KR102585080B1 - Apparatus for Providing Annotation Tool for Classifying ECG Wave - Google Patents

Abstract

Disclosed is an annotation tool providing device for classifying electrocardiogram waveforms.
This embodiment provides a tool for generating raw data for deep learning or AI learning, processing ECG waveform data that is different for each person, and enabling ECG waveforms to set modes and distinguish colors for specific sections based on medical judgment. Provides an annotation tool providing device for classifying.

Description

Apparatus for providing annotation tool for classifying ECG waveform {Apparatus for Providing Annotation Tool for Classifying ECG Wave}

One embodiment of the present invention relates to an annotation tool providing device for classifying electrocardiogram waveforms.

The content described below simply provides background information related to this embodiment and does not constitute prior art.

An electrocardiogram reading system has been developed to help medical staff analyze electrocardiograms. Conventional electrocardiogram reading systems detect the R, P, and T peaks of the waveform, and detect and classify arrhythmia based on rules.

A conventional electrocardiogram reading system receives the patient's entire electrocardiogram signal data, analyzes it, and outputs the results. Deep learning technology has recently been widely studied as an electrocardiogram reading algorithm because it has higher accuracy than existing methods.

Only qualified medical staff can determine arrhythmia using an electrocardiogram, but the reality is that there is a shortage of manpower compared to demand. When reading an ECG, it takes a lot of time because the ECG signal must be read from various perspectives, such as calculating the time difference between the shapes and sections of the P, QRS, and T waveforms, and analyzing the ECG rhythm. Medical staff must observe the patient's electrocardiogram in real time to keep an eye on the patient's condition, but continuous monitoring is difficult due to a lack of manpower. Because electrocardiogram analysis is directly related to the patient's life, it must be accurate and operate quickly in the event of an emergency.

Conventional electrocardiogram analysis sometimes uses the end and start points of the P, QRS, and T waveform sections, but the conventional technology only searches for peaks, which reduces its utility. When detecting and classifying arrhythmia, rule-based algorithm design is less accurate due to the diversity of waveforms, and when arrhythmia is added, a new rule-based algorithm must be designed.

Conventional electrocardiogram reading systems cannot be used in places where real-time reading is required, such as bedside patient monitoring. Deep learning algorithms for ECG analysis have different speeds depending on the implementation model, and cannot operate every hour for real-time operation. When visualizing ECG waveforms, one-dimensional data is output, which reduces readability during real-time reading.

Additionally, the conventional ECG reading system does not have an annotation tool that allows easy classification of one-dimensional ECG signals to secure data.

Therefore, it is important to have a way to visually display information about the section that the user wants to classify, and it is necessary to apply a graph directly to the classified data for confirmation. In addition, so that users can use it intuitively, it is necessary to select waveform classification through key input and accurately set the section through mouse input.
Prior art literature (patent document 1) KR 10-2322234 B1

This embodiment provides a tool for generating raw data for deep learning or AI learning, processing ECG waveform data that is different for each person, and enabling ECG waveforms to set modes and distinguish colors for specific sections based on medical judgment. The purpose is to provide an annotation tool provider for classifying.

According to one aspect of this embodiment, an electrocardiogram waveform acquisition unit that acquires electrocardiogram waveforms for a plurality of people; An ECG waveform display unit that images the ECG waveform and displays it on a screen; a mode conversion unit that converts to one of a plurality of modes depending on whether there is an input from the first input unit; a section setting unit that matches a specific section of the ECG waveform corresponding to an input from a second input unit to the ECG waveform displayed on the screen when converted to the mode; An ECG annotation providing device is provided, comprising an annotation unit that generates ECG annotation data displayed in a preset color corresponding to the mode for the specific section of the ECG waveform.

As described above, according to this embodiment, a tool for generating raw data for deep learning or AI learning is provided to process different ECG waveform data for each person, set a mode for each specific section based on medical judgment, and color This has the effect of classifying the ECG waveform.

According to this embodiment, the classified waveform can be applied to learning data in deep learning and learning algorithms, and using an annotation tool, ECG experts can easily classify specific waveforms, allowing a lot of data to be secured in a short time. It works.

