A data training system and method for implantable pulse generator

Through the data training system of the implanted pulse generator, the data reading and training module and the combination of preset parameters are compared, and the accuracy of determination of polarization potential or ER wave amplitude is solved, and automatic beater is improved Threshold captures the reliability of the function and reduces the computational volume and power consumption of the pacemaker.

CN113856046BActive Publication Date: 2024-11-22SHAANXI QINMING MEDICAL CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111235072.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-10-22
Publication Date
2024-11-22
Estimated Expiration
2041-10-22

AI Technical Summary

Technical Problem

The prior art has low reliability in the determination of acquisition and loss of acquisition by the size of the ER wave area or slope, resulting in the accuracy of the automatic pacing threshold acquisition function. Decreased sexually.

Method used

Provide a data training system for implantable pulse generator, including a data reading module, a data training module and a storage module, by reading the electrocardiogram data of the body surface and the electrocardiogram data in the cavity, and after preprocessing and preset parameters. The combination is used to compare, determine the capture and loss capture status, and obtain the accuracy based on the comparison results, and store and filter out the optimal preset parameter combination to improve the judgment accuracy.

Benefits of technology

Effectively improve the reliability of the automatic pacing threshold acquisition function, allowing more users to benefit from the automatic pacing threshold acquisition function, and reduce the calculation amount of the pacemaker by adjusting the preset parameter combination, reduce power consumption, and extend Service life.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113856046B_ABST
    Figure CN113856046B_ABST
Patent Text Reader

Abstract

The present application relates to the field of medical device technology, and in particular to a data training system and method for an implantable pulse generator, the system comprising a data reading module, a data training module and a storage module electrically connected in sequence, the method can firstly obtain surface electrocardiogram data and intracavitary electrocardiogram data through the system and perform preprocessing to obtain a basis for determination; then compare the intracavitary electrocardiogram data with a preset parameter combination, obtain a comparison result of the capture and loss of capture state of the intracavitary electrocardiogram data by the preset parameter combination, and then obtain the capture accuracy and loss of capture accuracy of the preset parameter combination; finally determine the optimal preset parameter combination. The system and method can effectively improve the reliability of the automatic pacing threshold capture function of the implantable pulse generator, so that more users can benefit from the function, and can reduce the amount of calculation of the pacemaker by changing the specific value of the preset parameter combination, thereby effectively reducing the power consumption of the pacemaker and extending its service life.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of medical device technology, and in particular to a data training system and method for an implantable pulse generator. Background Art

[0002] The pacing threshold capture function is a common function of automated implantable pulse generators, which includes two testing methods: one is the manual pacing threshold capture function, which requires the surface electrocardiogram to determine the capture and loss of pacing pulses under the supervision of a doctor, so it takes up limited medical resources; the other is the automatic pacing threshold capture function, which can directly determine the capture and loss of pacing pulses through intracardiac electrocardiogram data by the pulse generator when no one is monitoring, and adjust the pacing output amplitude of the atrium or ventricle based on the obtained results, so as to ensure that the pacing pulse can effectively capture the myocardium. The automatic ventricular pacing threshold capture function of some existing pulse generators has the function of beat-by-beat detection. The pulse generator can not only search for ventricular pacing thresholds at preset times, but also monitor the capture status of each ventricular pacing pulse at all times. Once loss of capture is detected, the pulse generator can immediately start the ventricular pacing threshold search and adjust the ventricular pacing output amplitude. Therefore, the automatic pacing threshold capture function has gradually become an indispensable function of implantable pulse generators.

[0003] The intracavitary electrocardiogram data generated after the pacing pulse is called the ER wave. The implantable pulse generator can collect the ER wave through the electrode wire, and determine the capture and loss of capture through the characteristic values ​​such as the ER wave area or slope. This method is the most commonly used method in various existing technologies to achieve the automatic pacing threshold capture function. The acquisition window of the ER wave is generally in the range of 0 to 100ms after the pacing pulse. However, due to the existence of the polarization potential, the ER wave morphology will be severely disturbed and significantly deformed in the range of 0 to 10ms. Therefore, the acquisition window of the ER wave is generally the interval of 10 to 100ms after the pacing pulse.

[0004] Figure 1 This is a schematic diagram of the morphology of the ER wave that is usually captured. Figure 1 As can be seen in the figure, there is an obvious polarization potential after the pacing pulse, which then quickly falls back to the baseline. Therefore, the loss of capture ER wave collected by the pulse generator is actually the polarization potential generated by the pacing pulse. Figure 2This is a schematic diagram of the captured ER wave under normal circumstances. Similarly, it can be seen that there is an obvious polarization potential after the pacing pulse, and there is an ER wave with large amplitude and width after the polarization potential area. Therefore, the captured ER wave collected by the pulse generator is actually the sum of the polarization potential and the ER wave. When the ER wave acquisition window is set to the interval of 10 to 100ms after the pacing pulse, under normal circumstances, the influence of the polarization electrode position on the ER wave can be filtered out. Therefore, the result of determining capture and loss of capture by characteristic values ​​such as the ER wave area or slope is reliable.

[0005] However, in actual applications, the width of the polarization potential and the shape of the ER wave are different for different users. When the polarization potential width is too large or the amplitude of the ER wave is too small, the accuracy of the existing technology in identifying capture and loss of capture will be significantly reduced, and even lead to errors in the test results of the automatic pacing threshold capture function. Figure 3 This is a schematic diagram of the morphology of the loss-of-capture ER wave collected when the polarization potential width is too large. It can be seen that there is an obvious polarization potential after the pacing pulse, and then it slowly falls back to the baseline. Therefore, after the polarization potential area, the loss-of-capture ER wave is still significantly affected by the polarization potential. Figure 4 This is a schematic diagram of the captured ER wave when the polarization potential width is too large. Similarly, it can be seen that there is an obvious polarization potential after the pacing pulse, and there is an ER wave with large amplitude and width after the polarization potential area. Therefore, after the polarization potential area, the captured ER wave will also be significantly affected by the polarization potential.

