Electrocardiogram analyzer and control method thereof
The electrocardiogram analysis device addresses the limitations of QRS-based classification by using P-wave features for supraventricular arrhythmia classification, enhancing efficiency and usability through user-assisted classification tools.
Patent Information
- Application Number
- JP2021177656
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-10-29
- Publication Date
- 2025-10-29
- Estimated Expiration
- 2041-10-29
AI Technical Summary
Existing automatic classification functions for electrocardiograms rely on QRS interval characteristics, which are inadequate for evaluating atrial electrical activity, necessitating manual assignment of supraventricular arrhythmia classifications, a time-consuming process.
An electrocardiogram analysis device and method that utilizes P-wave feature amounts for heartbeat classification, providing a classification screen for user-assisted assignment of supraventricular arrhythmias, with features like P-wave histogram and XY plot for intuitive classification and editing.
Enables efficient classification of supraventricular arrhythmias by allowing users to collectively assign classifications based on P-wave features, improving usability and reducing manual effort.
Smart Images

Figure 0007762039000001 
Figure 0007762039000002 
Figure 0007762039000003
Abstract
Description
[Technical Field]
[0001] The present invention relates to an electrocardiogram analyzer and a control method thereof. [Background technology]
[0002] To detect symptoms such as arrhythmia that are difficult to detect with short-term electrocardiogram measurements, long-term electrocardiogram measurements are performed using a Holter monitor. The quality of electrocardiograms measured with a Holter monitor is not consistent and they contain waveforms for tens of thousands of beats. Therefore, it is common for technicians and doctors to perform various analyses based on the results of automatic analysis by an electrocardiogram analyzer (Patent Documents 1 and 2).
[0003] One of the functions provided by the automatic analysis device is an automatic classification function that classifies the measured electrocardiogram beat by beat according to predetermined conditions. The user (technologist) of the automatic analysis device can evaluate the automatic classification results and change the classification unit or heartbeat unit, or create a new classification. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-95552 [Patent Document 2] Japanese Patent Publication No. 2020-130335 Summary of the Invention [Problem to be solved by the invention]
[0005] Previous automatic classification functions used the characteristics of the QRS interval of the electrocardiogram waveform. Because the QRS interval indicates the electrical activity of the ventricles, it is not suitable for evaluating the electrical activity of the atria. Therefore, in the past, technicians had to manually assign atrial (supraventricular) arrhythmia classifications beat by beat, which was a time-consuming process.
[0006] The present invention has been made in consideration of the problems of the conventional technology, and in one aspect thereof, provides an electrocardiogram analysis device and a control method thereof that can provide a function of assisting in classifying heartbeats related to supraventricular arrhythmias. [Means for solving the problem]
[0007] The above object is to provide a method for detecting a heartbeat of an electrocardiogram by using an acquisition means for acquiring electrocardiogram data, a calculation means for calculating a P-wave feature amount for each heartbeat of the electrocardiogram data, an output means for outputting a classification screen for classifying and displaying the heartbeats of the electrocardiogram data based on the P-wave feature amount, and an editing means for providing a function for collectively assigning a classification designated by a user to at least one of the heartbeats selected and the heartbeats not selected on the classification screen, and the classification screen is configured to display the selected heartbeat among the heartbeats that have been classified. Multiple Waveform for heart rate Overlay The area to display and the area not selected Multiple Waveform for heart rate Overlay The object of the present invention is achieved by an electrocardiogram analysis device having a display area. [Effects of the Invention]
[0008] With this configuration, the present invention can provide an electrocardiogram analysis device and a control method thereof that can provide information specific to electrocardiograms measured over multiple days. [Brief explanation of the drawings]
[0009] [Figure 1] 1 is a block diagram illustrating an example of the functional configuration of a general-purpose computer as an example of an electrocardiogram analysis apparatus according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram showing an example of a waveform list screen presented by the electrocardiogram analysis apparatus according to the embodiment. [Figure 3] FIG. 10 is a diagram showing an example of a P wave classification screen presented by the electrocardiogram analysis device according to the embodiment. [Figure 4] FIG. 10 is a diagram showing another example of a P wave classification screen presented by the electrocardiogram analysis device according to the embodiment. [Figure 5]4 is a flowchart illustrating the operation of the electrocardiogram analysis apparatus according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION
[0010] The present invention will be described in detail below based on exemplary embodiments with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the claimed invention. Furthermore, although multiple features are described in the embodiments, not all of them are necessarily essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same reference numerals are used to designate the same or similar components, and redundant explanations will be omitted.
