Method and system for cardiac rhythm recognition, medical device, medical system and storage medium

Abstract

The application is applied to the field of heart rhythm recognition, and provides a heart rhythm recognition method and system, a medical device, a medical system and a storage medium. The method comprises the following steps: acquiring first electrocardio characteristic data of a specified first user; performing heart rhythm recognition on the first electrocardio characteristic data based on a preset heart rhythm recognition algorithm and a rule model library of the medical device, to obtain a first heart rhythm recognition result, wherein the rule model library comprises a heart rhythm recognition rule generated based on historical electrocardio event records of the specified first user; obtaining a first defibrillation recognition result according to the first heart rhythm recognition result; if the first defibrillation recognition result is a defibrillatable result, acquiring second electrocardio characteristic data of the specified first user, performing heart rhythm recognition on the second electrocardio characteristic data, to obtain a second heart rhythm recognition result of the medical device; and obtaining a second defibrillation recognition result based on the aforementioned heart rhythm recognition result. The application greatly improves the accuracy of heart rhythm recognition.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a heart rhythm recognition method and system, medical device, medical system, and storage medium. Background Technology

[0002] Cardiac arrest is a life-threatening emergency that can occur in people of all ages. To provide timely treatment, numerous defibrillation devices have been developed. These devices identify the heart rhythm by collecting the patient's electrocardiogram (ECG) data and determine whether defibrillation is necessary based on the results. The existing rhythm identification process includes: monitoring the patient's ECG signal; initially determining if a malignant arrhythmia has occurred; if not, monitoring continues; if a malignant arrhythmia occurs, pre-charging of the high-voltage capacitor begins. After pre-charging, the algorithm secondly determines if the malignant arrhythmia persists. If the malignant arrhythmia persists, defibrillation is initiated; if the malignant arrhythmia terminates spontaneously, defibrillation is abandoned, and ECG monitoring continues.

[0003] While current heart rhythm recognition algorithms boast high accuracy, their development and optimization are based on general-purpose datasets. This limitation, stemming from the data integrity of these datasets, impacts accuracy. A common solution is to enhance algorithm robustness through continuous iteration and upgrades. However, once medical devices are deployed, safety considerations often limit algorithm upgrades to subsequent releases. Current algorithms are embedded in the main unit of implantable cardioverter-defibrillators (ICDs) at the factory. Once implanted, the algorithm cannot be upgraded or optimized; only a few parameters can be fine-tuned via a programmer. There remains no effective solution for post-market algorithm upgrade needs. Furthermore, individual patients possess unique heart rhythm characteristics that may differ from general datasets. Using existing algorithms uniformly could lead to misjudgments of these patient-specific rhythms, resulting in false or missed defibrillation. In particular, the misjudgment of the first heart rhythm recognition not only causes inconvenience to users, but also leads to repeated pre-charging. The energy loss during the pre-charging process greatly shortens the lifespan of medical devices with limited power.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To address the problems in the prior art, the present invention aims to provide a heart rhythm recognition method and system, medical device, medical system, and storage medium, which, combined with a rule model library specific to a given user and a preset heart rhythm recognition algorithm, greatly improves the accuracy of heart rhythm recognition.

[0006] This invention provides a method for heart rhythm recognition, comprising the following steps:

[0007] Obtain the first electrocardiogram (ECG) feature data of the specified first user;

[0008] Based on a preset heart rhythm recognition algorithm and rule model library of medical devices, heart rhythm recognition is performed on the first electrocardiogram feature data to obtain a first heart rhythm recognition result. The rule model library includes heart rhythm recognition rules generated based on the historical electrocardiogram event records of the specified first user.

[0009] The first defibrillation identification result is obtained based on the first heart rhythm identification result;

[0010] If the first defibrillation identification result is a defibrillable result, the second electrocardiogram feature data of the designated first user is obtained;

[0011] Based on the preset heart rhythm recognition algorithm of the medical device and / or the rule model library, heart rhythm recognition is performed on the second electrocardiogram feature data to obtain the second heart rhythm recognition result of the medical device.

[0012] A second defibrillation identification result is obtained based on the second heart rhythm identification result or based on the first heart rhythm identification result and the second heart rhythm identification result.

[0013] In some embodiments, obtaining a second defibrillation identification result based on the first heart rhythm identification result and the second heart rhythm identification result includes the following steps:

[0014] Obtain the first weight of the first heart rhythm recognition result and the second weight of the second heart rhythm recognition result;

[0015] The first heart rhythm recognition result and the second heart rhythm recognition result are weighted and averaged to obtain the final heart rhythm recognition result.

[0016] A second defibrillation identification result is obtained based on the final heart rhythm identification result.

[0017] In some embodiments, the rule model library is formed in the following manner:

[0018] The historical ECG event records of the medical device are obtained. Each historical ECG event record includes the historical ECG feature data of the designated first user and the second sub-defibrillation identification result of the medical device. The second sub-defibrillation identification result is obtained after performing heart rhythm identification based on a preset heart rhythm identification algorithm and the second ECG feature data.

[0019] The historical ECG event records are sent to the processing device;

[0020] The processing device receives heart rhythm recognition rules, each of which includes an electrocardiogram (ECG) characteristic condition and a standard defibrillation recognition result. The ECG characteristic condition is generated based on the historical ECG characteristic data of the designated first user.

[0021] The heart rhythm recognition rules are stored in the rule model library of the medical device.

[0022] In some embodiments, the processing device is configured to form or automatically update the heart rhythm recognition rules using the following steps:

[0023] Receive the historical ECG event records;

[0024] Obtain the standard defibrillation identification results corresponding to the historical ECG event records;

[0025] The heart rhythm recognition rules are obtained based on the historical electrocardiogram event records and the corresponding standard defibrillation recognition results;

[0026] The heart rhythm recognition rules are automatically updated each time a preset condition is met. The preset condition is one of the following: reaching a specific time, reaching a specific amount of data, or receiving a specific instruction.

[0027] In some embodiments, the processing device includes a processing terminal and a processing server, and the processing device obtains the standard defibrillation identification result corresponding to the historical electrocardiogram event record using the following steps:

[0028] The processing terminal displays the received historical ECG event records, receives several defibrillation identification results input by several second users, generates a standard defibrillation identification result based on the several defibrillation identification results, and sends the historical ECG event records and the standard defibrillation identification result to the processing server; or,

[0029] The processing terminal sends the received historical ECG event records to the processing server. The processing server pushes the received historical ECG event records to several second users, receives several defibrillation identification results input by several second users, generates a standard defibrillation identification result based on the several defibrillation identification results, and stores the historical ECG event records and the standard defibrillation identification results.

[0030] In some embodiments, generating a standard defibrillation identification result based on several defibrillation identification results includes: determining defibrillation identification results whose proportion is greater than a preset proportion threshold as standard defibrillation identification results.

[0031] In some embodiments, the processing device is configured to obtain the heart rhythm recognition rule by performing the following steps:

[0032] The processing server filters the historical ECG event records to obtain misjudgment records where the second sub-defibrillation identification result of the medical device is inconsistent with the standard defibrillation identification result.

[0033] The processing server pushes the misjudgment record of the designated first user to the second user;

[0034] The processing server obtains the heart rhythm recognition rules added by the second user.