[0006] In summary, when the user has a polarization potential width that is too large or the amplitude of the ER wave is too small, the reliability of the capture and loss-of-capture judgment based on the size of the ER wave area or slope is relatively low. At this time, the automatic pacing threshold capture function has to be abandoned. Summary of the invention

[0007] The present application provides a data training system and method for an implantable pulse generator to solve the problem of low reliability of the traditional method of determining capture and loss of capture based on the size of the ER wave area or slope when the user has a polarization potential width that is too large or the amplitude of the ER wave is too small.

[0008] The technical solutions adopted by this application to solve the above technical problems are as follows:

[0009] In a first aspect, the present application provides a data training system for an implantable pulse generator, comprising a data reading module, a data training module and a storage module;

[0010] The data reading module is used to read the surface ECG data and the intracavitary ECG data, and pre-process the surface ECG data and the intracavitary ECG data to obtain the judgment basis;

[0011] The data training module is electrically connected to the data reading module, and is used to compare the intracavitary electrocardiogram data with a preset parameter combination, obtain a comparison result of the capture and loss of capture state of each intracavitary electrocardiogram data for each preset parameter combination, and obtain the capture accuracy rate and loss of capture accuracy rate of the preset parameter combination according to the determination basis and the comparison result;

[0012] The storage module is electrically connected to the data training module, and is used to store the capture accuracy and loss-capture accuracy of all the preset parameter combinations.

[0013] Further, the data reading module includes a data preprocessing unit and a first storage unit;

[0014] The data preprocessing unit includes a filter, an amplifier and an AD converter which are electrically connected in sequence, and is used to preprocess the surface ECG data and the intracavitary ECG data to obtain a determination basis;

[0015] The first storage unit is electrically connected to the data preprocessing unit, and is used to store intracavitary electrocardiogram data, surface electrocardiogram data and the determination basis.

[0016] Furthermore, the data training module includes a memory, a plurality of registers, a plurality of comparators and a plurality of counters;

[0017] The memory is used to store the preset parameter combination, wherein the preset parameter combination includes a preset fluctuation value, a preset collection area and a preset critical value;

[0018] The registers include a first register, a second register, a third register, a fourth register, a fifth register and a sixth register;

[0019] The first register is electrically connected to the first storage unit, and is used to read the body surface ECG data corresponding to each of the intracavitary ECG data from the first storage unit, and read the determination basis corresponding to the body surface ECG data, wherein the determination basis includes capture determination and loss of capture determination;

[0020] The second register, the third register and the fourth register are electrically connected to the memory respectively, and all read the preset parameter combination from the memory, wherein the second register stores the preset fluctuation value, the third register stores the preset acquisition area, and the fourth register stores the preset critical value;

[0021] The fifth register is electrically connected to the storage unit and is used to read each ER wave data, wherein the ER wave data is data of the intracavitary electrocardiogram data in a preset time period after each pacing pulse;

[0022] The sixth register is electrically connected to the first comparator and is used to store the accumulated calculated value of each data point calculated value of the ER wave data that meets the preset parameter combination;

[0023] The comparator comprises a first comparator, a second comparator and a third comparator;

[0024] The first comparator is electrically connected to the second register, the third register and the fifth register respectively, and is used to compare the ER wave data with the preset fluctuation value and the preset acquisition area;

[0025] The second comparator is electrically connected to the fourth register and the sixth register respectively, and is used to compare the accumulated calculation value with the preset critical value and output a corresponding comparison result, wherein the comparison result includes the recognition of capture and the recognition of loss of capture;

[0026] The third comparator is electrically connected to the first register and the second comparator respectively, and is used to compare the comparison result output by the second comparator with the determination basis output by the first register, and output a recognition result;

[0027] The counter comprises a first counter, a second counter, a third counter and a fourth counter which are electrically connected to the third comparator respectively;

[0028] The first counter is used to count once when the determination capture and the identification capture are simultaneously input to the input terminal of the third comparator;

[0029] The second counter is used to count once when the determination loss of capture and the identification loss of capture are simultaneously input to the input terminal of the third comparator;

[0030] The third counter is used to count once when the determination loss capture and the identification capture are simultaneously input to the input terminal of the third comparator;

[0031] The fourth counter is used for counting once when the capture determination and the loss of capture identification are input to the input terminal of the third comparator at the same time.

[0032] Furthermore, the storage module includes a storage unit, and the storage unit is used to store the capture accuracy and loss-capture accuracy of all the preset parameter combinations.

[0033] In a second aspect, the present application provides a data training method for an implantable pulse generator, comprising the following steps:

[0034] Obtain surface ECG data and intracavitary ECG data;

[0035] Establishing a basis for determining capture and loss of capture based on the surface electrocardiogram data;

[0036] Comparing the intracavitary electrocardiogram data with a preset parameter combination to obtain a comparison result of the capture and loss of capture state of each intracavitary electrocardiogram data for each preset parameter combination, wherein the preset parameter combination includes a preset fluctuation value, a preset acquisition area, and a preset critical value;

[0037] Comparing the comparison result with the determination basis to obtain the capture accuracy rate and the loss-capture accuracy rate of each of the preset parameter combinations;

[0038] An optimal preset parameter combination is determined, wherein the optimal preset parameter combination is determined according to the capture accuracy rate and the loss-capture accuracy rate of each of the preset parameter combinations.