[0011] In the following embodiments, the present invention will be described with reference to a general-purpose computer such as a personal computer or a tablet terminal, but the present invention can also be implemented in any electronic device, such as a media player, a smartphone, or a game console.
[0012] 1 is a block diagram showing an example of the functional configuration of a general-purpose computer 100 capable of functioning as an electrocardiogram analysis device according to this embodiment. A CPU 1 functioning as a control unit implements functions according to the program by, for example, reading a program stored in a storage device 10 into a RAM 3 and executing the program. For example, by executing a specific application program (electrocardiogram analysis application) while the operating system (OS) is running on the general-purpose computer 100, the general-purpose computer 100 functions as the electrocardiogram analysis device according to this embodiment.
[0013] The storage device 10 is, for example, a hard disk drive (HDD) or a solid state drive (SSD), and stores operating system (OS), device drivers, applications, user data, etc. The storage device 10 also stores GUI (Graphical User Interface) data for displaying menu screens, user setting data, initial setting data for applications, etc.
[0014] ROM2 stores programs, firmware, and various setting information required for starting up the computer, such as a bootstrap loader. At least a portion of ROM2 may be rewritable.
[0015] The RAM 3 is used as an area for expanding the programs executed by the CPU 1 and as a temporary storage area for variables, data, etc. Furthermore, part of the RAM 3 may be used as a video memory.
[0016] The memory card 4 is a recording medium that can be inserted into the card slot 5 and removed from the card slot 5. The general-purpose computer 100 can read data from the memory card 4 inserted into the card slot 5 and write data to the memory card 4. In this embodiment, the general-purpose computer 100 acquires data such as a long-term electrocardiogram recorded by a Holter electrocardiograph through the memory card 4. Note that the general-purpose computer 100 may acquire the electrocardiogram data to be automatically analyzed by other methods. For example, the electrocardiogram data to be automatically analyzed may be acquired from the Holter electrocardiograph that performed the measurement or from another external device 200 that stores the measured electrocardiogram data, by communication via a communication interface 20 (described later).
[0017] The display unit 6 includes a display device such as a liquid crystal display (LCD) or an organic EL display, a display control circuit, etc. Although Fig. 1 shows a configuration in which the display unit 6 is built into the general-purpose computer 100, the display unit 6 may be external. Also, both a built-in display unit 6 and an external display unit 6 may be included.
[0018] The operation unit 8 is a device with which the user inputs instructions to the general-purpose computer 100, and is typically one or more input devices such as a keyboard, a pointing device (such as a mouse), or a contact-sensing device (such as a touch panel). The keyboard may be a hardware keyboard or a software keyboard. The touch panel may be provided on the display unit 6, or may be in the form of a touchpad, such as those commonly found on notebook computers.
[0019] The communication interface (I / F) 20 is hardware that enables the general-purpose computer 100 to communicate with the external device 200 in accordance with a predetermined standard. The communication I / F 20 has a configuration according to the communication standard it supports, such as a connector that complies with a wired communication standard and a wireless transmitter / receiver that complies with a wireless communication standard. The communication I / F 20 may support multiple communication standards. There are no particular restrictions on the communication standards that the communication I / F 20 supports, but typical examples of wired communication standards include Ethernet (registered trademark) and USB, and typical examples of wireless communication standards include Bluetooth (registered trademark) and wireless LAN (IEEE802.11x).