[0035] In some embodiments, the processing device is configured to obtain the heart rhythm recognition rule by performing the following steps:

[0036] The processing server obtains the historical ECG event records and the standard defibrillation identification results of the designated first user;

[0037] The processing server filters out misjudgment records where the second sub-defibrillation identification result of the medical device is inconsistent with the standard defibrillation identification result.

[0038] The processing server determines whether to generate pre-added heart rhythm recognition rules based on the historical electrocardiogram feature data corresponding to the misjudged records.

[0039] The processing server pushes the generated pre-added heart rhythm recognition rule to the second user, and upon receiving the second user's confirmation instruction, uses the pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule; or upon receiving the second user's modification instruction, uses the modified pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule.

[0040] In some embodiments, the processing server determines whether to generate pre-added heart rhythm recognition rules based on the historical electrocardiogram feature data corresponding to the misjudged records, including the following steps:

[0041] The processing server classifies the misjudged records according to the standard defibrillation identification results to obtain a first misjudged record that can be defibrillated and a second misjudged record that is not defibrillable.

[0042] For the first misjudged record and the second misjudged record, the feature values ​​in the historical electrocardiogram feature data are arranged and combined respectively;

[0043] Calculate the correlation coefficient between each combination and the misjudgment result;

[0044] If the correlation coefficient of at least one of the combinations is greater than a preset correlation threshold, a pre-added heart rhythm recognition rule is generated, the feature value of the combination is used as the ECG feature condition, and the category of the standard defibrillation recognition result is determined according to whether the historical ECG feature data belongs to the first misjudged record or the second misjudged record.

[0045] In some embodiments, after confirming the addition of the pre-added heart rhythm recognition rule as the heart rhythm recognition rule, the following steps are further included:

[0046] The processing server calculates the probability of misjudging historical ECG feature data that conforms to the confirmed added heart rhythm recognition rules based on the historical ECG feature data of the designated first user, the second defibrillation recognition result of the medical device, and the standard defibrillation recognition result;

[0047] The processing server sets a third weight value for the heart rhythm recognition rule based on the probability of misjudgment.

[0048] In some embodiments, the process of performing heart rhythm recognition on the first electrocardiogram feature data based on a preset heart rhythm recognition algorithm and rule model library of the medical device to obtain the first heart rhythm recognition result of the medical device includes the following steps:

[0049] Based on the first electrocardiogram feature data and the preset heart rhythm recognition algorithm, the first sub-heart rhythm recognition result is obtained;

[0050] Determine whether the first ECG feature data matches a certain heart rhythm recognition rule in the rule model library. If so, determine the second sub-heart rhythm recognition result based on the standard defibrillation recognition result of the matched heart rhythm recognition rule.

[0051] Obtain the third weight value of the matching heart rhythm recognition rule, use it as the third weight value of the second sub-heart rhythm recognition result, and calculate the fourth weight value of the first sub-heart rhythm recognition result based on the third weight value;

[0052] The first sub-rhythm recognition result is obtained by weighting and averaging the first sub-rhythm recognition result and the second sub-rhythm recognition result.

[0053] In some embodiments, the historical ECG feature data includes the ECG feature data of the first user at the time of the historical ECG event, and the ECG feature data of the first user during a first time period before the time of the historical ECG event and during a second time period after the time of the historical ECG event.

[0054] In some embodiments, the first electrocardiogram feature data includes at least one of real-time acquired electrocardiogram signal graphs and electrocardiogram feature parameters, and the second electrocardiogram feature data includes at least one of real-time acquired electrocardiogram signal graphs and electrocardiogram feature parameters.

[0055] This invention also provides a heart rhythm recognition system, the system comprising:

[0056] The first acquisition module is used to acquire the first electrocardiogram feature data of a specified first user;

[0057] The first judgment module is used to perform heart rhythm recognition on the first electrocardiogram feature data based on the preset heart rhythm recognition algorithm and rule model library of the medical device, to obtain a first heart rhythm recognition result, and to obtain a first defibrillation recognition result based on the first heart rhythm recognition result. The rule model library includes heart rhythm recognition rules generated based on the historical electrocardiogram event records of the specified first user.

[0058] The second acquisition module is used to acquire the second electrocardiogram feature data of the designated first user if the first defibrillation identification result is a defibrillable result.

[0059] The second judgment module is used to perform heart rhythm recognition on the second electrocardiogram feature data based on the preset heart rhythm recognition algorithm of the medical device and / or the rule model library, to obtain the second heart rhythm recognition result of the medical device, and to obtain the second defibrillation recognition result based on the second heart rhythm recognition result or based on the first heart rhythm recognition result and the second heart rhythm recognition result.

[0060] This invention also provides a medical device, comprising:

[0061] processor;

[0062] The memory stores a preset heart rhythm recognition algorithm, a rule model library, and executable instructions of the processor.

[0063] The communication module is used to communicate with the processing device;

[0064] The processor is configured to perform the steps of the heart rhythm recognition method by executing the executable instructions.

[0065] This invention also provides a medical system, comprising: a processing device for receiving historical electrocardiogram (ECG) event records from a medical device, obtaining standard defibrillation identification results corresponding to the historical ECG event records, and obtaining heart rhythm identification rules based on the historical ECG event records and the corresponding standard defibrillation identification results; and the medical device.

[0066] This invention also provides a computer-readable storage medium for storing a program, which, when executed by a processor, implements the steps of the heart rhythm recognition method.

[0067] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0068] The heart rhythm recognition method and system, medical device, medical system, and storage medium of the present invention have the following beneficial effects:

[0069] By employing this invention, during the first heart rhythm recognition, a rule model library specific to a designated user and a preset heart rhythm recognition algorithm are combined to improve the accuracy of the first heart rhythm recognition, avoid repeated pre-charging due to misjudgment in the first heart rhythm recognition, save pre-charging losses, effectively conserve the limited power of medical devices, and avoid unnecessary second heart rhythm recognition. Furthermore, the rule model library stores heart rhythm recognition rules specific to the designated user, serving as a supplement to the universal preset heart rhythm recognition algorithm in medical devices, effectively reducing the risk of misjudgment by medical devices for individual patients. Attached Figure Description

[0070] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0071] Figure 1 This is a flowchart of a heart rhythm recognition method according to an embodiment of the present invention;

[0072] Figure 2 This is a flowchart of a generation rule model library according to an embodiment of the present invention;

[0073] Figure 3 This is a flowchart illustrating the formation or automatic updating of heart rhythm recognition rules according to an embodiment of the present invention;

[0074] Figure 4 This is a flowchart of a process server automatically acquiring the heart rhythm recognition rules according to an embodiment of the present invention;

[0075] Figure 5 This is a flowchart illustrating how the first heart rhythm recognition result of the medical device is obtained according to an embodiment of the present invention;

[0076] Figure 6 This is a schematic diagram of the structure of a heart rhythm recognition system according to an embodiment of the present invention;

[0077] Figure 7 This is a schematic diagram of the structure of a medical device according to an embodiment of the present invention;

[0078] Figure 8 This is a schematic diagram of the structure of a medical system according to an embodiment of the present invention. Detailed Implementation

[0079] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0080] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0081] The flowchart shown in the attached diagram is merely an illustrative example and does not necessarily include all steps. For example, some steps may be broken down, while others may be combined or partially combined. Therefore, the actual execution order may change depending on the specific circumstances.