[0039] Furthermore, the determining of the optimal preset parameter combination comprises the following steps:

[0040] Determining an accurate preset parameter combination, wherein the accurate preset parameter combination is the preset parameter combination in which the capture accuracy rate and the loss of capture accuracy rate are both 100%;

[0041] Determine a minimum acquisition area preset parameter combination, wherein the minimum acquisition area preset parameter combination is the accurate preset parameter combination with the smallest preset acquisition area;

[0042] An optimal preset parameter combination is determined, where the optimal preset parameter combination is the minimum acquisition area preset parameter combination in which the preset fluctuation value and the preset critical value are both centered.

[0043] Furthermore, the method comprises the following steps:

[0044] S1: The second register, the third register and the fourth register read the preset parameter combination from the memory respectively;

[0045] S2: The first register reads the surface ECG data corresponding to the intracavitary ECG data after a pacing pulse from the first storage unit, and inputs the corresponding determination basis to the second input terminal of the third comparator, wherein if the surface ECG data is determined to be captured, the second input terminal inputs a high level; if the surface ECG data is determined to be lost capture, the second input terminal inputs a low level;

[0046] S3: the fifth register reads the ER wave data after the pacing pulse of the intracavitary electrocardiogram data from the first storage unit;

[0047] S4: Acquire each data point of the ER wave data from the fifth register;

[0048] S5: If the data point is within the preset acquisition area stored in the third register, proceed to S6, otherwise proceed to S8;

[0049] S6: triggering the first comparator to work, if the data point is outside the fluctuation range defined by the preset fluctuation value stored in the second register, proceed to S7, otherwise proceed to S8;

[0050] S7: adding the calculated value of the data point to the calculated value of the previous data point to obtain an accumulated calculated value, and storing the accumulated calculated value in a sixth register;

[0051] S8: If the data point is not the last data point of the ER wave data defined by the preset acquisition area, proceed to S9, otherwise proceed to S10;

[0052] S9: Obtain the next data point of the ER wave data, and proceed to S5;

[0053] S10: If the accumulated calculation value stored in the sixth register is greater than the preset critical value stored in the fourth register, the second comparator inputs a high level to the first input terminal of the third comparator, indicating that the preset parameter combination identifies the state of the ER wave data as recognition capture; otherwise, the second comparator inputs a low level to the first input terminal of the third comparator, indicating that the preset parameter combination identifies the state of the ER wave data as recognition loss capture;

[0054] S11: If the first input terminal and the second input terminal of the third comparator are both at high level, the first counter counts once;

[0055] If the first input terminal and the second input terminal of the third comparator are both at low level, the second counter performs one counting;

[0056] If the first input terminal of the third comparator is at a high level and the second input terminal is at a low level, the third counter counts once;

[0057] If the first input terminal of the third comparator is at a low level and the second input terminal is at a high level, the fourth counter counts once;

[0058] S12: If the body surface ECG data in the first register is not the body surface ECG data corresponding to the intracavitary ECG data after the last pacing pulse in the first storage unit, proceed to S13; otherwise, proceed to S14;

[0059] S13: the first register reads the surface ECG data corresponding to the intracavitary ECG data after the next pacing pulse from the first storage unit, and performs S3;

[0060] S14: determining the capture accuracy rate and the loss-capture accuracy rate of the preset parameter combination according to the counting results of the first counter, the second counter, the third counter and the fourth counter, and saving the capture accuracy rate and the loss-capture accuracy rate of the preset parameter combination to a second storage unit;

[0061] S15: Clear the data in the sixth register, the first counter, the second counter, the third counter and the fourth counter;

[0062] S16: If the preset parameter combination is not the last preset parameter combination in the memory, proceed to S17, otherwise the data training ends;

[0063] S17: The second register, the third register and the fourth register read the next set of preset parameter combinations from the memory respectively, and perform S2.

[0064] Furthermore, the capture accuracy rate and the loss capture accuracy rate are calculated by the following formulas:

[0065]

[0066]

[0067] The technical solution provided by this application includes the following beneficial technical effects:

[0068] The present application provides a data training system and method for an implantable pulse generator, the system comprising a data reading module, a data training module and a storage module; the data reading module is used to read surface ECG data and intracavitary ECG data, and preprocess the surface ECG data and the intracavitary ECG data; the data training module is electrically connected to the data reading module, and is used to compare the intracavitary ECG data with a preset parameter combination, to obtain an identification result of the capture and loss of capture state of each intracavitary ECG data for each preset parameter combination, and to obtain the capture accuracy and loss of capture accuracy of the preset parameter combination according to the identification result; the storage module is electrically connected to the data training module, and is used to store the capture accuracy and loss of capture accuracy of all preset parameter combinations. The method includes acquiring surface ECG data and intracavitary ECG data; establishing a basis for determining capture and loss of capture based on the surface ECG data; comparing the intracavitary ECG data with a preset parameter combination to obtain a comparison result of the capture and loss of capture status of each intracavitary ECG data for each preset parameter combination, wherein the preset parameter combination includes a preset fluctuation value, a preset acquisition area, and a preset critical value; comparing the comparison result with the basis for determination to obtain the capture accuracy and loss of capture accuracy of each preset parameter combination; and determining the optimal preset parameter combination. The data training system and method for an implantable pulse generator provided in the present application can effectively improve the reliability of the automatic pacing threshold capture function, thereby enabling more users to benefit from the automatic pacing threshold capture function, and the system and method can achieve the purpose of reducing the amount of calculation of the pacemaker by changing the specific values ​​of the preset parameter combination, thereby effectively reducing the power consumption of the pacemaker and extending its service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a schematic diagram of the waveform of the pacing pulse failure under normal circumstances;