[0020] A user (e.g., a technician) who uses automatic analysis of an electrocardiogram measured by a Holter electrocardiograph starts the automatic analysis application stored in storage device 10 by an operation method corresponding to the OS running on general-purpose computer 100. CPU 1 reads the automatic analysis application from storage device 10 into RAM 3 and executes it, causing general-purpose computer 100 to function as an electrocardiogram analyzer. Hereinafter, general-purpose computer 100 functioning as an electrocardiogram analyzer will be referred to as electrocardiogram analyzer 100.
[0021] Next, we will explain the operation of the electrocardiogram analysis device 100, which is realized by the CPU 1 executing an automatic analysis application. The CPU 1 acquires a data file storing electrocardiogram data to be automatically analyzed in response to an operation from the operation unit 8. The CPU 1 may, for example, present a file browser screen provided by the OS on the display unit 6, and allow the user to specify the data file to acquire. Alternatively, the CPU 1 may search a predetermined location, such as the memory card 4, and automatically acquire a data file that meets predetermined conditions (for example, a data file with a specified file name).
[0022] The CPU 1 stores the acquired data file in the storage device 10. Instead of acquiring the entire data file all at once, the electrocardiogram data stored in the data file may be acquired in fixed amounts at a time. In this embodiment, the electrocardiogram data is assumed to be digital data that has been A / D converted under predetermined conditions. The type and number of leads included in the electrocardiogram data are not particularly limited. The electrocardiogram data may include one or more leads measured by a typical Holter electrocardiograph, such as NASA lead, CC5 lead, and CM5 lead, or the standard 12 leads. The measurement period for electrocardiogram data using a Holter electrocardiograph is generally 24 hours or more, but a measurement period of 24 hours or more is not required.
[0023] The CPU 1 first applies a quality check process to the electrocardiogram data to detect poor signal intervals that are unsuitable for analysis. For example, the CPU 1 evaluates the baseline level and superimposed noise, and identifies intervals where the superimposed level exceeds a threshold or noise intervals as poor signal intervals and excludes them from the analysis process. When measurements are taken on multiple channels (leads) using a Holter electrocardiograph, the CPU 1 applies a quality check process to the data on each channel.
[0024] The CPU 1 applies quality check processing to all data from the start to the end of measurement, and stores information that can identify the detected bad signal intervals (e.g., the start date and time and the end date and time of the interval) in association with identification information for the data file (e.g., the file name) in the storage device 10. The CPU 1 applies the following processing to intervals of the electrocardiogram data excluding the bad signal intervals (referred to as analysis intervals or valid signal intervals).
[0025] Next, CPU 1 detects QRS interval candidates from the electrocardiogram data. If multiple channels (leads) are measured using a Holter electrocardiograph, CPU 1 detects QRS interval candidates from the data of each channel. CPU 1 then detects the candidate that is determined to be highly reliable as the final QRS interval. Reliability can be determined based on one or more of the following: signal quality, degree of noise contamination, validity of the RR interval, and whether or not it has been detected in other leads.
[0026] The CPU 1 then calculates a characteristic amount (parameter) based on the QRS interval. The calculated parameters include, but are not limited to, the RR interval (RRI) between adjacent QRS intervals, the width of the QRS interval, the height of the QRS interval, the direction of the QRS interval, and the area of the QRS interval. Furthermore, the CPU 1 can calculate other parameters, such as heart rate variability (HRV), from the parameters based on the QRS interval.
[0027] The CPU 1 also executes an automatic classification function that divides the electrocardiogram data into beats and assigns a classification to each beat based on multiple predetermined conditions related to parameters based on the QRS interval, the shape of the QRS interval, etc. The classifications assigned here include, but are not limited to, normal heartbeat (N), premature ventricular contraction (S), bundle branch block (B), paced heartbeat (P), etc. Each classification may have subclassifications. For example, normal heartbeat (N) may be further subdivided into multiple subclassifications (N1 to Nx, where x is an integer equal to or greater than 2). The CPU 1 also classifies heartbeats included in poor signal intervals as noise (A).