1A and 1B are diagrams showing a waveform classification annotation tool screen according to this embodiment.
Figure 2 is a diagram showing an annotation tool providing device according to this embodiment.
Figure 3 is a diagram showing a color table according to the graph mode according to this embodiment.
Figure 4 is a diagram showing keyboard input and color according to the waveform according to this embodiment.
Figure 5 is a diagram showing an example guide box (PVC-QRS) according to waveform selection according to this embodiment.
Figure 6 is a diagram showing an example section selection guide box (Normal QRS) according to this embodiment.
Figure 7 is a diagram illustrating an example of a graph representation of the section selection result (Normal QRS) according to this embodiment.
Figure 8 is a diagram showing the ECG waveform classification annotation tool GUI according to this embodiment.
Figure 9 is a diagram showing an example of ECG waveform classification selection according to this embodiment.

Hereinafter, this embodiment will be described in detail with reference to the attached drawings.

Figure 1 is a diagram showing a waveform classification annotation tool screen according to this embodiment.

The ECG annotation providing device 200 according to this embodiment uses an annotation tool to configure and output a single graph screen for user readability. The ECG annotation providing device 200 uses an annotation tool to express both waveform classification operations and results within a graph screen. The ECG annotation providing device 200 can be used not only for ECG waveforms but also for classifying important sections of other one-dimensional signals.

The ECG annotation providing device 200 is one-dimensional in the ECG waveform and can classify sections with specific meaning based on the shape of the signal and medical judgment. The ECG annotation providing device 200 can classify sections of the ECG waveform into P waveform, T waveform, QRS waveform (Normal, PVC, PAC), etc., and each section can be a standard for medical judgment. The ECG annotation providing device 200 can accurately classify ECG waveforms based on a recent data learning-based algorithm.

The ECG annotation providing device 200 provides an annotation tool that allows classification of one-dimensional ECG waveforms using an intuitive, user-friendly interface.

The ECG annotation providing device 200 provides tools for performing annotation. When the ECG annotation providing device 200 classifies a special section of the cardiac ECG waveform using deep learning or machine learning, it displays the waveforms in different colors.

The ECG annotation providing device 200 provides a tool that allows each ECG waveform to have a special section and perform annotation by dividing each section.

The heart is a pump that circulates blood throughout the body and regularly repeats contraction and expansion. The pumping action of the heart is achieved by contraction of the myocardium, and weak electricity is generated each time the heart beats. Electricity generated when the heart beats causes an electric current to flow within the body, and the electric current flowing within the body generates distribution of potential on the surface of the body.

An electrocardiogram (ECG) is an electrocardiogram (ECG) that induces, amplifies, and records small electrical potential changes resulting from heart activity in an appropriate area of the body surface using a certain method.

As shown in Figure 1b, the waveforms of the electrocardiogram are P wave (P-Wave), Q wave (Q-Wave), R wave (R-Wave), S wave (S-Wave), and T wave (T-Wave). ) includes. The P wave occurs during atrial depolarization, the QRS wave occurs during ventricular depolarization, and the T wave occurs during ventricular repolarization.

The P wave depolarization of the atrium begins near the sinoatrial crystal and proceeds from right to left across the atrium. The first part of the P wave represents depolarization of the right atrium. The latter part of the P wave represents depolarization of the left atrium. Normally, the P wave occurs during ventricular diastole. Generally, P waves are upward and have a slightly rounded shape.

The T wave represents normal repolarization of the ventricle. Normal repolarization begins at the epicardial surface of the ventricle and progresses through the ventricular wall toward the endocardium. The T wave occurs during the end of ventricular systole. The point where the T wave returns to the baseline is the end of the T wave, and the direction of the peak is upward. From QRS to T, the bedroom is in a relaxed state, so knowing the important sections can help you analyze your heartbeats.

Since ECG analysis data is different for each person, it is necessary to collect ECG analysis data. The ECG annotation providing device 200 provides an annotation tool that allows ECG waveform classification to be easily performed after quickly collecting ECG analysis data.