[0070] Figure 2 This is a schematic diagram of the waveform of the pacing pulse captured under normal circumstances;

[0071] Figure 3 It is a schematic diagram of the waveform of the pacing pulse when the polarization potential is wide;

[0072] Figure 4 It is a schematic diagram of the waveform captured by the pacing pulse when the polarization potential is wide;

[0073] Figure 5 A schematic diagram of an application scenario of a data training system for an implantable pulse generator provided in an embodiment of the present application;

[0074] Figure 6 A circuit block diagram of a data training system for an implantable pulse generator provided in an embodiment of the present application;

[0075] Figure 7A flow chart of a data training method for an implantable pulse generator provided in an embodiment of the present application;

[0076] Figure 8 A screening method for the optimal preset parameter combination provided in the embodiment of the present application;

[0077] Fig. 9 A schematic diagram of a reference standard for establishing capture and loss of capture states provided in an embodiment of the present application;

[0078] Fig.10 A schematic diagram showing that the reliability of the prior art in determining loss of capture is reduced when the polarization potential provided in the embodiment of the present application is wide;

[0079] Fig.11 A schematic diagram showing that the prior art determines that the capture reliability is reduced when the polarization potential provided in the embodiment of the present application is wide;

[0080] Fig.12 A schematic diagram showing that when the polarization potential provided in the embodiment of the present application is wide, the reliability of the present application in determining the loss of capture is high;

[0081] Fig.13 This is a schematic diagram showing that when the polarization potential provided in the embodiment of the present application is wide, the present application determines that the capture reliability is higher.

[0082] Explanation of the accompanying drawings: 1-program controller, 2-pulse generator, 3-program control head, 4-data training system, 100-data reading module, 101-data preprocessing unit, 102-first storage unit, 200-data training module, 201-memory, 202-first register, 203-second register, 204-third register, 205-fourth register, 206-fifth register, 207-first comparator, 208-sixth register, 209-second comparator, 210-third comparator, 211-first counter, 212-second counter, 213-third counter, 214-fourth counter, 300-storage module, 301-second storage unit. DETAILED DESCRIPTION

[0083] To facilitate the description and understanding of the technical solution of the application, the technical solution of the application is further explained below in conjunction with the accompanying drawings and embodiments.

[0084] The automatic pacing threshold capture function of the prior art is not applicable only to a small proportion of users with excessively large polarization potential width or small ER wave amplitude, but the base number of users is very large, for example, there are 1 million users, and the proportion of users for whom the automatic pacing threshold capture function is not applicable is 5%, then 50,000 people will not be able to benefit from the automatic pacing threshold capture function. In view of this situation, the embodiment of the present application provides a data training system for an implantable pulse generator.

[0085] See also Figure 5 , is a schematic diagram of an application scenario of a data training system for an implantable pulse generator provided in an embodiment of the present application. Figure 5 As shown in the figure, the basic operation process of the system is as follows:

[0086] In routine follow-up, surface ECG data and intracavitary ECG data were obtained through manual threshold test function;

[0087] Importing surface ECG data and intracavitary ECG data into the data training system;

[0088] The data training system extracts all ER wave data from the intracavitary electrocardiogram data. The ER wave data is the intracavitary electrocardiogram data 10 to 100 ms after each pacing pulse.

[0089] The data training system completes the ER wave data training for all preset parameter combinations with the above ER wave data as the object, thereby obtaining the capture accuracy and loss of capture accuracy of all preset parameter combinations;

[0090] Filter out an optimal preset parameter combination and feed it back to the program controller;

[0091] The optimal preset parameter combination is programmed for the user and the automatic pacing threshold function is turned on.

[0092] For details, see Figure 6 , is a circuit block diagram of a data training system for an implantable pulse generator provided in an embodiment of the present application. Figure 6 As shown, the circuit block diagram of the data training system mainly includes three basic modules: a data reading module 100, a data training module 200 and a storage module 300.

[0093] The data reading module 100 includes a data preprocessing unit 101 and a storage unit 102. The data preprocessing unit 101 includes a filter, an amplifier and an AD converter which are electrically connected in sequence. During routine follow-up of the user of the implantable pulse generator, the programmer can obtain the user's surface ECG data and intracavitary ECG data. After the data reading module 100 reads the surface ECG data and intracavitary ECG data from the programmer, the data preprocessing unit 101 preprocesses the original ECG data of the surface ECG data and the intracavitary ECG data to obtain a judgment basis, and saves the preprocessed original ECG data and the judgment basis to the first storage unit 102.