[0028] In this embodiment, the CPU 1 further detects P waves for each beat and obtains P-wave-related feature quantities (parameters). P-wave-related feature quantities may be, for example, but are not limited to, the PR interval, PQ interval, P-wave height (level), P-wave width, and P-wave polarity (convex upward or convex downward). Note that when P waves cannot be detected or when P waves are abnormal and feature quantities cannot be obtained, these are used as feature quantities.
[0029] The CPU 1 stores all the parameters calculated for each heartbeat in the storage device 10 in association with the identification information of the heartbeat (for example, a consecutive number starting from 1 assigned to the heartbeats from the start to the end of the measurement section) and the electrocardiogram data or the identification information of the electrocardiogram data (such as a file name).
[0030] 2 is a diagram showing an example of a waveform list screen 200 presented on the display unit 6 by the electrocardiogram analysis device 100. The waveform list screen 200 is a screen that displays a list of representative waveforms for each heart rate group that have been assigned the same classification by the automatic classification function.
[0031] A classification tab 210 is provided for each classification assigned by the automatic classification function. The CPU 1 displays the representative waveform of the classification corresponding to the selected tab among the classification tabs 210 in the waveform display area 230. FIG. 2 shows a state in which the tab corresponding to normal heartbeat (N) is selected. Therefore, the representative waveform of the heartbeat classified as normal heartbeat (N) is displayed in the waveform display area 230. Here, since the normal heartbeat (N) classification has nine sub-classifications N1 to N9, the representative waveform for each sub-classification is displayed.
[0032] A representative waveform is a waveform that represents the waveforms of heartbeats assigned to the same classification. The representative waveform may be a waveform selected according to any criteria, such as the waveform that most closely matches the criteria of the classification. The representative waveform is displayed together with the classification (subclassification) and heartbeat number. The heartbeat number is a sequential number assigned starting with 1 for the first heartbeat, and serves as identification information for the heartbeat. In this case, two leads out of the multiple leads included in the electrocardiogram data are set as targets for automatic analysis, so a representative waveform is displayed for each lead.
[0033] The user checks the representative waveform and, if necessary, other waveforms to determine the validity of the automatic classification. By operating the waveform list screen 200, the user can perform editing tasks such as combining multiple sub-classifications or changing classifications by sub-classification unit.
[0034] As described above, the automatic classification function classifies heartbeats according to the characteristics of the QRS interval of the heartbeat waveform, and therefore does not classify supraventricular arrhythmias. In this embodiment, to assist in the classification of supraventricular arrhythmias, a function is provided in which the user collectively assigns classifications to multiple heartbeats based on P-wave feature quantities. Specifically, upon detecting an operation of the P-wave classification button 220 on the waveform list screen 200, the CPU 1 causes the display unit 6 to display a P-wave classification screen, which provides a function in which the user collectively assigns classifications to multiple heartbeats based on P-wave feature quantities.
[0035] 3 is a diagram showing an example of a P-wave classification screen 300. The P-wave classification screen 300 provides a function for a user to select multiple heartbeats based on P-wave features and assign a classification to them collectively. In the example of FIG. 3, the P-wave classification screen 300 includes a histogram 310 of the heartbeats with respect to the P-wave features, allowing a user to collectively select multiple heartbeats based on the histogram 310.
[0036] The feature amount of the P wave used in the histogram 310 can be changed using a pull-down menu 311. Fig. 3 shows a case where the PQ interval is specified as the feature amount of the P wave. The CPU 1 generates a heartbeat histogram 310 using predetermined conditions (number and width of bins, range of values) for the feature amount selected by operating the pull-down menu 311, and displays the histogram 310 on the P-wave classification screen 300.
[0037] For example, if the height of the P wave is specified as the feature of the P wave, the histogram 310 can be generated so that the horizontal axis has a bin containing 0.0 mV in the center, with negative values on the left and positive values on the right. Note that heartbeats in which a P wave cannot be detected and heartbeats in which the P wave feature for generating the histogram 310 has a clearly abnormal value are excluded from the histogram 310 as unclassified heartbeats.