The ECG annotation providing device 200 sets each color for the P waveform, QRS waveform, and T waveform using an annotation tool. The ECG annotation providing device 200 changes to the corresponding mode when there is input from the keyboard using an annotation tool.

If there is an input for P or p from the keyboard using an annotation tool, the ECG annotation providing device 200 selects the P waveform and then matches the information input through the mouse event to the section corresponding to the P waveform to create a specific section (P waveform).

In other words, when there is an input for P or p from the keyboard using an annotation tool, the ECG annotation providing device 200 defines a section corresponding to the P waveform among the entire ECG waveform.

The ECG annotation providing device 200 uses an annotation tool to define a specific section of the ECG waveform in a mode corresponding to input from the keyboard, then inputs the data into deep learning and outputs the learned results.

The ECG annotation providing device 200 uses an annotation tool to define a section corresponding to the P waveform, a section corresponding to the QRS wave, and a section corresponding to the T waveform in a plurality of ECG waveforms, and then generates an ECG learning result by learning these. . The ECG annotation providing device 200 automatically extracts a section corresponding to a specific section (P waveform, QRS waveform, T waveform) from the ECG waveform using the ECG learning results.

Figure 2 is a diagram showing an annotation tool providing device according to this embodiment.

The ECG annotation providing device 200 according to this embodiment includes an ECG waveform acquisition unit 210, an ECG waveform display unit 220, a mode conversion unit 230, a section setting unit 240, an annotation unit 250, and a learning unit. Includes (260). Components included in the ECG annotation providing device 200 are not necessarily limited to this.

Each component included in the ECG annotation providing device 200 is connected to a communication path connecting software modules or hardware modules within the device and can operate organically with each other. These components communicate using one or more communication buses or signal lines.

Each component of the ECG annotation providing device 200 shown in FIG. 2 refers to a unit that processes at least one function or operation, and may be implemented as a software module, a hardware module, or a combination of software and hardware.

The ECG waveform acquisition unit 210 acquires ECG waveforms for multiple people. The ECG waveform display unit 220 images the ECG waveform and displays it on the screen.

The mode conversion unit 230 converts the mode to one of a plurality of modes depending on whether there is an input from the keyboard (first input unit).

The mode conversion unit 230 converts to the P waveform setting mode when the preset first key (‘P’ or ‘p’) is input from the keyboard (first input unit). The mode conversion unit 230 converts to the T waveform setting mode when a preset second key (‘T’ or ‘t’) is input from the keyboard (first input unit).

The mode conversion unit 230 converts to the normal QRS waveform setting mode when a preset third key (‘N’ or ‘n’) is input from the keyboard (first input unit). The mode conversion unit 230 converts the mode to the PAC QRS waveform setting mode when the preset fourth key (‘S’ or ‘s’) is input from the keyboard (first input unit).

The mode conversion unit 230 converts the mode to the PVC QRS waveform setting mode when the preset fifth key (‘V’ or ‘v’) is input from the keyboard (first input unit). The mode conversion unit 230 converts to the noise waveform setting mode when the preset sixth key (‘W’ or ‘w’) is input from the keyboard (first input unit). The mode conversion unit 230 converts to the base waveform setting mode when the preset seventh key (‘B’ or ‘b’) is input from the keyboard (first input unit).

The mode conversion unit 230 repeats the process of setting the specific section for the next mode when the specific section for one of the plurality of modes is set and ECG annotation data is generated.

In the mode conversion state, the section setting unit 240 matches a specific section corresponding to the input of the mouse (second input unit) among the ECG waveforms to the ECG waveform displayed on the screen.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the P waveform setting mode. The section setting unit 240 sets the P waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the T waveform setting mode. The section setting unit 240 sets the T waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the normal QRS waveform setting mode. The section setting unit 240 sets the normal QRS waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the PAC QRS waveform setting mode. The section setting unit 240 sets the PAC QRS waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the PVC QRS waveform setting mode. The section setting unit 240 sets the PVC QRS waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the noise waveform setting mode. The section setting unit 240 sets the noise waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

The section setting unit 240 sets the starting point with the first click of the mouse (second input unit) input while converted to the base waveform setting mode. The section setting unit 240 sets the base waveform section, the end point of which is set by the second click of the mouse (second input unit), by matching it with the ECG waveform.