[0094] The data training module 200 is composed of a memory 201, a plurality of registers, a plurality of comparators and a plurality of counters. The memory 201 is used to store all preset parameter combinations including three parameters, wherein the three parameters are a preset acquisition area consisting of an acquisition start point and an acquisition end point, a preset fluctuation value and a preset critical value. The first register 202 is electrically connected to the first storage unit 102, and is used to read the surface ECG data corresponding to the intracavitary ECG data after each pacing pulse and the judgment basis from the first storage unit 102. The second register 203, the third register 204 and the fourth register 205 are electrically connected to the memory 201 respectively, and are used to read each preset parameter combination in the memory 201, wherein the second register 203 stores the preset fluctuation value, the third register 204 stores the preset acquisition area (acquisition start point and acquisition end point), and the fourth register 205 stores the preset critical value. The fifth register 206 is electrically connected to the first storage unit 102, and is used to read the ER wave data after each pacing pulse, wherein the ER wave data is the data of the intracavitary ECG data after each pacing pulse within a time period of 10 to 100 ms. The first comparator 207 is electrically connected to the second register 203, the third register 204 and the fifth register 206, respectively, and is used to compare the ER wave data with the preset fluctuation value and the preset acquisition area. The sixth register 208 is electrically connected to the first comparator 207, and is used to store the cumulative calculation value of the ER wave data that meets the preset parameter combination, and the cumulative calculation value is the sum of the calculation values ​​of each data point of the ER wave data that meets the preset parameter combination, and the calculation value can be the amplitude of each data point based on the baseline. The second comparator 209 is electrically connected to the fourth register 205 and the sixth register 208, respectively, and is used to compare the cumulative calculation value with the preset critical value, and output the corresponding comparison result, and the comparison result includes the recognition capture and the recognition loss capture. The third comparator 210 is electrically connected to the first register 202 and the second comparator 209, respectively, and is used to compare the comparison result output by the second comparator 209 with the judgment basis output by the first register 202, and output the recognition result. The counters are electrically connected to the third comparator 210, respectively, and are used to record the determination results of each preset parameter combination on the ER wave, wherein the first counter 211 is used to record the number of times a preset parameter combination correctly identifies a captured ER wave; the second counter 212 is used to record the number of times a preset parameter combination correctly identifies a lost-capture ER wave; the third counter 213 is used to record the number of times a preset parameter combination incorrectly identifies a lost-capture ER wave as a captured ER wave; and the fourth counter 214 is used to record the number of times a parameter combination incorrectly identifies a captured ER wave as a lost-capture ER wave.

[0095] The storage module 300 includes a storage unit 301 for storing the capture accuracy and the loss-capture accuracy of all preset parameter combinations.

[0096] See also Figure 7, is a flow chart of a data training method for an implantable pulse generator provided in an embodiment of the present application. Figure 7 As shown in , the main implementation process of this method is:

[0097] Acquire body surface electrocardiogram data and intracavitary electrocardiogram data, and pre-process the original body surface electrocardiogram data and intracavitary electrocardiogram data;

[0098] Establish the basis for determining capture and loss of capture based on surface ECG data;

[0099] Comparing the intracavitary electrocardiogram data (i.e., ER wave data) with a preset parameter combination, obtaining a comparison result of the capture and loss of capture state of each ER wave data for each preset parameter combination, wherein the preset parameter combination includes a preset fluctuation value, a preset acquisition area, and a preset critical value;

[0100] Comparing the comparison result with the determination basis to obtain a recognition result of each preset parameter combination, and determining the capture accuracy rate and the loss capture accuracy rate of each preset parameter combination according to the recognition result;

[0101] Determine the optimal preset parameter combination according to the capture accuracy and loss-of-capture accuracy of each preset parameter combination;

[0102] The optimal preset parameter combination is applied to the implantable pulse generator to enable the automatic pacing threshold function.

[0103] Specifically, the embodiments of the present application provide Figure 6 Based on the data training system for an implantable pulse generator shown in FIG. 1 , the data training method for an implantable pulse generator provided in an embodiment of the present application is as follows:

[0104] S1: The second register 203, the third register 204 and the fourth register 205 read the preset parameter combination from the memory 201 respectively;

[0105] S2: The first register 202 reads the surface ECG data corresponding to the intracavitary ECG data after a pacing pulse from the first storage unit 102, and inputs the corresponding determination basis to the second input terminal of the third comparator 210, wherein if the surface ECG data is determined to be captured, the second input terminal inputs a high level; if the surface ECG data is determined to be lost capture, the second input terminal inputs a low level;

[0106] S3: The fifth register 206 reads the ER wave data after the pacing pulse of the intracavitary electrocardiogram data from the first storage unit 102;

[0107] S4: Obtain each data point of the ER wave data from the fifth register 206;

[0108] S5: If the data point is within the preset acquisition area stored in the third register 204, proceed to S6, otherwise proceed to S8;

[0109] S6: Trigger the first comparator 207 to work. If the data point is outside the fluctuation range defined by the preset fluctuation value stored in the second register 203, proceed to S7; otherwise, proceed to S8.

[0110] S7: adding the calculated value of the data point to the calculated value of the previous data point to obtain an accumulated calculated value, and storing the accumulated calculated value in the sixth register 208;

[0111] S8: If the data point is not the last data point of the ER wave data defined by the preset acquisition area, proceed to S9, otherwise proceed to S10;

[0112] S9: Get the next data point of ER wave data and proceed to S5;

[0113] S10: If the accumulated calculation value stored in the sixth register 208 is greater than the preset critical value stored in the fourth register 205, the second comparator 209 inputs a high level to the first input terminal of the third comparator 210, indicating that the preset parameter combination identifies the state of the ER wave data as recognition capture; otherwise, the second comparator 209 inputs a low level to the first input terminal of the third comparator 210, indicating that the preset parameter combination identifies the state of the ER wave data as recognition loss capture;

[0114] S11: If the first input terminal and the second input terminal of the third comparator 210 are both at high levels, it indicates that the preset parameter combination correctly identifies the capture, and the first counter 211 counts once;

[0115] If the first input terminal and the second input terminal of the third comparator 210 are both at low levels, it indicates that the preset parameter combination correctly identifies the loss of capture, and the second counter 212 counts once;

[0116] If the first input terminal of the third comparator 210 is at a high level and the second input terminal is at a low level, it indicates that the preset parameter combination mistakenly identifies the loss of capture as capture, and the third counter 213 counts once.