[0038] When there are multiple leads to be the subject of automatic analysis, the CPU 1 generates the histogram 310 for one predetermined lead. The user can change the lead for which the histogram 310 is generated by selecting a radio button 312. In this way, the user can display the histogram 310 for the desired P-wave feature quantity for the desired lead on the P-wave classification screen 300.
[0039] The histogram 310 allows the selection of one or multiple consecutive bins of P-wave features by operating the operation unit 8. When a selection operation on the histogram 310 is detected, the CPU 1 recognizes this as an operation to select heartbeats classified into the selected range or bin of P-wave features. Figure 3 shows a state in which a large value range 313 is selected in the histogram 310 for PQ intervals.
[0040] When the CPU 1 detects a selection operation on the histogram 310, (1) Heartbeats included in the selection range (selected), (2) heartbeats that are not included in the selection range (not selected), and (3) Unclassified heartbeats (heartbeats not displayed in the histogram 310 and heartbeats for which P-wave features could not be obtained) For each of these, waveforms are displayed in overlapping order.
[0041] In FIG. 3, the P-wave classification screen 300 displays overlapping heartbeat waveforms in a waveform display area 320 located to the right of a histogram 310. "Inside selected range," "Outside selected range," and "Unclassified" correspond to the heartbeats (1) to (3) above, respectively. The overlapping waveform display can be achieved by overlapping the waveforms of the corresponding multiple heartbeats, aligning their horizontal positions with the R waves. For convenience, FIG. 3 shows the same overlapping display for (1) to (3).
[0042] By displaying the waveforms of the heart rate groups selected by the user according to the selection conditions set by the user in an overlapping manner, the appropriateness of assigning the same classification to the selected heart rate groups can be intuitively grasped. The overlapping display of waveforms is sufficient if it is performed for at least the selected heart rate groups.
[0043] However, by presenting a waveform superimposed display for the heart rate groups that were not selected, it becomes possible to select the heart rate to be excluded and collectively assign a classification to the remaining heart rate groups, thereby improving usability.Furthermore, by presenting a waveform superimposed display for the heart rate groups that were not displayed in the histogram 310, it becomes possible to check the waveforms of the heart rate groups that were not displayed in the histogram 310 or collectively classify them as noise, again improving usability.
[0044] The display color of the coordinates constituting the superimposed display may be changed depending on the frequency of the waveforms overlapping the coordinates. This makes it possible to distinguish between parts with a high degree of commonality and parts with a low degree of commonality in the superimposed heartbeat waveform group, which is useful for the user to adjust the heartbeat selection conditions.
[0045] Furthermore, some of the heartbeats whose waveforms are displayed in an overlaid fashion may be selectable in response to an operation on the overlaid display via the operation unit 8, and the selected heartbeat may be excluded from targets for which a classification is to be assigned collectively. For example, the CPU 1 detects a position or area designation operation on the overlaid display as an operation for selecting a heartbeat corresponding to a waveform passing through the designated position or area. The CPU 1 then erases the waveform of the selected heartbeat from the overlaid display and also excludes it from targets for which a classification is to be assigned to the heartbeat group whose waveforms are displayed in an overlaid fashion.
[0046] A judgment code change button 321 is disposed adjacent to the superimposed display of waveforms. When the CPU 1 detects an operation on the judgment code change button 321 via the operation unit 8, it displays a judgment code change menu 322. The judgment code change menu 322 displays assignable classifications so that they can be selected. When the CPU 1 detects a selection operation on the judgment code change menu 322 via the operation unit 8, it collectively assigns the selected classifications to the heart rate groups whose waveforms are superimposed and closes the P-wave classification screen 300. Note that the P-wave classification screen 300 may remain displayed until an explicit instruction is received from the user.
[0047] 4 is a diagram showing another example of a P-wave classification screen, in which the same components as those in the P-wave classification screen 300 shown in Fig. 3 are given the same reference numerals and their descriptions will be omitted. Instead of the histogram 310, the P-wave classification screen 400 has an area (XY plot) 410 in which individual heartbeats are plotted in a two-dimensional area in which a first feature amount of the P wave is assigned to the first axis and a second feature amount of the P wave is assigned to the second axis.