[0082] A patient's own heart rhythm characteristics have inherent features, whether it is a normal sinus rhythm or a malignant rhythm. For a given patient, there are certain similarities between multiple malignant rhythms. Based on this, this invention provides a novel heart rhythm recognition method. A pre-set rule model library is incorporated into the medical device to store heart rhythm recognition rules specific to each patient, supplementing the pre-set heart rhythm recognition algorithm within the medical device. This significantly reduces the risk of misdiagnosis of individual patients after the medical device is released to the market. Furthermore, it is precisely because of this medical theoretical foundation that the algorithm proposed in this invention has high practical value.

[0083] like Figure 1 As shown, in order to solve the technical problems in the prior art, the present invention provides a heart rhythm recognition method, comprising the following steps:

[0084] S100: Obtain the first electrocardiogram feature data of the specified first user;

[0085] The first user is specified here, for example, a patient wearing a medical device. The first electrocardiogram feature data includes at least one of the real-time acquired electrocardiogram signal and electrocardiogram feature parameters. Specifically, the first electrocardiogram feature data includes, but is not limited to: heart rate, overall and local morphology of electrocardiogram signal, QRS pulse width, PP interval, RR interval, PR interval, etc.

[0086] S200: Based on the preset heart rhythm recognition algorithm and rule model library of the medical device, heart rhythm recognition is performed on the first electrocardiogram feature data to obtain the first heart rhythm recognition result. The rule model library includes heart rhythm recognition rules generated based on the historical electrocardiogram event records of the specified first user.

[0087] The preset heart rhythm recognition algorithm is the existing heart rhythm recognition algorithm fixed in the medical device. It collects the patient's electrocardiogram signal, analyzes the electrocardiogram feature data, determines whether there is a malignant heart rhythm event (defibrillable heart rhythm event), and obtains a heart rhythm recognition result. This heart rhythm recognition result is the risk value of whether it is a defibrillable heart rhythm.

[0088] The heart rhythm recognition rules in the rule model library here are generated based on the historical ECG event records of a designated first user. The first user corresponding to the historical ECG event records is the patient wearing the medical device, which is the same user as the designated first user in step S100.

[0089] In step S200 here, a first heart rhythm recognition result is obtained by combining the preset heart rhythm recognition algorithm and rule model library of the medical device. The specific method of obtaining the result will be described below. This first heart rhythm recognition result is the risk value of whether it is a defibrillable heart rhythm.

[0090] S300: Obtain the first defibrillation recognition result based on the first heart rhythm recognition result;

[0091] The first defibrillation identification result may be one of two things: a defibrillable result or a non-defibrillable result. If the first defibrillation identification result is a non-defibrillable result, the current judgment process ends, that is, step S400 is not executed, and the process waits for the next heart rhythm identification process to be executed.

[0092] Specifically, for example, a heart rhythm recognition result threshold is set. When the first heart rhythm recognition result is greater than or equal to the heart rhythm recognition result threshold, the first defibrillation recognition result is a defibrillable result. When the first heart rhythm recognition result is less than the heart rhythm recognition result threshold, the first defibrillation recognition result is a non-defibrillable result.

[0093] S400: If the first defibrillation identification result is a defibrillable result, obtain the second electrocardiogram feature data of the designated first user;

[0094] For a defibrillation medical device that can automatically identify heart rhythm and automatically defibrillate, if the first defibrillation identification result is defibrillable, the high-voltage capacitor of the defibrillation actuator is automatically charged, and the second electrocardiogram feature data of the designated first user is acquired, and the subsequent steps S500 and S600 are continued; in another embodiment, if the first defibrillation identification result is defibrillable, the high-voltage capacitor of the defibrillation actuator is automatically charged, and after a preset charging time, the second electrocardiogram feature data of the designated first user is acquired, and the subsequent steps S500 and S600 are continued.

[0095] The second electrocardiogram feature data includes at least one of the real-time acquired electrocardiogram signal graph and electrocardiogram feature parameters. Specifically, the second electrocardiogram feature data includes, but is not limited to: heart rate, overall and local morphology of electrocardiogram signal, QRS pulse width, PP interval, RR interval, PR interval, etc. The second electrocardiogram feature data may be the same as or different from the first electrocardiogram feature data.

[0096] S500: Based on the preset heart rhythm recognition algorithm of the medical device and / or the rule model library, perform heart rhythm recognition on the second electrocardiogram feature data to obtain the second heart rhythm recognition result of the medical device;

[0097] In the first embodiment, during the second heart rhythm recognition, only the preset heart rhythm recognition algorithm of the medical device is used to perform heart rhythm recognition on the second-collected electrocardiogram feature data to obtain the second heart rhythm recognition result of the medical device; in the second embodiment, during the second heart rhythm recognition, only the rule model library is used to perform heart rhythm recognition on the second-collected electrocardiogram feature data to obtain the second heart rhythm recognition result of the medical device; in the third embodiment, during the second heart rhythm recognition, both the preset heart rhythm recognition algorithm of the medical device and the rule model library are combined to perform heart rhythm recognition on the second-collected electrocardiogram feature data to obtain the second heart rhythm recognition result of the medical device.

[0098] S600: Obtain a second defibrillation recognition result based on the second heart rhythm recognition result or based on the first heart rhythm recognition result and the second heart rhythm recognition result.

[0099] In a first embodiment, in step S600, the second defibrillation identification result is obtained directly based on the second heart rhythm identification result. For example, a heart rhythm identification result threshold is set. When the second heart rhythm identification result is greater than or equal to the heart rhythm identification result threshold, the second defibrillation identification result is a defibrillable result; when the second heart rhythm identification result is less than the heart rhythm identification result threshold, the second defibrillation identification result is a non-defibrillable result.

[0100] In the second embodiment, the second defibrillation identification result is obtained by combining the first heart rhythm identification result and the second heart rhythm identification result, taking into account the results of the two identifications. Specifically, step S600: obtaining the second defibrillation identification result based on the first heart rhythm identification result and the second heart rhythm identification result includes the following steps:

[0101] Obtain the first weight k1 of the first heart rhythm recognition result and the second weight k2 of the second heart rhythm recognition result;

[0102] The first heart rhythm recognition result m1 and the second heart rhythm recognition result m2 are weighted and averaged to obtain the final heart rhythm recognition result (m1*k1+m2*k2) / (k1+k2);

[0103] A second defibrillation identification result is obtained based on the final heart rhythm identification result; for example, when the final heart rhythm identification result is greater than or equal to the heart rhythm identification result threshold, the second defibrillation identification result is a defibrillable result, and when the final heart rhythm identification result is less than the heart rhythm identification result threshold, the second defibrillation identification result is a non-defibrillable result.

[0104] This invention employs a heart rhythm recognition algorithm. Through steps S100 to S200, during the first heart rhythm recognition, it combines a rule model library specific to a given user with a preset heart rhythm recognition algorithm to improve the accuracy of the first heart rhythm recognition. Only when defibrillation is determined to be necessary in step S300 is a second heart rhythm recognition performed through steps S400 and S500, effectively avoiding unnecessary second heart rhythm recognition and improving the accuracy of heart rhythm recognition. The rule model library stores heart rhythm recognition rules specific to the given user, supplementing the preset heart rhythm recognition algorithm commonly used in medical devices, and can effectively reduce the risk of misjudgment of individual patients by medical devices.