[0117] If the first input terminal of the third comparator 210 is at a low level and the second input terminal is at a high level, it indicates that the preset parameter combination mistakenly identifies the capture as a loss of capture, and the coefficient fourth counter 214 counts once;

[0118] S12: If the body surface ECG data in the first register 202 is not the body surface ECG data corresponding to the intracavitary ECG data after the last pacing pulse in the first storage unit 102, proceed to S13, otherwise proceed to S14;

[0119] S13: the first register 202 reads the surface ECG data corresponding to the intracavitary ECG data after the next pacing pulse from the first storage unit 102, and performs S3;

[0120] S14: Determine the capture accuracy and the loss-of-capture accuracy of the preset parameter combination according to the counting results of the first counter 211, the second counter 212, the third counter 213 and the fourth counter 214, and save the capture accuracy and the loss-of-capture accuracy of the preset parameter combination to the second storage unit 301, completing the data training of the current preset parameter combination, wherein:

[0121]

[0122]

[0123] S15: Clear the data in the sixth register 208, the first counter 211, the second counter 212, the third counter 213 and the fourth counter 214;

[0124] S16: If the preset parameter combination is not the last preset parameter combination in the memory 201, proceed to S17, otherwise the data training ends;

[0125] S17: The second register 203, the third register 204 and the fourth register 205 read the next set of preset parameter combinations from the memory 201 respectively, and perform S2.

[0126] After data training is completed, there are generally multiple sets of preset parameter combinations that can accurately identify capture and loss of capture states. At this time, it is necessary to screen out the optimal preset parameter combination.

[0127] See also Figure 8 , is a screening method for the optimal preset parameter combination provided in the embodiment of the present application. Specifically, the steps of the screening method are:

[0128] The first step is to select a preset parameter combination with a capture accuracy rate and a loss of capture accuracy rate of 100%, which is the basis for ensuring that the automatic pacing threshold capture function can operate correctly.

[0129] The second step is to select the preset parameter combination with the smallest preset acquisition area, that is, the largest acquisition starting point and the smallest acquisition end point. First, selecting the preset parameter combination with the largest acquisition starting point can avoid the influence of polarization potential to the greatest extent, which is beneficial to improving the reliability of the automatic pacing threshold capture function; secondly, selecting the preset parameter combination with the smallest acquisition end point is beneficial to narrowing the acquisition area of ​​ER waves and reducing the amount of calculation of the pacemaker, especially for the pulse generator with beat-by-beat detection function, which can effectively reduce power consumption and thus extend its service life.

[0130] The third step is to select the preset parameter combination with both the preset fluctuation value and the preset critical value in the middle as the optimal preset parameter combination. For example, if the three groups of the acquisition start point, acquisition end point, preset fluctuation value and preset critical value are 8-15-4-10, 8-15-4-11 and 8-15-4-12 respectively, which are preset parameter combinations that meet the screening conditions of the first two steps, then 8-15-4-11 is selected as the optimal preset parameter combination. The preset fluctuation value and the preset critical value are centered, and the error range allowed by the algorithm is the largest, which is conducive to improving the reliability of the automatic pacing threshold capture function.

[0131] After the screening of the optimal preset parameter combination is completed, the optimal preset parameter combination selected by the user is programmed and the automatic pacing threshold function is turned on.

[0132] See also Fig. 9 , is a schematic diagram of a reference standard for capture and loss of capture provided in an embodiment of the present application. Fig. 9 As shown in , the ventricular pacing threshold is tested by the manual ventricular pacing threshold capture function. In order to obtain the user's actual ventricular pacing threshold, the manual ventricular pacing threshold test function will gradually reduce the ventricular pacing output amplitude until the ventricular pacing pulse cannot capture the myocardium. Therefore, the obtained surface ECG data and intracavitary ECG data must include capture status and loss-of-capture status. Since intracavitary ECG data may be affected by polarization potential, the capture and loss-of-capture status is generally not directly determined through intracavitary ECG data. Since the surface ECG is not affected by polarization potential, the capture and loss-of-capture status can be directly determined through the surface ECG data, thereby establishing a reference standard for capture and loss-of-capture status. That is, the data training system and method for implantable pulse generators provided in the above-mentioned embodiments of the present application are completely feasible.

[0133] See also Fig.10 , which is a schematic diagram showing that the reliability of the prior art in determining loss of capture is reduced when the polarization potential provided in the embodiment of the present application is wide; see Fig.11 , which is a schematic diagram showing that the prior art determines that the capture reliability is reduced when the polarization potential provided in the embodiment of the present application is wide; see Fig.12 , which is a schematic diagram showing that the reliability of the present application in determining the loss of capture is high when the polarization potential provided in the embodiment of the present application is large; see Fig.13 , which is a schematic diagram of the present invention determining that the capture reliability is high when the polarization potential provided in the present invention is wide. Figures 10 to 13 , through a peak-to-peak value example, the advantages of the data training system and method for an implantable pulse generator provided by the embodiments of the present application are explained.

[0134] for Figure 1 and Figure 2In the general case, it is obvious that the peak-to-peak value of the decaptured ER wave is significantly smaller than the peak-to-peak value of the captured ER wave within the ER wave window. Therefore, the capture and decapture states can be accurately determined. Fig.10 and Fig.11 As shown, at this time, the peak-to-peak value P of the ER wave is lost in the ER wave window. LOC The peak value of the captured ER wave P CAP The peak-to-peak value is similar, which reduces the reliability of the peak-to-peak value determination result and even leads to the risk of wrong determination.