[0048] Here, an XY plot 410 is shown in which the PQ interval is assigned as the first feature amount and the P wave height is assigned as the second feature amount. The first and second feature amounts may be changeable by the user through a pull-down menu operation, similar to the P wave classification screen 300, for example.
[0049] When an operation to select a two-dimensional area within the XY plot 410 is detected, the CPU 1 recognizes this as an operation to select a heartbeat corresponding to the plot included in the selected two-dimensional area. 4 shows a state in which a rectangular area 411 is selected via the operation unit 8, for example, and the CPU 1 displays the waveform of the heartbeat corresponding to the plot included in the rectangular area 411 superimposed on the area within the "selected range" of the waveform display area 320.
[0050] Furthermore, CPU1 superimposes and displays the waveforms of heartbeats corresponding to plots not included in rectangular area 411 in the "outside selected range" area of waveform display area 320, and the waveforms of heartbeats not plotted in XY plot 410 in the "P wave not detected" area of waveform display area 320.
[0051] The assignment of classifications using the determination code change button 321 and determination code change menu 322, and the exclusion of heartbeats in response to operations on the superimposed display are as described with reference to FIG.
[0052] By using the XY plot 410, it becomes possible to select a heartbeat in consideration of the relationship between the two P-wave feature amounts. Furthermore, by operating the operation unit 8, the user can switch between displaying the histogram 310 and the XY plot 410, allowing the user to select a heartbeat group based on various selection conditions related to the P-wave feature amounts.
[0053] Next, the waveform editing operation of the electrocardiogram analyzer 100, which is realized by the CPU 1 executing the automatic analysis application, will be described with reference to the flowchart shown in FIG. In S501, the CPU 1 acquires a data file storing electrocardiogram data to be automatically analyzed in response to an operation from the operation unit 8. The CPU 1 may, for example, present a file browser screen provided by the OS on the display unit 6, and allow the user to specify the data file to be acquired. Alternatively, the CPU 1 may search a predetermined location such as the memory card 4, and automatically acquire a data file that meets predetermined conditions (for example, a data file with a predetermined file name).
[0054] The CPU 1 stores the acquired data file in the storage device 10. It should be noted that, instead of acquiring the entire data file all at once, the electrocardiogram data stored in the data file may be acquired in fixed amounts at a time. In this embodiment, the electrocardiogram data is assumed to be digital data that has been A / D converted under predetermined conditions. The electrocardiogram data includes the types and number of leads from which X, Y, and Z leads can be derived. For example, the electrocardiogram data may include three leads, eV1, eV5, and eVF, or the standard 12 leads, but is not limited to these.
[0055] In S503, CPU1 applies preprocessing to the electrocardiogram data. The preprocessing may be a quality check process that detects poor signal intervals unsuitable for analysis from the electrocardiogram data. In the quality check process, CPU1 evaluates, for example, the baseline level and superimposed noise, and determines intervals where the superimposed level exceeds a threshold or noise intervals as poor signal intervals and excludes them from the analysis process. When measurements are taken on multiple channels (leads) using a Holter electrocardiograph, CPU1 applies preprocessing to the data on each channel.
[0056] The CPU 1 applies preprocessing to all data from the start to the end of measurement, and stores information that can identify the detected bad signal interval (e.g., the start date and time and the end date and time of the interval) in association with identification information for the data file (e.g., the file name) in the storage device 10. The CPU 1 applies the subsequent processing to intervals of the electrocardiogram data excluding the bad signal intervals (referred to as analysis intervals or valid signal intervals).
[0057] In S505, the CPU 1 calculates predetermined feature quantities (parameters) for the electrocardiogram data related to the leads to be automatically analyzed among one or more leads included in the electrocardiogram data. The CPU 1 divides the electrocardiogram data for the leads for which feature quantities are to be calculated into one heartbeat and detects QRS intervals. For example, the CPU 1 detects QRS interval candidates based on signal levels, and selects the candidate determined to be highly reliable as the final QRS interval. The reliability can be determined based on one or more of, for example, signal quality, the degree of noise contamination, the validity of the RR interval, and whether or not the QRS interval has been detected in other leads.