[0105] The heart rhythm recognition method can be executed by a medical device, which can be an implantable cardioverter-defibrillator (ICD) or other implantable cardiac monitoring / treatment device, or it can be an external cardiac monitoring / treatment device. This medical device has a pre-set general heart rhythm recognition algorithm and a rule model library for expansion. When the medical device is a cardiac monitoring device, it can collect electrocardiogram (ECG) characteristic data, analyze the collected ECG characteristic data, determine the risk of malignant arrhythmia, and determine whether defibrillation treatment is needed based on the determined risk, using this determination result as the heart rhythm recognition result. When the medical device is a cardiac treatment device, after collecting ECG characteristic data, analyzing the collected ECG characteristic data, determining the risk of malignant arrhythmia as the heart rhythm recognition result, it also determines whether to perform electrical defibrillation treatment based on the heart rhythm recognition result.

[0106] For implantable cardioverter defibrillators (ICDs), the current preset rhythm recognition algorithm is fixed in the defibrillator unit at the factory. Once the defibrillator unit is implanted, the algorithm cannot be upgraded or optimized; only a few parameters can be fine-tuned via a programmer. Since fine-tuning these parameters makes it difficult to optimize the algorithm as a whole, this invention addresses this by setting up an additional rule model library. Rhythm recognition rules generated based on the historical ECG event records of a specified first user can be added as needed, effectively improving the accuracy of rhythm recognition without removing the defibrillator unit. Similarly, for external defibrillators (EDBs), the accuracy and reliability of rhythm recognition can be further improved without altering the internal algorithm.

[0107] like Figure 2 As shown, in this embodiment, the rule model library is formed in the following manner:

[0108] S011: Obtain historical ECG event records from the medical device. Each historical ECG event record includes historical ECG feature data of the designated first user and the second sub-defibrillation identification result of the medical device. The second sub-defibrillation identification result is obtained after performing heart rhythm identification based on a preset heart rhythm identification algorithm and the second ECG feature data.

[0109] The historical ECG feature data includes the ECG feature data of the first user at the time of the historical ECG event, and the ECG feature data of the first user in the first time period before the time of the historical ECG event and in the second time period after the time of the historical ECG event.

[0110] S012: Send the historical ECG event records to the processing device;

[0111] S013: Receive heart rhythm recognition rules from the processing device, each heart rhythm recognition rule including electrocardiogram feature conditions and standard defibrillation recognition results, wherein the electrocardiogram feature conditions are generated based on the historical electrocardiogram feature data of the designated first user;

[0112] S014: Store the heart rhythm recognition rules in the rule model library of the medical device.

[0113] This rule model library is a pre-installed interface in medical devices. It can be set to empty at the factory and expanded upon receiving new heart rhythm recognition rules from the processing device. Because this heart rhythm recognition rule corresponds to the electrocardiogram characteristic data and standard defibrillation recognition results of the same specified patient, it is more consistent with the individual heart rhythm characteristics of the patient.

[0114] like Figure 3 As shown, in this embodiment, the processing device is configured to form or automatically update the heart rhythm recognition rules using the following steps:

[0115] S021: Receive the historical ECG event records;

[0116] S022: Obtain the standard defibrillation identification result corresponding to the historical ECG event record;

[0117] S023: Obtain the heart rhythm recognition rule based on the historical ECG event records and the corresponding standard defibrillation recognition results;

[0118] S024: When a preset condition is met each time, the heart rhythm recognition rule is automatically updated. The preset condition is one of the following: reaching a specific time, reaching a specific amount of data, or receiving a specific instruction. The specific time may include periodic time nodes or specific time points, and the specific instruction may include a heart rhythm recognition rule update instruction, etc.

[0119] In this embodiment, the processing device includes a processing terminal and a processing server. The processing terminal is a terminal device capable of communicating with medical devices (wireless and / or wired communication), such as an external programmer, or it can be a terminal device used by a second user, such as a mobile phone or computer. The processing server is, for example, a cloud server that wirelessly communicates with the processing terminal, or a local server that wirelessly / wirelessly communicates with the processing terminal.

[0120] In one embodiment, in step S022, the processing device obtains the standard defibrillation identification result corresponding to the historical ECG event record using the following steps:

[0121] The processing terminal displays received historical ECG event records, receives several defibrillation identification results input by several second users, generates a standard defibrillation identification result based on these results, and sends the historical ECG event records and the standard defibrillation identification result to the processing server. In this embodiment, the second users are, for example, doctors or specific professionals. The processing terminal can display historical ECG event records to multiple second users, who each input defibrillation identification results, and each second user is unaware of the defibrillation identification results input by other second users. The defibrillation identification results input by the second users can be of two types: defibrillable results and non-defibrillable results.

[0122] Specifically, generating a standard defibrillation identification result based on several defibrillation identification results includes: determining defibrillation identification results whose percentage is greater than a preset percentage threshold as standard defibrillation identification results. For example, after obtaining multiple defibrillation identification results, if more than 75% of the defibrillation identification results are defibrillable, then the standard defibrillation identification results of that historical ECG event record are set as defibrillable; if more than 75% of the defibrillation identification results are not defibrillable, then the standard defibrillation identification results of that historical ECG record are set as not defibrillable. If a defibrillation identification result greater than 75% cannot be obtained, for example, 55% are defibrillable and 45% are not defibrillable, then the standard defibrillation identification result is not determined. Here, 75% is only an example of a preset percentage threshold, and this threshold can also be set to other values ​​as needed, such as 70%, 80%, etc.

[0123] In another embodiment, in step S022, the processing device obtains the standard defibrillation identification result corresponding to the historical ECG event record using the following steps:

[0124] The processing terminal sends the received historical ECG event records to the processing server. The processing server pushes the received historical ECG event records to several second users, receives several defibrillation identification results input by the second users, generates a standard defibrillation identification result based on the several defibrillation identification results, and stores the historical ECG event records and the standard defibrillation identification result. Similarly, the process of the processing server generating the standard defibrillation identification result can also be to determine the defibrillation identification results with a proportion greater than a preset proportion threshold as the standard defibrillation identification result.

[0125] In one implementation, a second user can determine whether to generate a heart rhythm recognition rule based on historical data and select to add the rule. Specifically, in step S023, the processing device is configured to obtain the heart rhythm recognition rule using the following steps:

[0126] The processing server filters the historical ECG event records to obtain misjudged records where the second sub-defibrillation identification result of the medical device and the standard defibrillation identification result are inconsistent. For example, if the second sub-defibrillation identification result of the medical device is a defibrillable identification result, while the standard defibrillation identification result is a non-defibrillable identification result, or if the second sub-defibrillation identification result of the medical device is a non-defibrillable identification result, while the standard defibrillation identification result is a defibrillable identification result, both of these are considered inconsistent results, and the corresponding historical ECG event records are misjudged records.

[0127] The processing server pushes the misjudgment records of the designated first user to the second user. Specifically, the processing server can select all the accumulated misjudgment records of the same patient within a certain period of time, directly organize them into a list and push it to the second user. Before pushing, it can be classified according to the similarity of the historical ECG feature data. The historical ECG feature data of ECG event records in a category have a certain similarity. Then, the classified and organized ECG event records and standard defibrillation identification results are pushed to the doctor for the convenience of the second user to view.

[0128] The second user, based on their professional judgment, uses historical ECG feature data from the misjudged records and previous standard defibrillation identification results to determine whether the historical ECG feature data that caused the misjudgment has certain commonalities, and then determines whether supplementary heart rhythm identification rules need to be added.