When the first click occurs by inputting a mouse (second input unit) on the ECG waveform, the section setting unit 240 sets the corresponding point as the first starting point of a specific section and then generates a second click by inputting a mouse (second input unit). A guide in the form of a box matching the color of the waveform is displayed on the ECG waveform so that the operator can recognize the section setting in advance.

The section setting unit 240 sets the end point of a specific section at the corresponding point when the second click occurs by inputting the mouse (second input unit) on the ECG waveform, and when setting the specific section is completed, the color corresponding to the mode is displayed on the graph screen. It should be displayed immediately on the screen so that the operator can check it.

The annotation unit 250 generates ECG annotation data displayed in a preset color corresponding to the mode for a specific section of the ECG waveform.

The annotation unit 250 generates ECG annotation data displayed in the P waveform section in a preset P waveform color (light blue) corresponding to the P waveform setting mode. The annotation unit 250 generates ECG annotation data displayed in the T waveform section in a preset T waveform color (purple) corresponding to the T waveform setting mode.

The annotation unit 250 generates ECG annotation data displayed in the normal QRS waveform section in a preset normal QRS waveform color (green) corresponding to the normal QRS waveform setting mode.

The annotation unit 250 generates ECG annotation data displayed in the PAC QRS waveform section in a preset PAC QRS waveform color (blue) corresponding to the PAC QRS waveform setting mode. The annotation unit 250 generates ECG annotation data displayed in the PVC QRS waveform section in a preset PVC QRS waveform color (red) corresponding to the PVC QRS waveform setting mode.

The annotation unit 250 generates ECG annotation data displayed in a preset noise waveform color (gray) corresponding to the noise waveform setting mode in the noise waveform section. The annotation unit 250 generates ECG annotation data displayed in the base waveform section in a preset base waveform color (graph line color) corresponding to the base waveform setting mode.

The learning unit 260 generates an ECG learning result by performing learning by inputting ECG annotation data generated by a plurality of modes for each of a plurality of people.

Figure 3 is a diagram showing a color table according to the graph mode according to this embodiment.

Since the preferred graph color may vary depending on the user, the ECG annotation providing device 200 can provide a total of three graph colors as shown in FIG. 3.

Figure 4 is a diagram showing keyboard input and color according to the waveform according to this embodiment.

The ECG annotation providing device 200 uses the letters (P, p, T, t, N, n, S, s, V, v, W) written on the table as shown in FIG. 4 so that waveforms can be selected by keyboard input. , w, B, b) are arranged so that medical staff can intuitively select them. As shown in FIG. 4, the ECG annotation providing device 200 sets different colors displayed on the graph screen depending on the waveform. The ECG annotation providing device 200 displays a guide box according to the selected waveform.

The ECG annotation providing device 200 provides a UI that is intuitive and easy to work with because it requires matching some sections of the ECG waveform with keyboard input and mouse event input using an annotation tool.

The ECG annotation providing device 200 determines whether there is input (P, p, T, t, N, n, S, s, V, v, W, w, B, b) from the keyboard using an annotation tool. Check .

If there is input from the keyboard using an annotation tool, the ECG annotation providing device 200 converts it into a mode (P waveform, T waveform, Normal QRS, PAC QRS, PVC QRS, Noise waveform, base waveform) corresponding to the keyboard input. do.

The ECG annotation providing device 200 converts to the ‘P waveform setting mode’ where ‘P’ or ‘p’ is input from the keyboard. The ECG annotation providing device 200 converts to the ‘T waveform setting mode’ in which ‘T’ or ‘t’ input exists from the keyboard. The ECG annotation providing device 200 converts to ‘normal QRS waveform setting mode’ in which ‘N’ or ‘n’ input is present from the keyboard. The ECG annotation providing device 200 converts to ‘PAC QRS waveform setting mode’ in which ‘S’ or ‘s’ input is input from the keyboard. The ECG annotation providing device 200 converts to ‘PVC QRS waveform setting mode’ in which ‘V’ or ‘v’ input is present from the keyboard. The ECG annotation providing device 200 converts to the ‘noise waveform setting mode’ where ‘W’ or ‘w’ is input from the keyboard. The ECG annotation providing device 200 converts to the ‘base waveform setting mode’ in which ‘B’ or ‘b’ input exists from the keyboard.

The ECG annotation providing device 200 uses an annotation tool to select modes corresponding to keyboard input (P waveform setting mode, T waveform setting mode, normal QRS waveform setting mode, PAC QRS waveform setting mode, PVC QRS waveform setting mode, noise waveform Setting mode, base waveform setting mode), the section corresponding to the mouse event in the ECG waveform (the first click sets the start point, the second click sets the end point) is changed to a specific section corresponding to that mode. Set as section.

When the ECG annotation providing device 200 sets a section corresponding to a mouse event (the first click sets the starting point and the second click sets the end point) using an annotation tool, it changes the color of the corresponding section. .

The ECG annotation providing device 200 changes to ‘P waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘light blue’. The ECG annotation providing device 200 changes to the ‘T waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘purple’.

The ECG annotation providing device 200 changes to ‘normal QRS waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘green’. The ECG annotation providing device 200 changes to ‘PAC QRS waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘blue’. The ECG annotation providing device 200 changes to ‘PVC QRS waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘red’.

The ECG annotation providing device 200 changes to ‘noise waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘gray’. The ECG annotation providing device 200 changes to the ‘base waveform setting mode’ using an annotation tool and changes the section corresponding to the mouse event in the ECG waveform to ‘graph line color’. Here, the base waveform refers to a general waveform.

The ECG annotation providing device 200 uses an annotation tool to configure a plurality of modes (P waveform setting mode, T waveform setting mode, normal QRS waveform setting mode, PAC QRS waveform setting mode, PVC QRS waveform setting mode, Repeat the process of specifying the start and end points for the first mode (noise waveform setting mode, base waveform setting mode) and then specifying the start and end points for the next mode.

Since the heartbeat is different for each person, the ECG annotation providing device 200 sets sections for each mode in the ECG waveform for each person and generates ECG annotation data with a color for each section.

The ECG annotation providing device 200 collects a plurality of ECG annotation data and performs learning.

Figure 5 is a diagram showing an example guide box (PVC-QRS) according to waveform selection according to this embodiment.

The ECG annotation providing device 200 sets the first starting point of the section when the first click occurs on the ECG waveform using a click-and-click method using a mouse input. The ECG annotation providing device 200 sets the end point of the section when the second click occurs on the ECG waveform. The ECG annotation providing device 200 displays a box-type guide according to the color of the waveform on the graph screen in order for the operator to recognize in advance that the section is set between clicks of the section.

The ECG annotation providing device 200 collects different ECG waveforms for each person. The ECG annotation providing device 200 images the ECG waveform and displays it on the screen.

The ECG annotation providing device 200 changes modes using a keyboard, and for each mode, uses a mouse to define specific sections on the input ECG waveform and generate stored ECG annotation data by matching it with the ECG waveform.

The ECG annotation providing device 200 generates ECG annotation data in a stacked structure in which specific sections are matched to ECG waveforms for each mode. The ECG annotation providing device 200 can learn ECG annotation data using deep learning.

The ECG annotation providing device 200 creates ECG annotation data by imaging the ECG waveform and storing a type of metadata defining specific sections for each mode on the ECG waveform in time series. The ECG annotation providing device 200 stores ECG annotation data in each label according to a preset timing.

Figure 6 is a diagram showing an example section selection guide box (Normal QRS) according to this embodiment.

When a second click occurs on the ECG waveform and the setting for the corresponding section is completed, the ECG annotation providing device 200 immediately reflects the color of the specified waveform on the graph screen so that the user can check it.