[0058] The CPU 1 can then calculate feature quantities such as the level of the R wave and the width of the QRS interval based on the detected QRS interval. Furthermore, the CPU 1 can calculate HR and its variability (HRV) from the R-R interval (RRI) of adjacent QRS intervals. The CPU 1 also detects P waves from one beat of electrocardiogram data and calculates feature quantities related to the P waves. The CPU 1 stores the calculated feature quantity data in the storage device 10, for example, for each lead.
[0059] In S507, the CPU 1 executes an automatic classification process to assign a classification to each heartbeat. In the automatic classification process, the CPU 1 can assign a classification to each heartbeat based on the feature amount calculated for the QRS interval, the shape of the QRT interval, predetermined criteria, etc.
[0060] In S509, the CPU 1 outputs a P-wave classification screen to the display unit 6. As described above, the P-wave classification screen may be called from the waveform display screen that displays a list of representative waveforms of heartbeats classified by the automatic classification process. When the P-wave classification screen is displayed, it is assumed that whether a histogram or an XY plot is to be displayed, which P-wave feature value is to be used, and so on are set in advance.
[0061] In S511, the CPU 1 determines whether or not an operation on the P-wave classification screen has been detected, and if it is determined that an operation has been detected, executes S513, and if not, executes S511 again.
[0062] In S513, the CPU 1 determines whether the detected operation is a range specification operation for the histogram 310 or the XY plot 410. The range specification operation may be, for example, a mouse drag or an equivalent touch operation, or any range specification operation using a keyboard. If it is determined that a range specification operation has been detected, the CPU 1 executes S515; if not, the CPU 1 executes S517.
[0063] In S515, the CPU 1 displays overlapping waveforms for heartbeats that correspond to the specified range (selected heartbeats). As described above, the CPU 1 can also display overlapping waveforms for heartbeats that do not correspond to the specified range (non-selected heartbeats) or that are not included in the histogram 310 or XY plot 410. After displaying overlapping waveforms, the CPU 1 executes S511.
[0064] In S517, the CPU 1 determines whether the detected operation is a category designation operation. The category designation operation may be, for example, a selection operation on the judgment code change menu 322 described above. If it is determined that a category designation operation has been detected, the CPU 1 executes S519. If another operation, such as an operation to change the P-wave feature amount, an operation to switch between the histogram 310 and the XY plot 410, or an operation for displaying overlapping waveforms, is detected, the CPU 1 executes an operation corresponding to the operation and then executes S511 again.
[0065] In S519, the CPU 1 collectively assigns the classification specified by the classification specifying operation to the heart rate group that is the target of the classification specifying operation. The heart rate group that is the target of the classification specifying operation is the heart rate group whose waveform is displayed in the superimposed display of the waveform that is the target of the classification specifying operation. Then, the CPU 1 executes S511 again. Alternatively, the CPU 1 ends the display of the P wave classification screen and executes an editing operation, for example, via the waveform list screen.
[0066] As described above, the electrocardiogram analysis device of this embodiment has a function of presenting a classification screen for classifying heartbeats according to P-wave feature amounts and collectively assigning classifications to heartbeat groups selected by the user from the classification screen. This effectively supports the task of assigning supraventricular arrhythmia classifications to heartbeats, which could not be assigned using the automatic classification function based on QRS intervals.
[0067] (Other embodiments) The present invention can also be implemented as a program that causes a computer to function as the electrocardiogram analyzer described in the above embodiment. Furthermore, the present invention is not limited to the content of the above embodiment, and various changes and modifications can be made without departing from the spirit and scope of the invention. Therefore, the following claims are appended to clarify the scope of the invention. [Explanation of symbols]
[0068] 100... general-purpose computer (electrocardiogram analyzer), 1... CPU, 4... memory card, 10... storage device, 200... external device
Claims
1. an acquisition means for acquiring electrocardiogram data; a calculation means for calculating a feature value of a P wave for each heartbeat of the electrocardiogram data; an output means for outputting a classification screen that classifies and displays the heartbeats of the electrocardiogram data based on the feature amount of the P wave; an editing means for providing a function of collectively assigning a classification designated by a user to at least one of the heartbeats selected and the heartbeats not selected on the classification screen, An electrocardiogram analysis device characterized in that the classification screen has an area for displaying overlapping waveforms for multiple selected heartbeats among the classified and displayed heartbeats, and an area for displaying overlapping waveforms for multiple non-selected heartbeats.
2. 2. The electrocardiogram analyzer according to claim 1, wherein the classification screen includes a histogram relating to the feature amount of the P wave.
3. 2. The electrocardiogram analysis device according to claim 1, wherein the classification screen includes an area in which individual heartbeats are plotted within a two-dimensional area in which a first feature of the P wave is assigned to a first axis and a second feature of the P wave is assigned to a second axis.
4. The classification screen includes: a histogram relating to the feature amount of the P wave; a region in which individual heartbeats are plotted within a two-dimensional region in which a first feature value of the P wave is assigned to a first axis and a second feature value of the P wave is assigned to a second axis; 2. The electrocardiogram analyzer according to claim 1, wherein the display is switchable.
5. The electrocardiogram analysis device according to any one of claims 1 to 4, characterized in that the editing means provides a function of collectively assigning a classification specified by the user to heartbeats that are not displayed on the classification screen.
6. 6. The electrocardiogram analysis device according to claim 1, wherein at least one of the waveforms of the selected plurality of heartbeats and the waveforms of the non-selected plurality of heartbeats are displayed superimposed with their horizontal positions aligned.
7. the classification screen has an area for displaying waveforms for heartbeats that are not displayed on the classification screen, 7. The electrocardiogram analyzer according to claim 6, wherein waveforms for a plurality of heartbeats that are not displayed on the classification screen are displayed in an overlapping manner with their positions aligned in the horizontal direction.
8. 8. The electrocardiogram analyzer according to claim 6, wherein the display color of the coordinates constituting the overlapped display of the waveforms varies depending on the frequency of the waveforms overlapping the coordinates.
9. 9. The electrocardiogram analysis device according to claim 6, wherein a heartbeat corresponding to a waveform selected in an area where the waveforms of the plurality of heartbeats are displayed in an overlapping manner is excluded from targets for which a classification is assigned collectively.
10. 10. The electrocardiogram analyzer according to claim 1, wherein a classification relating to supraventricular arrhythmia can be designated as the classification.
11. 11. The electrocardiogram analysis device according to claim 1, wherein the P wave feature quantity includes one or more of a PR interval, a PQ interval, a P wave width, a P wave height, and a P wave polarity.
12. an acquisition step of acquiring electrocardiogram data; a calculation step of calculating a feature value of a P wave for each heartbeat of the electrocardiogram data; an output step of outputting a classification screen that classifies and displays the heartbeats of the electrocardiogram data based on the feature amount of the P wave; an editing step of providing a function of collectively assigning a classification designated by a user to at least one of the heartbeats selected and the heartbeats not selected on the classification screen; and A method for controlling an electrocardiogram analysis device, characterized in that the classification screen has an area for displaying overlapping waveforms for multiple selected heartbeats from the classified and displayed heartbeats, and an area for displaying overlapping waveforms for multiple heartbeats not selected.
13. A program for causing a computer to function as each of the means included in the electrocardiogram analyzer according to any one of claims 1 to 11.
Citation Information
Patent Citations
Arrhythmia classification system and method using deep neural network
CN111281371A
Device for classifying electrocardiogram
JP2006116207A
Heart-rate variability analysis method and analysis device
JP2009095552A
Biological information processing device, biological information processing method and program
JP2019097639A
Time feature value calculation device, calculation method and its program
JP2020130335A