[0129] The processing server obtains the heart rhythm recognition rules added by the second user. Specifically, it receives the ECG determination conditions and standard defibrillation recognition results of the new heart rhythm recognition rules added by the second user.

[0130] In another implementation, the processing device can automatically determine whether a new heart rhythm recognition rule needs to be added based on the misjudgment records, and after the need for adding a new heart rhythm recognition rule is met and confirmed by a second user, the new heart rhythm recognition rule is officially added. Specifically, such as... Figure 4 As shown, in step S023, the processing device is configured to obtain the heart rhythm recognition rule using the following steps:

[0131] S0231: The processing server obtains the historical ECG event records and the standard defibrillation identification results of the designated first user;

[0132] S0232: The processing server filters out misjudgment records where the second sub-defibrillation identification result of the medical device is inconsistent with the standard defibrillation identification result;

[0133] S0233: The processing server determines whether to generate pre-added heart rhythm recognition rules based on the historical ECG feature data corresponding to the misjudged record;

[0134] S0234: The processing server pushes the generated pre-added heart rhythm recognition rule to the second user, and when it receives the confirmation instruction from the second user, it uses the pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule; or when it receives the modification instruction from the second user, it uses the modified pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule.

[0135] Therefore, when the second user receives the pre-added heart rhythm recognition rule generated by the processing server, he / she can decide to allow the addition of this heart rhythm recognition rule, or if he / she thinks it is unnecessary, he / she can refuse to add this heart rhythm recognition rule, or if he / she thinks this heart rhythm recognition rule is inaccurate, he / she can modify it and add the heart rhythm recognition rule, such as modifying the feature value range of a feature in the ECG feature conditions, deleting or adding the feature value of a feature, etc.

[0136] In this embodiment, in step S0233, the processing server determines whether to generate pre-added heart rhythm recognition rules based on the historical ECG feature data corresponding to the misjudged record, including the following steps:

[0137] The processing server classifies the misjudged records according to the standard defibrillation identification results to obtain a first misjudged record that can be defibrillated and a second misjudged record that cannot be defibrillated. Here, the first misjudged record is all the misjudged records whose standard defibrillation identification results are defibrillable, and the second misjudged record is all the misjudged records whose standard defibrillation identification results are not defibrillable.

[0138] For the first misjudged record, the feature values ​​in the historical ECG feature data of all first misjudged records are arranged and combined. Here, the arrangement and combination can be exhaustively enumerated to exhaust all possible arrangements and combinations to obtain multiple first combinations. In addition, the first combination can also be a further combination of several combinations. For example, if there are three features a, b, and c, the first combination A is the combination of features a and b, the first combination B is the combination of features b and c, and the first combination C is the combination of the first combination A and the first combination B.

[0139] The correlation coefficient between each first combination in the first misjudged record and the misjudged result is calculated respectively. For a certain first combination A, the correlation coefficient is calculated as follows: the number of times the first combination A appears in all first misjudged records is divided by the number of times all first misjudged records are recorded to obtain the correlation coefficient between the first combination A and the misjudged result. The calculation method here is only an example. Other optional methods for calculating the correlation coefficient between combinations and misjudged results are also within the protection scope of this invention.

[0140] If at least one of the first combinations has a correlation coefficient greater than a preset correlation threshold (the specific value of the preset correlation threshold can be set as needed), a pre-added heart rhythm recognition rule is generated, the feature value of the first combination is used as the ECG feature condition, and the standard defibrillation recognition result is determined to be a defibrillable recognition result.

[0141] For the second misjudged record, the feature values ​​in the historical ECG feature data of all second misjudged records are arranged and combined. Here, the arrangement and combination can be exhaustively enumerated to exhaust all possible arrangements and combinations, resulting in multiple second combinations.

[0142] Calculate the correlation coefficient between each second combination and the misjudgment result in the second misjudgment record. For example, the correlation coefficient of a certain second combination B is calculated by dividing the number of times the second combination B appears in all second misjudgment records by the number of times all second misjudgment records appear.

[0143] If the correlation coefficient of at least one of the second combinations is greater than a preset correlation threshold (the specific value of the preset correlation threshold can be set as needed), a pre-added heart rhythm recognition rule is generated, the feature value of the second combination is used as the ECG feature condition, and the standard defibrillation recognition result is determined to be a non-defibrillable recognition result.

[0144] In another implementation, the processing server analyzes and summarizes the common features of misjudged records to determine whether to add heart rhythm recognition rules, which can be done using existing machine learning algorithms or data statistical methods. For example, a combination of feature values ​​can be used as input data, and the output data can be the probability of that combination causing a medical device to misjudge the record. If the probability is greater than a preset probability threshold, a pre-added heart rhythm recognition rule is generated.

[0145] like Figure 5 As shown, in this embodiment, step S200: based on the preset heart rhythm recognition algorithm and rule model library of the medical device, heart rhythm recognition is performed on the first electrocardiogram feature data to obtain the first heart rhythm recognition result of the medical device, including the following steps:

[0146] S210: Based on the first electrocardiogram feature data and the preset heart rhythm recognition algorithm, obtain the first sub-heart rhythm recognition result;

[0147] The preset heart rhythm recognition algorithm is the algorithm that is already fixed inside the medical device when it leaves the factory. Subsequent adjustments to the algorithm's architecture are only required via an external programmer to modify some parameters. This invention does not limit the specific algorithm type of the preset heart rhythm recognition algorithm. Depending on the needs of different medical devices at the time of manufacture, various existing heart rhythm recognition algorithms can be selected. For example, a heart rhythm recognition algorithm based on electrocardiogram (ECG) signal morphology analysis, or a heart rhythm recognition algorithm based on time-domain and frequency-domain analysis of ECG signals, or inputting ECG feature data into a trained convolutional neural network or binary classification network for heart rhythm recognition, etc. The algorithm can also be a combination of the above algorithms, or a combination of the above algorithms with other algorithms.

[0148] S220: Determine whether the first ECG feature data matches a certain heart rhythm recognition rule in the rule model library. If so, determine the second sub-heart rhythm recognition result based on the standard defibrillation recognition result of the matched heart rhythm recognition rule.

[0149] This rule model library is a preset model library interface in medical devices. It can be empty when it leaves the factory, that is, no heart rhythm recognition rules are set in it. In subsequent use, there may be cases where no heart rhythm recognition rules have been added, cases where a small number of heart rhythm recognition rules have been added, and cases where multiple heart rhythm recognition rules have been added.

[0150] The ECG feature conditions may include, for example, the numerical range conditions of one or more ECG features and / or the relationship between the numerical ranges of multiple ECG features; the numerical range conditions of an ECG feature may include the numerical range conditions of one ECG feature, for example, the ECG feature conditions of a heart rhythm recognition rule include the heart rate being between (x1, x2), and the values ​​of x1 and x2 can be selected and set as needed. The numerical range conditions for ECG features can also include a combination of two or more ECG feature numerical range conditions. For example, the ECG feature conditions for a rhythm recognition rule include a heart rate between (x1, x2) and a QRS pulse width between (y1, y2), where the values ​​of y1 and y2 can be selected and set as needed. The relationship between the numerical ranges of multiple ECG features can include, for example, the PP interval being m1 times the RR interval, or the value of feature a being a combination of the values ​​of feature b and feature c, where the value of m1 can also be selected and set as needed. The numerical range conditions of ECG features can also be combined with the relationship between the numerical ranges of multiple ECG features. For example, the ECG feature conditions for a rhythm recognition rule include a heart rate between (x1, x2) and a PP interval being m1 times the RR interval. This is merely an example of several rhythm recognition rule ECG feature conditions and is not intended to limit the scope of this invention. The rule model library can, for example, store each rhythm recognition rule in the form of a table, with each row representing a rhythm recognition rule, corresponding to a column of ECG feature conditions and a column of recognition judgment results.