[0135] The data training system and method for an implantable pulse generator provided in the embodiment of the present application adopts an effective data extraction scheme based on four parameters: the acquisition starting point, the acquisition end point, the preset fluctuation value, and the preset critical value. That is to say, the embodiment of the present application only collects points between the preset acquisition area determined by the acquisition starting point and the acquisition end point, and regards the points outside the range defined by the preset fluctuation value as valid points, and the rest as invalid points. When the peak-to-peak value obtained is greater than the preset critical value, it is determined to be captured, otherwise it is determined to be lost. For the case where the polarization potential is wide, after adopting this scheme, the following is obtained Fig.12 and Fig.13 , that is, the collection starting point and collection end point are limited to Fig.12 and Fig.13 The collection area is shown. Fig.12 As shown in Figure 1, the entire preset acquisition area is within the preset fluctuation value, that is, the positive and negative fluctuation range defined by the positive and negative fluctuation lines. Therefore, the number of valid points corresponding to the lost ER wave is 0, and the peak-to-peak value P of the lost ER wave is LOC is 0. Fig.13 As shown in the figure, there are many points in the preset acquisition area that are outside the positive and negative fluctuation range defined by the preset fluctuation value, and the peak-to-peak value P of the ER wave is captured. CAP It reflects the difference between the peak and the baseline. Obviously, for the captured ER wave and the lost ER wave obtained under the condition of wide polarization potential, the peak-to-peak value established on the above effective data extraction scheme has a significant difference. Based on this difference, the capture and loss of capture can be accurately judged, so the user can benefit from the automatic pacing threshold capture function.

[0136] From the above embodiments, it can be seen that adjusting the acquisition starting point, that is, setting the value of the acquisition starting point to be larger, can effectively reduce or even eliminate the influence of the wide polarization wave on the ER wave; adjusting the acquisition end point, that is, setting the value of the acquisition end point to be smaller, can effectively narrow the acquisition range of the ER wave, which is beneficial to reducing the amount of calculation of the pacemaker, especially for a pulse generator with a beat-by-beat detection function, which can significantly reduce the amount of calculation of the pacemaker. Adjusting the preset fluctuation value, that is, setting the value of the preset fluctuation value to be larger, is beneficial to expanding the difference between the characteristic values ​​of the captured ER wave and the uncaptured ER wave, thereby improving the sensitivity of the algorithm.

[0137] The data training system and method for an implantable pulse generator provided in the embodiment of the present application can screen out the optimal preset parameter combination through data training, and apply the optimal preset parameter combination to the automatic pacing threshold capture function, which can effectively improve the reliability of the automatic pacing threshold capture function, and can enable more users to benefit from the automatic pacing threshold capture function, and can also reduce the amount of calculation of the pacemaker, especially for a pulse generator with a beat-by-beat detection function, which can effectively reduce power consumption and thus extend its service life.

Claims

1. A data training system for an implantable pulse generator, It is characterized in that It comprises a data reading module (100), a data training module (200) and a storage module (300); The data reading module (100) is used to read body surface electrocardiogram data and intracavitary electrocardiogram data, and pre-process the body surface electrocardiogram data and the intracavitary electrocardiogram data to obtain a determination basis; The data reading module (100) comprises a data preprocessing unit (101) and a first storage unit (102); The data preprocessing unit (101) comprises a filter, an amplifier and an AD converter which are electrically connected in sequence and are used to preprocess the surface electrocardiogram data and the intracavitary electrocardiogram data to obtain a basis for determination; The first storage unit (102) is electrically connected to the data preprocessing unit (101) and is used to store intracavitary electrocardiogram data, surface electrocardiogram data and the determination basis; The data training module (200) is electrically connected to the data reading module (100) and is used to compare the intracavitary electrocardiogram data with a preset parameter combination, obtain a comparison result of the capture and loss of capture state of each intracavitary electrocardiogram data for each preset parameter combination, and obtain the capture accuracy rate and loss of capture accuracy rate of the preset parameter combination according to the determination basis and the comparison result; The data training module (200) comprises a memory (201), a plurality of registers, a plurality of comparators and a plurality of counters; The memory (201) is used to store the preset parameter combination, wherein the preset parameter combination includes a preset fluctuation value, a preset collection area and a preset critical value; The registers include a first register (202), a second register (203), a third register (204), a fourth register (205), a fifth register (206) and a sixth register (208); The first register (202) is electrically connected to the first storage unit (102) and is used to read the body surface electrocardiogram data corresponding to each of the intracavitary electrocardiogram data from the first storage unit (102), and read the determination basis corresponding to the body surface electrocardiogram data, wherein the determination basis includes capture determination and loss of capture determination; The second register (203), the third register (204) and the fourth register (205) are respectively electrically connected to the memory (201), and all read the preset parameter combination from the memory (201), wherein the second register (203) stores the preset fluctuation value, the third register (204) stores the preset acquisition area, and the fourth register (205) stores the preset critical value; The fifth register (206) is electrically connected to the storage unit (102) and is used to read each ER wave data, wherein the ER wave data is the data of the intracavitary electrocardiogram data in a preset time period after each pacing pulse; The sixth register (208) is electrically connected to the first comparator (207) and is used to store the accumulated calculated value of each data point calculated value of the ER wave data that meets the preset parameter combination; The comparator comprises a first comparator (207), a second comparator (209) and a third comparator (210); The first comparator (207) is electrically connected to the second register (203), the third register (204) and the fifth register (206) respectively, and is used to compare the ER wave data with the preset fluctuation value and the preset acquisition area; The second comparator (209) is electrically connected to the fourth register (205) and the sixth register (208) respectively, and is used to compare the accumulated calculation value with the preset critical value and output a corresponding comparison result, wherein the comparison result includes the recognition of capture and the recognition of loss of capture; The third comparator (210) is electrically connected to the first register (202) and the second comparator (209) respectively, and is used to compare the comparison result output by the second comparator (209) with the determination basis output by the first register (202), and output a recognition result; The counter comprises a first counter (211), a second counter (212), a third counter (213) and a fourth counter (214) which are electrically connected to the third comparator (210) respectively; The first counter (211) is used to count once when the determination capture and the identification capture are simultaneously input to the input end of the third comparator (210); The second counter (212) is used to count once when the determination loss of capture and the identification loss of capture are simultaneously input to the input end of the third comparator (210); The third counter (213) is used to count once when the determination loss of capture and the identification capture are simultaneously input to the input end of the third comparator (210); The fourth counter (214) is used to count once when the capture determination and the loss of capture identification are simultaneously input to the input end of the third comparator (210); The storage module (300) is electrically connected to the data training module (200) and is used to store the capture accuracy and loss-of-capture accuracy of all the preset parameter combinations.