The ECG annotation providing device 200 classifies the effects on the ventricles and parts other than the ventricles into S or V and sets them to PAC or PVC mode according to the operator's operation or command. The ECG annotation providing device 200 can distinguish between the QRS waveform and the T waveform as S when the shape of the QRS waveform and the T waveform are changed by the operator's manipulation or command.

The ECG annotation providing device 200 is preferably applied to ECG waveforms, but is not necessarily limited thereto, and can be applied as a tool that can distinguish specific waveforms from other biometric data or one-dimensional time series data.

The ECG annotation providing device 200 can use ECG annotation data as correct answer data when using it as raw data for deep learning or AI learning.

The ECG annotation providing device 200 generates ECG annotation data and later inputs the ECG annotation data into a learning model to learn it. This can automatically set a specific section based on medical judgment in the ECG waveform and display a color. .

In order to input the ECG annotation data as raw data for deep learning or AI learning, the ECG annotation providing device 200 selects data with noise (outside the threshold), determines it as noise data with low value as sample data, and filters it. do.

Figure 7 is a diagram illustrating an example of a graph representation of the section selection result (Normal QRS) according to this embodiment.

The ECG annotation providing device 200 can be used not only for ECG waveforms but also for classifying important sections of other one-dimensional signals.

Figure 8 is a diagram showing the ECG waveform classification annotation tool GUI according to this embodiment.

The ECG annotation providing device 200 acquires ECG waveforms for a plurality of people, images the ECG waveforms, and outputs them on the screen. The ECG annotation providing device 200 outputs a list of ECG waveforms for a plurality of people in a GUI format. The ECG annotation providing device 200 outputs the ECG waveform selected from the ECG waveform list on the screen.

Figure 9 is a diagram showing an example of ECG waveform classification selection according to this embodiment.

The ECG annotation providing device 200 outputs the ECG waveform on the screen.

As shown in (a) of FIG. 9, the ECG annotation providing device 200 selects one of a plurality of modes (P waveform, T waveform, Normal QRS, PAC QRS) according to the input from the keyboard (first input unit). , PVC QRS, Noise waveform, and Base waveform).

As shown in (b) of FIG. 9, the ECG annotation providing device 200 changes the mode according to the input from the keyboard (first input unit) and with the first click of the input from the mouse (second input unit) Set the starting point.

As shown in (c) of FIG. 9, the ECG annotation providing device 200 sets the starting point with the first click and outputs the selection section while moving the mouse (second input unit). The ECG annotation providing device 200 specifies a waveform section.

As shown in (d) of FIG. 9, the ECG annotation providing device 200 sets the end point with a second click of the mouse (second input unit) and outputs the selected waveform in a preset color. The ECG annotation providing device 200 stores the selected waveform and outputs the screen.

The above description is merely an illustrative explanation of the technical idea of the present embodiment, and those skilled in the art will be able to make various modifications and variations without departing from the essential characteristics of the present embodiment. Accordingly, the present embodiments are not intended to limit the technical idea of the present embodiment, but rather to explain it, and the scope of the technical idea of the present embodiment is not limited by these examples. The scope of protection of this embodiment should be interpreted in accordance with the claims below, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of rights of this embodiment.

200: ECG annotation providing device
210: ECG waveform acquisition unit
220: ECG waveform display unit
230: Mode conversion unit
240: Section setting unit
250: Annotation part
260: Learning Department

Claims (12)