[0151] When determining whether the first ECG feature data matches a certain heart rhythm recognition rule in the rule model library, it is determined whether the first ECG feature data conforms to the ECG feature conditions of a certain heart rhythm recognition rule in the rule model library; for example, the ECG feature conditions of a heart rhythm recognition rule include a heart rate between (x1, x2) and a QRS pulse width between (y1, y2), then it is determined whether the heart rate and QRS pulse width in the ECG feature data to be identified conform to the corresponding numerical range. If yes, the ECG feature data to be identified conforms to the heart rhythm recognition rule; if no, the ECG feature data to be identified does not conform to the heart rhythm recognition rule.

[0152] S230: Obtain the third weight value of the matched heart rhythm recognition rule as the third weight value of the second sub-heart rhythm recognition result, and calculate the fourth weight value of the first sub-heart rhythm recognition result based on the third weight value;

[0153] S240: The first sub-rhythm recognition result and the second sub-rhythm recognition result are weighted and averaged to obtain the first rhythm recognition result; for example, if the third weight value is k3, the fourth weight value is k4, the first sub-rhythm recognition result is n1, and the second sub-rhythm recognition result is n2, then the first rhythm recognition result m1 is (n1*k4+n2*k3) / (k3+k4).

[0154] In this embodiment, the first sub-rhythm identification result is a malignant arrhythmia risk value, for example, a value between 0 and 5. The smaller the value, the greater the probability of malignant arrhythmia requiring defibrillation. A value of 0 indicates the lowest probability of needing defibrillation, while a value of 5 indicates the highest probability. The second sub-rhythm identification result includes two cases: one is a malignant arrhythmia risk value corresponding to a defibrillable identification result, for example, 5; the other is a malignant arrhythmia risk value corresponding to a non-defibrillable identification result, for example, 0. Alternatively, when the first sub-rhythm identification result is 0 to other values ​​z, the second sub-rhythm identification result is the highest value z or 0. The value of z can be selected and set as needed, for example, 1, 10, 20, etc.

[0155] The fourth weight of the first sub-rhythm recognition result and the third weight of the second sub-rhythm recognition result determine the magnitude of the impact of the second sub-rhythm recognition result on the final malignant rhythm assessment result. This third weight can be set by the second user after evaluation, or it can be automatically set by the processing server based on the probability of misjudgment. Specifically, in this embodiment, after step S0234, following the step of using the pre-added rhythm recognition rule as the confirmed added rhythm recognition rule, the following steps are also included:

[0156] The processing server calculates the probability of misjudging historical ECG feature data that conforms to the confirmed added heart rhythm recognition rule based on the historical ECG feature data of the designated first user, the second defibrillation recognition result of the medical device, and the standard defibrillation recognition result. The probability is calculated, for example, by dividing the number of historical ECG event records that conform to the heart rhythm recognition rule by the total number of historical ECG event records that conform to the heart rhythm recognition rule.

[0157] The processing server sets a third weight value for the heart rhythm recognition rule based on the probability of misjudgment. The higher the probability of misjudgment, the larger the third weight value.

[0158] The sum of the third and fourth weight values ​​can be preset to 1 or other fixed values. After obtaining the third weight value, the fourth weight value can be calculated. For example, if the preset total weight is 1, and both the third and fourth weights are values ​​between 0 and 1, and the sum of the third and fourth weights is 1, then when the third weight is set to 0, the fourth weight is 1; when the third weight is set to 50%, the fourth weight is 50%; and when the third weight is set to 1, the fourth weight is 0.

[0159] For example, as illustrated above, the preset range for heart rhythm recognition results is 0-5. A result of 0 indicates the lowest risk of a malignant heart rhythm, while a result of 5 indicates the highest risk. For other values ​​between 0 and 5, the higher the score, the greater the risk of a malignant heart rhythm. The preset threshold for heart rhythm recognition results is 4. A score of 4 or higher indicates a malignant heart rhythm that can be defibrillated. Suppose that in a given recognition, the first sub-rhythm recognition result is 4.5, meaning the preset heart rhythm recognition algorithm identifies it as a defibrillable rhythm. The second sub-rhythm recognition result is 0, meaning the standard defibrillation recognition result is a non-defibrillable rhythm. If the weight of the second sub-rhythm recognition result is 0, then the first heart rhythm recognition result is 4.5, and the medical device's heart rhythm recognition result is a defibrillable rhythm, unaffected by the second sub-rhythm recognition result. If the weight of the second sub-rhythm recognition result is 50% and the weight of the first sub-rhythm recognition result is 50%, then the first rhythm recognition result is (4.5+0)*50%=2.25, which means that the first rhythm recognition result has become a non-defibrillable rhythm.

[0160] In this embodiment, in step S600, in the first implementation, during the second heart rhythm recognition, only the preset heart rhythm recognition algorithm of the medical device is used to perform heart rhythm recognition on the second-collected electrocardiogram feature data, that is, the risk assessment value obtained by the preset heart rhythm recognition algorithm is directly used as the second heart rhythm recognition result. In the second implementation, only the rule model library is used for the second heart rhythm recognition to determine whether the second electrocardiogram feature data matches a certain heart rhythm recognition rule in the rule model library. If so, the second heart rhythm recognition result is determined according to the standard defibrillation recognition result of the matched heart rhythm recognition rule. In the third implementation, the preset heart rhythm recognition algorithm and the rule model library are combined for heart rhythm recognition. It can be based on the same heart rhythm recognition method as step S200, but the recognition object changes from the first electrocardiogram feature data to the second electrocardiogram feature data.

[0161] Those skilled in the art will understand that various aspects of the present invention can be implemented as systems, methods, or program products. Therefore, various aspects of the present invention can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects.

[0162] like Figure 6 As shown, this embodiment of the invention also provides a heart rhythm recognition system, the system comprising:

[0163] The first acquisition module M100 is used to acquire the first electrocardiogram feature data of a specified first user;

[0164] The first judgment module M200 is used to perform heart rhythm recognition on the first electrocardiogram feature data based on the preset heart rhythm recognition algorithm and rule model library of the medical device, to obtain a first heart rhythm recognition result, and to obtain a first defibrillation recognition result based on the first heart rhythm recognition result. The rule model library includes heart rhythm recognition rules generated based on the historical electrocardiogram event records of the specified first user.

[0165] The second acquisition module M300 is used to acquire the second electrocardiogram feature data of the designated first user if the first defibrillation identification result is a defibrillable result.

[0166] The second judgment module M400 is used to perform heart rhythm recognition on the second electrocardiogram feature data based on the preset heart rhythm recognition algorithm of the medical device and / or the rule model library, to obtain the second heart rhythm recognition result of the medical device, and to obtain the second defibrillation recognition result based on the second heart rhythm recognition result or based on the first heart rhythm recognition result and the second heart rhythm recognition result.

[0167] This invention employs a heart rhythm recognition system. Through a first acquisition module M100 and a first judgment module M200, the system combines a user-specific rule model library and a preset heart rhythm recognition algorithm during the first heart rhythm recognition, improving the accuracy of the initial recognition. Only when the first judgment module M200 determines that defibrillation is necessary is a second heart rhythm recognition performed via a second acquisition module M300 and a second judgment module M400. This effectively avoids unnecessary second heart rhythm recognition and further improves accuracy. The rule model library stores heart rhythm recognition rules specific to each user, supplementing the universally used preset heart rhythm recognition algorithm in medical devices, effectively reducing the risk of misjudgment by medical devices for individual patients.

[0168] like Figure 7As shown, this embodiment of the invention also provides a medical device 1, including: a processor 11; a memory 12, storing the preset heart rhythm recognition algorithm, the rule model library, and executable instructions of the processor; and a communication circuit module 13 for communicating with the processing device; wherein the processor 11 is configured to execute the steps of the heart rhythm recognition method by executing the executable instructions. The medical device 1 can be an implantable cardioverter-defibrillator or other implantable cardiac monitoring / treatment device, or it can be an external cardiac monitoring / treatment device. The communication circuit module 13 can be a wireless communication circuit module and / or a wired communication circuit module. The processor 11 can be implemented using an MCU, and the memory 12 can be the MCU's own storage area, an external storage device, various ROM memories, various Flash memories, etc.

[0169] This invention, by pre-setting a rule model library in medical device 1, eliminates the need to update the pre-set heart rhythm recognition algorithm stored in the memory of medical device 1. Furthermore, the rule model library can strengthen the pre-set heart rhythm recognition algorithm, flexibly addressing algorithm upgrade needs after the medical device 1 is launched on the market. This improves the accuracy and safety of ECG monitoring, effectively avoiding the additional risks associated with modifying the pre-set heart rhythm recognition algorithm architecture. Supplementing the rule model library allows for flexible responses to specific patient situations, and the upgrade process is simple and quick. This rule model library is established based on physicians' evaluation of patients' individual ECG event records. It flexibly addresses the risk of algorithm misjudgment for individual patients by considering the unique characteristics and patterns of their heart rhythms. It can scientifically and effectively correct potential defects of general pre-set heart rhythm recognition algorithms for specific patients, significantly reducing the risk of false defibrillation and missed defibrillation for each patient.

[0170] like Figure 8 As shown, embodiments of the present invention also provide a medical system, including as follows: Figure 7The medical device 1 and the processing device 2 are shown. The processing device 2 includes a processing terminal 21 and a processing server 22. The processing terminal 21 is a terminal device capable of communicating with the medical device (wireless and / or wired communication), such as an external programmer. After receiving ECG event records from the medical device 1, the processing terminal 21 displays them to the doctor. The doctor can view the ECG event records, evaluate the heart rhythm recognition results of the medical device 1 in the ECG event records, and input the doctor's recognition result. Alternatively, the processing terminal 21 can send the ECG event records to the processing server 22, which then displays the ECG event records to the doctor and obtains the doctor's input recognition result. The processing server 22 can function as a database, storing ECG event records and doctor's recognition results for multiple patients, and classifying the stored data according to the patient. The processing server 22 pre-stores the association between patient IDs and medical device IDs. When the processing terminal 21 receives an ECG event record and a doctor's input recognition result from a medical device 1, it determines which patient's data the obtained ECG event record and doctor's recognition result belong to. This embodiment uses an implantable medical device as an example, and the communication circuit module of the medical device is a wireless communication circuit module. Its wireless communication methods include, but are not limited to, WIFI, Bluetooth, radio frequency, transceiver coils, etc. The processing device is an external programmable device, but the present invention is not limited thereto.

[0171] This invention also provides a computer-readable storage medium for storing a program that, when executed by a processor, implements the steps of the heart rhythm recognition method. In some possible implementations, various aspects of the invention can also be implemented as a program product comprising program code that, when executed on a terminal device, causes the terminal device to perform the steps described in the above-described heart rhythm recognition method section of this specification according to various exemplary embodiments of the invention.

[0172] This computer-readable storage medium can be executed on a medical device or other type of electronic device. The program product of this invention is not limited thereto; in this document, the readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0173] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0174] The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0175] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0176] When the program in the computer storage medium is executed by the processor, it implements the steps of the heart rhythm recognition method. Therefore, the computer storage medium can also achieve the technical effects of the heart rhythm recognition method.

[0177] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for recognizing heart rhythm, characterized in that, Includes the following steps: Obtain the first electrocardiogram (ECG) feature data of the specified first user; Based on the preset heart rhythm recognition algorithm and rule model library of the medical device, the first electrocardiogram feature data is used to perform heart rhythm recognition to obtain the first heart rhythm recognition result. The rule model library includes heart rhythm recognition rules generated based on the historical electrocardiogram event records of the designated first user, who is a user wearing the medical device. The first defibrillation identification result is obtained based on the first heart rhythm identification result; If the first defibrillation identification result is a defibrillable result, the second electrocardiogram feature data of the designated first user is obtained; Based on the preset heart rhythm recognition algorithm of the medical device, heart rhythm recognition is performed on the second electrocardiogram feature data to obtain the second heart rhythm recognition result of the medical device. Obtaining a second defibrillation recognition result based on the first heart rhythm recognition result and the second heart rhythm recognition result includes: obtaining a first weight of the first heart rhythm recognition result and a second weight of the second heart rhythm recognition result; averaging the first heart rhythm recognition result and the second heart rhythm recognition result to obtain a final heart rhythm recognition result; and obtaining a second defibrillation recognition result based on the final heart rhythm recognition result. The rule model library is formed in the following way: The historical ECG event records of the medical device are obtained. Each historical ECG event record includes the historical ECG feature data of the designated first user and the second sub-defibrillation identification result of the medical device. The second sub-defibrillation identification result is obtained after heart rhythm identification based on a preset heart rhythm identification algorithm and the second ECG feature data. The historical ECG feature data includes the ECG feature data of the first user at the time of the historical ECG event, the ECG feature data of the first user in the first time period before the time of the historical ECG event, and the ECG feature data of the first user in the second time period after the time of the historical ECG event. The historical ECG event records are sent to the processing device; The processing device receives heart rhythm recognition rules, each of which includes an electrocardiogram (ECG) characteristic condition and a standard defibrillation recognition result. The ECG characteristic condition is generated based on the historical ECG characteristic data of the designated first user. The heart rhythm recognition rules are stored in the rule model library of the medical device.

2. The heart rhythm recognition method according to claim 1, characterized in that, The processing device is configured to form or automatically update the heart rhythm recognition rules using the following steps: Receive the historical ECG event records; Obtain the standard defibrillation identification results corresponding to the historical ECG event records; The heart rhythm recognition rules are obtained based on the historical electrocardiogram event records and the corresponding standard defibrillation recognition results; The heart rhythm recognition rules are automatically updated each time a preset condition is met. The preset condition is one of the following: reaching a specific time, reaching a specific amount of data, or receiving a specific instruction.

3. The heart rhythm recognition method according to claim 2, characterized in that, The processing device includes a processing terminal and a processing server. The processing device obtains the standard defibrillation identification results corresponding to the historical electrocardiogram event records using the following steps: The processing terminal displays the received historical ECG event records, receives several defibrillation identification results input by several second users, generates a standard defibrillation identification result based on the several defibrillation identification results, and sends the historical ECG event records and the standard defibrillation identification result to the processing server; or, The processing terminal sends the received historical ECG event records to the processing server. The processing server pushes the received historical ECG event records to several second users, receives several defibrillation identification results input by several second users, generates a standard defibrillation identification result based on the several defibrillation identification results, and stores the historical ECG event records and the standard defibrillation identification results.

4. The heart rhythm recognition method according to claim 3, characterized in that, The process of generating a standard defibrillation identification result based on several defibrillation identification results includes: determining the defibrillation identification results whose proportion is greater than a preset proportion threshold as the standard defibrillation identification results.

5. The heart rhythm recognition method according to claim 3, characterized in that, The processing device is configured to acquire the heart rhythm recognition rules using the following steps: The processing server filters the historical ECG event records to obtain misjudgment records where the second sub-defibrillation identification result of the medical device is inconsistent with the standard defibrillation identification result. The processing server pushes the misjudgment record of the designated first user to the second user; The processing server obtains the heart rhythm recognition rules added by the second user.

6. The heart rhythm recognition method according to claim 4, characterized in that, The processing device is configured to acquire the heart rhythm recognition rules using the following steps: The processing server obtains the historical ECG event records and the standard defibrillation identification results of the designated first user; The processing server filters out misjudgment records where the second sub-defibrillation identification result of the medical device is inconsistent with the standard defibrillation identification result. The processing server determines whether to generate pre-added heart rhythm recognition rules based on the historical electrocardiogram feature data corresponding to the misjudged records. The processing server pushes the generated pre-added heart rhythm recognition rule to the second user, and upon receiving the second user's confirmation instruction, uses the pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule; or upon receiving the second user's modification instruction, uses the modified pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule.

7. The heart rhythm recognition method according to claim 6, characterized in that, The processing server determines whether to generate pre-added heart rhythm recognition rules based on the historical ECG feature data corresponding to the misjudged records, including the following steps: The processing server classifies the misjudged records according to the standard defibrillation identification results to obtain a first misjudged record that can be defibrillated and a second misjudged record that is not defibrillable. For the first misjudged record and the second misjudged record, the feature values ​​in the historical electrocardiogram feature data are arranged and combined respectively; Calculate the correlation coefficient between each combination and the misjudgment result; If the correlation coefficient of at least one of the combinations is greater than a preset correlation threshold, a pre-added heart rhythm recognition rule is generated, the feature value of the combination is used as the ECG feature condition, and the category of the standard defibrillation recognition result is determined according to whether the historical ECG feature data belongs to the first misjudged record or the second misjudged record.

8. The heart rhythm recognition method according to claim 6, characterized in that, After using the pre-added heart rhythm recognition rule as the confirmed heart rhythm recognition rule, the following steps are also included: The processing server calculates the probability of misjudging historical ECG feature data that conforms to the confirmed added heart rhythm recognition rules based on the historical ECG feature data of the designated first user, the second defibrillation recognition result of the medical device, and the standard defibrillation recognition result; The processing server sets a third weight value for the heart rhythm recognition rule based on the probability of misjudgment.

9. The heart rhythm recognition method according to claim 8, characterized in that, The preset heart rhythm recognition algorithm and rule model library based on medical devices are used to perform heart rhythm recognition on the first electrocardiogram feature data to obtain the first heart rhythm recognition result of the medical devices, including the following steps: Based on the first electrocardiogram feature data and the preset heart rhythm recognition algorithm, the first sub-heart rhythm recognition result is obtained; Determine whether the first ECG feature data matches a certain heart rhythm recognition rule in the rule model library. If so, determine the second sub-heart rhythm recognition result based on the standard defibrillation recognition result of the matched heart rhythm recognition rule. Obtain the third weight value of the matching heart rhythm recognition rule, use it as the third weight value of the second sub-heart rhythm recognition result, and calculate the fourth weight value of the first sub-heart rhythm recognition result based on the third weight value; The first sub-rhythm recognition result is obtained by weighting and averaging the first sub-rhythm recognition result and the second sub-rhythm recognition result.

10. The heart rhythm recognition method according to claim 1, characterized in that, The first electrocardiogram feature data includes at least one of real-time acquired electrocardiogram signal graphs and electrocardiogram feature parameters, and the second electrocardiogram feature data includes at least one of real-time acquired electrocardiogram signal graphs and electrocardiogram feature parameters.

11. A heart rhythm recognition system, characterized in that, The system includes: The first acquisition module is used to acquire the first electrocardiogram feature data of a specified first user; The first judgment module is used to perform heart rhythm recognition on the first electrocardiogram feature data based on the preset heart rhythm recognition algorithm and rule model library of the medical device, to obtain a first heart rhythm recognition result, and to obtain a first defibrillation recognition result based on the first heart rhythm recognition result. The rule model library includes heart rhythm recognition rules generated based on the historical electrocardiogram event records of the designated first user, where the first user is a user wearing the medical device. The second acquisition module is used to acquire the second electrocardiogram feature data of the designated first user if the first defibrillation identification result is a defibrillable result. The second judgment module is used to perform heart rhythm recognition on the second electrocardiogram feature data based on the preset heart rhythm recognition algorithm of the medical device to obtain the second heart rhythm recognition result of the medical device, and to obtain a second defibrillation recognition result based on the first heart rhythm recognition result and the second heart rhythm recognition result, including: obtaining a first weight of the first heart rhythm recognition result and a second weight of the second heart rhythm recognition result; weighting and averaging the first heart rhythm recognition result and the second heart rhythm recognition result to obtain a final heart rhythm recognition result; and obtaining a second defibrillation recognition result based on the final heart rhythm recognition result. The rule model library is formed in the following way: The historical ECG event records of the medical device are obtained. Each historical ECG event record includes the historical ECG feature data of the designated first user and the second sub-defibrillation identification result of the medical device. The second sub-defibrillation identification result is obtained after heart rhythm identification based on a preset heart rhythm identification algorithm and the second ECG feature data. The historical ECG feature data includes the ECG feature data of the first user at the time of the historical ECG event, the ECG feature data of the first user in the first time period before the time of the historical ECG event, and the ECG feature data of the first user in the second time period after the time of the historical ECG event. The historical ECG event records are sent to the processing device; The processing device receives heart rhythm recognition rules, each of which includes an electrocardiogram (ECG) characteristic condition and a standard defibrillation recognition result. The ECG characteristic condition is generated based on the historical ECG characteristic data of the designated first user. The heart rhythm recognition rules are stored in the rule model library of the medical device.

12. A medical device, characterized in that, include: processor; The memory stores a preset heart rhythm recognition algorithm, a rule model library, and executable instructions of the processor. The communication module is used to communicate with the processing device; The processor is configured to perform the steps of the heart rhythm recognition method according to any one of claims 1 to 10 by executing the executable instructions.

13. A medical system, characterized in that, include: A processing device is configured to receive historical electrocardiogram (ECG) event records from medical devices, obtain standard defibrillation identification results corresponding to the historical ECG event records, and obtain heart rhythm identification rules based on the historical ECG event records and the corresponding standard defibrillation identification results; and The medical device according to claim 12.

14. A computer-readable storage medium for storing a program, characterized in that, When the program is executed by the processor, it implements the steps of the heart rhythm recognition method according to any one of claims 1 to 10.

Images