2. The data training system for an implantable pulse generator according to claim 1, It is characterized in that The storage module (300) comprises a storage unit (301), and the storage unit (301) is used to store the capture accuracy and the loss-capture accuracy of all the preset parameter combinations.

3. A data training method for an implantable pulse generator, It is characterized in that The following steps are involved: Obtain surface ECG data and intracavitary ECG data; Establishing a basis for determining capture and loss of capture based on the surface electrocardiogram data; Comparing the intracavitary electrocardiogram data with a preset parameter combination to obtain a comparison result of the capture and loss of capture state of each intracavitary electrocardiogram data for each preset parameter combination, wherein the preset parameter combination includes a preset fluctuation value, a preset acquisition area, and a preset critical value; Comparing the comparison result with the determination basis to obtain the capture accuracy rate and the loss-capture accuracy rate of each of the preset parameter combinations; An optimal preset parameter combination is determined, wherein the optimal preset parameter combination is determined according to the capture accuracy rate and the loss-capture accuracy rate of each preset parameter combination.

4. The data training method for an implantable pulse generator according to claim 3, It is characterized in that Determining the optimal preset parameter combination comprises the following steps: Determining an accurate preset parameter combination, wherein the accurate preset parameter combination is the preset parameter combination in which the capture accuracy rate and the loss of capture accuracy rate are both 100%; Determine a minimum acquisition area preset parameter combination, wherein the minimum acquisition area preset parameter combination is the accurate preset parameter combination with the smallest preset acquisition area; An optimal preset parameter combination is determined, where the optimal preset parameter combination is the minimum acquisition area preset parameter combination in which the preset fluctuation value and the preset critical value are both centered.

5. A data training method for an implantable pulse generator, It is characterized in that The method is applied to the data training system for an implantable pulse generator according to claim 2, comprising the following steps: S1: the second register (203), the third register (204) and the fourth register (205) respectively read the preset parameter combination from the memory (201); S2: the first register (202) reads the surface electrocardiogram data corresponding to the intracavitary electrocardiogram data after a pacing pulse from the first storage unit (102), and inputs the corresponding determination basis to the second input end of the third comparator (210), wherein if the surface electrocardiogram data is determined to be captured, the second input end inputs a high level; if the surface electrocardiogram data is determined to be lost capture, the second input end inputs a low level; S3: the fifth register (206) reads the ER wave data after the pacing pulse of the intracavitary electrocardiogram data from the first storage unit (102); S4: Obtaining each data point of the ER wave data from the fifth register (206); S5: If the data point is within the preset acquisition area stored in the third register (204), proceed to S6; otherwise, proceed to S8; S6: triggering the first comparator (207) to work, if the data point is outside the fluctuation range defined by the preset fluctuation value stored in the second register (203), proceed to S7, otherwise proceed to S8; S7: adding the calculated value of the data point to the calculated value of the previous data point to obtain an accumulated calculated value, and storing the accumulated calculated value in a sixth register (208); S8: If the data point is not the last data point of the ER wave data defined by the preset acquisition area, proceed to S9, otherwise proceed to S10; S9: Obtain the next data point of the ER wave data, and proceed to S5; S10: If the accumulated calculation value stored in the sixth register (208) is greater than the preset critical value stored in the fourth register (205), the second comparator (209) inputs a high level to the first input terminal of the third comparator (210), indicating that the state of the ER wave data identified by the preset parameter combination is recognition capture; otherwise, the second comparator (209) inputs a low level to the first input terminal of the third comparator (210), indicating that the state of the ER wave data identified by the preset parameter combination is recognition loss capture; S11: If the first input terminal and the second input terminal of the third comparator (210) are both at a high level, the first counter (211) performs one counting; If the first input terminal and the second input terminal of the third comparator (210) are both at a low level, the second counter (212) performs one count; If the first input terminal of the third comparator (210) is at a high level and the second input terminal is at a low level, the third counter (213) performs one counting; If the first input terminal of the third comparator (210) is at a low level and the second input terminal is at a high level, the fourth counter (214) performs one counting; S12: If the body surface electrocardiogram data in the first register (202) is not the body surface electrocardiogram data corresponding to the intracavitary electrocardiogram data after the last pacing pulse in the first storage unit (102), then proceed to S13; otherwise, proceed to S14; S13: the first register (202) reads the surface electrocardiogram data corresponding to the intracardiac electrocardiogram data after the next pacing pulse from the first storage unit (102), and performs S3; S14: determining the capture accuracy rate and the loss-capture accuracy rate of the preset parameter combination according to the counting results of the first counter (211), the second counter (212), the third counter (213) and the fourth counter (214), and saving the capture accuracy rate and the loss-capture accuracy rate of the preset parameter combination to the second storage unit (301); S15: Clearing the data in the sixth register (208), the first counter (211), the second counter (212), the third counter (213) and the fourth counter (214); S16: if the preset parameter combination is not the last preset parameter combination in the memory (201), proceed to S17; otherwise, the data training ends; S17: The second register (203), the third register (204) and the fourth register (205) respectively read the next set of preset parameter combinations from the memory (201) and perform S2.

6. The data training method for an implantable pulse generator according to claim 5, It is characterized in that The capture accuracy rate and the loss capture accuracy rate are calculated by the following formulas: