department 36: Similarity calculation unit 37: Judgment Department 38: Output 39: Control Department 40: Storage 100: Electrocardiogram.
Claims
1. An electrocardiogram signal analysis system, which obtains electrocardiogram information and performs processing, wherein: The electrocardiogram signal analysis system comprises: An input unit, receiving input of the electrocardiogram information; A filter processing unit, applying a filter that allows characteristic frequencies to pass through the electrocardiographic information, and outputting the electrocardiographic information after the filter processing; A unit waveform data extraction unit, which extracts unit waveform data by segmenting the waveform data of the electrocardiographic information processed by the filter based on the peak value of the R wave or the S wave; A data processing unit that generates processed data by adjusting the wave height of the unit waveform data; a reference data selection unit that selects reference data serving as a reference for a measurement state from the processed data; a similarity calculation unit for calculating a similarity between the reference data and the processed data to be determined; as well as The determination unit determines whether the electrocardiographic information corresponding to the processed data as a determination target includes noise based on the similarity.
2. The electrocardiogram signal analysis system according to claim 1, wherein: The data processing unit includes: The peak defect is detected by determining whether the potential of the minimum point slightly before or after the peak of the R wave exceeds a reference, or whether the potential of the maximum point slightly before or after the peak of the S wave is lower than a reference, and interpolation data is generated by interpolating the peak defect portion. The wave height is adjusted for the interpolated data.
3. The electrocardiogram signal analysis system according to claim 1, wherein: The data processing unit includes: The peak defect is detected by determining whether the potential of the maximum point slightly before or after the peak of the R wave exceeds a reference, or whether the potential of the minimum point slightly before or after the peak of the S wave is lower than a reference, and interpolation data is generated by interpolating the peak defect portion. The wave height is adjusted for the interpolated data.
4. The electrocardiogram signal analysis system according to claim 1, wherein: The data processing unit includes: By determining whether the potential of the inflection point slightly before or after the peak of the R wave exceeds a reference, or whether the potential of the inflection point slightly before or after the peak of the S wave is lower than a reference, a peak defect portion is detected, and interpolation data is generated by interpolating the peak defect portion. The wave height is adjusted for the interpolated data.
5. The electrocardiogram signal analysis system according to claim 1, wherein: The data processing department: For the unit waveform data or the processed data, the wave height is reversed. The wave height of the inverted data is adjusted.
6. The electrocardiogram signal analysis system according to claim 1, wherein: The data processing department: For the unit waveform data or the processed data, time is reversed. The wave height of the inverted data is adjusted.
7. The electrocardiogram signal analysis system according to claim 1, wherein: The similarity calculation unit calculates the similarity between the dynamically time-stretched reference data and the processed data to be determined.
8. The electrocardiogram signal analysis system according to any one of claims 1 to 7, wherein: The reference data selection unit selects reference data from determination data generated based on electrocardiographic information acquired from a wearer who wears a measurement device that acquires the electrocardiographic information.
9. The electrocardiogram signal analysis system according to any one of claims 1 to 7, wherein: The reference data selection unit selects reference data from determination data generated based on electrocardiographic information acquired from a person other than a wearer, the wearer being equipped with a measuring device that acquires the electrocardiographic information.