An electrocardiogram waveform acquisition unit that acquires electrocardiogram waveforms for a plurality of people;
An ECG waveform display unit that images the ECG waveform and displays it on a screen;
a mode conversion unit that converts to one of a plurality of modes depending on whether there is an input from the first input unit;
a section setting unit that matches a specific section of the ECG waveform corresponding to an input from a second input unit to the ECG waveform displayed on the screen when converted to the mode;
An annotation unit that generates ECG annotation data displayed in a preset color corresponding to the mode for the specific section of the ECG waveform.
The mode conversion unit,
When the specific section for one of the plurality of modes is set and the ECG annotation data is generated, the setting process for the specific section for the next mode is repeated. According to paragraph 1,
The mode conversion unit converts to the P waveform setting mode when a preset first key is input from the first input unit,
In the state in which the section setting unit is converted to the P waveform setting mode, the P waveform section is set by the first click of the input of the second input unit and the end point by the second click of the second input unit. It is set by matching the ECG waveform,
The annotation unit generates the ECG annotation data displayed in a preset P waveform color corresponding to the P waveform setting mode in the P waveform section. According to paragraph 1,
The mode conversion unit converts to the T waveform setting mode when a preset second key is input from the first input unit,
In the state in which the section setting unit is converted to the T waveform setting mode, the T waveform section is set by the first click of the input of the second input unit and the end point by the second click of the second input unit. It is set by matching the ECG waveform,
The annotation unit generates the ECG annotation data displayed in the T waveform section in a preset T waveform color corresponding to the T waveform setting mode. According to paragraph 1,
The mode converter converts the mode to a normal QRS waveform setting mode when a preset third key is input from the first input unit,
In a state where the section setting unit is converted to the normal QRS waveform setting mode, the starting point is set with the first click of the input of the second input unit, and the normal QRS waveform section is set with the ending point being set with the second click of the second input unit. It is set by matching the electrocardiogram waveform,
The annotation unit generates the ECG annotation data displayed in the normal QRS waveform section in a preset normal QRS waveform color corresponding to the normal QRS waveform setting mode. According to paragraph 1,
The mode conversion unit converts to the PAC QRS waveform setting mode when a preset fourth key is input from the first input unit,
In a state in which the section setting unit is converted to the PAC QRS waveform setting mode, the PAC QRS waveform section sets the starting point with the first click of the input of the second input unit and sets the ending point with the second click of the second input unit. is set by matching the ECG waveform,
The annotation unit generates the ECG annotation data displayed in the PAC QRS waveform section in a preset PAC QRS waveform color corresponding to the PAC QRS waveform setting mode. According to paragraph 1,
The mode converter converts the PVC QRS waveform setting mode when a preset fifth key is input from the first input unit,
The section setting unit sets the starting point with the first click of the input of the second input unit in a state converted to the PVC QRS waveform setting mode, and sets the end point with the second click of the second input unit. is set by matching the ECG waveform,
The annotation unit generates the ECG annotation data displayed in the PVC QRS waveform section in a preset PVC QRS waveform color corresponding to the PVC QRS waveform setting mode. According to paragraph 1,
The mode conversion unit converts to the noise waveform setting mode when a preset sixth key is input from the first input unit,
The section setting unit sets the starting point with the first click of the input of the second input unit in a state converted to the noise waveform setting mode, and sets the noise waveform section with the ending point with the second click of the second input unit. It is set by matching the ECG waveform,
The annotation unit generates the ECG annotation data displayed in the noise waveform section in a preset noise waveform color corresponding to the noise waveform setting mode. According to paragraph 1,
The mode conversion unit converts to the base waveform setting mode when a preset seventh key is input from the first input unit,
In the state where the section setting unit is converted to the base waveform setting mode, the base waveform section is set by the first click of the input of the second input unit, and the end point is set by the second click of the second input unit. It is set by matching the ECG waveform,
The annotation unit generates the ECG annotation data displayed in the base waveform section in a preset base waveform color corresponding to the base waveform setting mode. delete According to paragraph 1,
A learning unit that generates an ECG learning result by performing learning by inputting the ECG annotation data generated for each of the plurality of modes for each of the plurality of people.
An electrocardiogram annotation providing device further comprising: According to paragraph 1,
The section setting section
When the first click occurs on the ECG waveform by the input of the second input unit, the corresponding point is set as the first starting point of the specific section, and then the operator is made to recognize the section setting in advance before the second click occurs by the input of the second input unit. An electrocardiogram annotation providing device, characterized in that it displays a box-type guide matched to the color of the waveform on the electrocardiogram waveform. According to paragraph 1,
The section setting unit,
When a second click occurs due to an input from the second input unit on the ECG waveform, the corresponding point is set as the end point of the specific section so that when the specific section is set, the color corresponding to the mode is immediately displayed on the graph screen. An ECG annotation providing device that allows the operator to check.