Electrocardiogram lead pattern detection method, system, and storage medium

Abstract

The application relates to an electrocardio lead mode detection method, system and storage medium, wherein the method comprises the following steps: performing electrocardio feature extraction on current electrocardio data to obtain target signal features; the current electrocardio data is used to indicate original electrocardio data meeting a full connection state within a preset time; based on the target signal features and corresponding original electrocardio data, lead channel analysis is performed on the current electrocardio data to obtain a to-be-output lead mode; when the target signal features and the to-be-output lead mode meet a preset output condition, a first lead mode is output. Through the application, the target signal features in the current electrocardio data can be extracted, lead channel analysis is performed according to the target signal features, and the output of the lead mode is limited, so that the problem that the lead mode cannot be automatically detected is solved.

Description

Technical Field

[0001] The present application relates to the technical field of electrocardiogram (ECG) information processing, and in particular to an ECG lead pattern detection method, system, and storage medium. Background Art

[0002] ECG signals, which reflect the heart's electrical activity over time, are collected using skin electrodes. These electrodes are then connected to ECG lead wires to form different leads, transmitting the ECG signals to the ECG measurement circuit to produce an ECG. Skin electrodes are typically placed on various parts of the body, such as the limbs and chest. Commonly used types include single-lead, 3-lead, 12-lead, 18-lead, and 24-lead electrodes.

[0003] In different application scenarios, when users need to switch to different lead modes to collect ECG signals, they need to identify the lead mode accordingly. Currently, after replacing the electrode assembly, some lead modes are determined based on the user's pre-set information and the lead detachment status. This requires the user to set it in advance, resulting in the problem of being unable to automatically detect the lead mode.

[0004] With respect to the problem that the lead mode cannot be automatically detected in related technologies, no effective solution has been proposed so far. Summary of the Invention

[0005] In this embodiment, a method, system, and storage medium for detecting an electrocardiogram lead pattern are provided to solve the problem in related arts that lead patterns cannot be automatically detected.

[0006] In a first aspect, this embodiment provides an electrocardiogram (ECG) lead pattern detection method, comprising:

[0007] Extracting ECG features from the current ECG data to obtain target signal features; the current ECG data is used to indicate the original ECG data that meets the full connection state within the current preset time;

[0008] Based on the target signal characteristics and the corresponding original ECG data, performing lead channel analysis on the current ECG data to obtain a lead pattern to be output;

[0009] When the target signal characteristics and the lead pattern to be output meet a preset output condition, a first lead pattern is output.

[0010] In some embodiments, the step of performing lead channel analysis on the current ECG data based on the target signal characteristics and the corresponding original ECG data to obtain a lead pattern to be output includes:

[0011] Determining a number of signal channels corresponding to the original electrocardiogram data;

[0012] analyze similarity between each two of the signal channels based on the target signal feature and the corresponding original electrocardio data, to obtain a single-channel similarity result;

[0013] For any one of the signal channels in the plurality of signal channels: determine, based on the single-channel similarity result, a difference of the signal channel relative to all other signal channels, to obtain a difference degree score corresponding to the signal channel;

[0014] determine, based on the difference degree score corresponding to each of the signal channels, a to-be-output lead mode corresponding to the current electrocardio data.

[0015] In some embodiments, the determining, based on the difference degree score corresponding to each of the signal channels, the to-be-output lead mode corresponding to the current electrocardio data comprises:

[0016] determine, based on the difference degree score corresponding to each of the signal channels, a difference degree total score corresponding to all signal channels;

[0017] determine, based on a preset threshold value judging the difference degree total score, the to-be-output lead mode corresponding to the current electrocardio data.

[0018] In some embodiments, the outputting the first lead mode when the target signal feature and the to-be-output lead mode satisfy a preset output condition comprises:

[0019] adopt a corresponding output strategy based on a preset time period in which the current preset time is located; wherein,

[0020] in a first preset time period, if the target signal feature satisfies a first preset output condition and the to-be-output lead mode satisfies a second preset output condition, output the first lead mode based on the to-be-output lead mode; or,

[0021] in a second preset time period, if the to-be-output lead mode satisfies the second preset output condition, output the first lead mode based on the to-be-output lead mode.

[0022] In some embodiments, the above method further comprises:

[0023] if the current preset time satisfies a third preset time period when the target signal feature and the to-be-output lead mode do not satisfy the preset output condition, perform weighted processing on a lead mode of historical electrocardio data, and output a second lead mode.

[0024] In some embodiments, the extracting an electrocardio feature from the current electrocardio data to obtain a target signal feature comprises:

[0025] Preprocess the current electrocardio data to obtain first and second quasi-electrocardio data;

[0026] Extract features from the first quasi-electrocardio data to generate first target signal features;

[0027] Extract features from the second quasi-electrocardio data based on historical target signal features to generate second target signal features;

[0028] Detect the quality of the current electrocardio data based on the first and second target signal features;

[0029] Determine the first and second target signal features corresponding to the current electrocardio data that meets preset conditions in the quality detection result as target signal features.

[0030] In some embodiments, the preprocessing of the current electrocardio signal data to obtain first and second quasi-electrocardio data includes:

[0031] Perform pacing processing on the current electrocardio signal data to generate pacing processing data;

[0032] Perform baseline offset processing on the pacing processing data to obtain first and second preprocessed electrocardio data;

[0033] Perform noise reduction processing and data augmentation processing on the second preprocessed electrocardio data in sequence to obtain second quasi-electrocardio data.

[0034] In some embodiments, the above method further includes:

[0035] Collect original electrocardio data of each signal channel in the current preset time;

[0036] Determine the lead off state of the signal channel based on the original electrocardio data;

[0037] Determine the original electrocardio data in the full connection state based on the lead off state of the signal channel.

[0038] In a second aspect, an electrocardio lead mode detection system is provided in the present embodiment, which includes a feature extraction module, a lead detection module, and a lead output module;

[0039] The feature extraction module is configured to extract electrocardio features from current electrocardio data to obtain target signal features, wherein the current electrocardio data is used to indicate original electrocardio data in a current preset time that meets a full connection state;

[0040] The lead detection module is configured for performing lead channel analysis on the current electrocardio data based on the target signal feature and the corresponding original electrocardio data, to obtain a to-be-output lead mode.

[0041] The lead output module is configured for outputting a first lead mode when the target signal feature and the to-be-output lead mode satisfy a preset output condition.

[0042] In a third aspect, a storage medium is provided in the present embodiment, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the electrocardio lead mode detection method in the first aspect.

[0043] Compared with the related art, the electrocardio lead mode detection method, system and storage medium provided in the present embodiment can extract a target signal feature from current electrocardio data, perform lead channel analysis based on the target signal feature and corresponding original electrocardio data, and output a first lead mode when the target signal feature and the to-be-output lead mode satisfy a preset output condition. The present embodiment can extract a target signal feature from current electrocardio data, perform lead channel analysis based on the target signal feature, and limit the output of the lead mode, thereby solving the problem that the lead mode cannot be automatically detected.

[0044] The details of one or more embodiments of the present application are presented in the following drawings and description to make other features, objects and advantages of the present application more apparent. BRIEF DESCRIPTION OF DRAWINGS

[0045] The accompanying drawings illustrated herein are used to provide further understanding of the present application, and constitute a part of the present application. The illustrative embodiments of the present application and their description serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0046] Figure 1 FIG. 1 is a hardware structure block diagram of a terminal for an electrocardio lead mode detection method according to an embodiment;

[0047] Figure 2 FIG. 2 is a flowchart of an electrocardio lead mode detection method according to an embodiment;

[0048] Figure 3 FIG. 3 is a schematic diagram of signal channels in different lead modes according to an embodiment;

[0049] Figure 4 FIG. 4 is a schematic diagram of an electrocardio detection scene to which the electrocardio lead mode detection method according to an embodiment is applied;

[0050] Figure 5 FIG. 1 is a schematic diagram of an embodiment of a central electro feature extraction;

[0051] Figure 6 FIG. 2 is a schematic diagram of different time periods in an embodiment of a lead pattern detection process;

[0052] Figure 7 FIG. 3 is a flow chart of another embodiment of a central electro lead pattern detection method;

[0053] Figure 8 FIG. 4 is a structural block diagram of an embodiment of a central electro lead pattern detection system.

[0054] In the figure: 102, processor; 104, memory; 106, transmission device; 108, input and output device; 10, feature extraction module; 20, lead detection module; 30, lead output module. DETAILED DESCRIPTION

[0055] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and explained below in conjunction with the accompanying drawings and embodiments.

[0056] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the general meaning understood by a person with ordinary skill in the art to which the present application belongs. In the present application, "one", "a", "an", "the", "these" and similar words do not represent a quantitative limitation, but can be singular or plural. In the present application, the terms "include", "contain", "have" and any variants thereof have the purpose of covering non-exclusive inclusion; for example, a process, method and system, product or device containing a series of steps or modules (units) are not limited to the listed steps or modules (units), but can include steps or modules (units) not listed, or can include other steps or modules (units) inherent to the process, method, product or device. In the present application, the terms "connected", "connected", "coupled" and similar words do not limit to physical or mechanical connection, but can include electrical connection, whether direct or indirect. In the present application, "multiple" means two or more. The association between the associated objects is described by the term "and / or", which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. In general, the character " / " represents an "or" relationship between the associated objects. In the present application, the terms "first", "second", "third" and the like are only used to distinguish similar objects, and do not represent a specific order of the objects.

[0057] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. For example, when running on a terminal, the terminal can be a personal computer, a laptop, a smart phone, a tablet computer and a portable wearable device. Figure 1 FIG. 1 is a hardware structure diagram of a terminal of the ECG lead pattern detection method of this embodiment. Figure 1 As shown, the terminal may include one or more ( Figure 1 Only one is shown) a processor 102 and a memory 104 for storing data, wherein the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0058] Memory 104 can be used to store computer programs, such as software programs and modules of application software, such as the computer program corresponding to the ECG lead pattern detection method in this embodiment. Processor 102 executes the computer program stored in memory 104 to execute various functional applications and data processing, thereby implementing the above-mentioned method. Memory 104 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 may further include memory remotely located from processor 102, and such remote memory may be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0059] The transmission device 106 is used to receive or send data via a network. The network may include a wireless network provided by the terminal's telecommunications provider. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0060] The electrocardiosignal is a reaction of electrical activity of the heart in a period of time, collected by skin electrodes, and then the skin electrodes are connected with electrocardio lead wires to form different leads, transmit the electrocardiosignal to an electrocardio measurement circuit, and obtain an electrocardiogram. The skin electrodes are generally placed at various parts of limbs and chest, and commonly used are single lead, 3-lead, 12-lead, 18-lead or 24-lead, etc.

[0061] In different application scenarios, when the user needs to switch different lead modes for electrocardiosignal collection, the corresponding lead mode needs to be identified. At present, after replacing the electrode assembly, the switched lead mode is determined according to the user pre-setting information and the lead off state, which still needs the user to pre-set, resulting in the problem of unable to automatically detect the lead mode.

[0062] In the embodiment, an electrocardio lead mode detection method is provided, Figure 2 The flowchart of the electrocardio lead mode detection method in the embodiment is shown in Figure 2 The method comprises the following steps:

[0063] In step S210, electrocardio features of current electrocardio data are extracted to obtain target signal features; the current electrocardio data are used to indicate original electrocardio data meeting a full connection state in a current preset time.

[0064] In the lead mode detection process, the electrocardio data in the detection process is divided by a plurality of preset times, and the current electrocardio data are obtained in the current preset time.

[0065] Specifically, the dynamic original electrocardio data can be obtained in real time by a wearable electrocardio device and an integrated monitoring device, or the local electrocardio data are obtained as offline original electrocardio data. Correspondingly, the electrocardio lead mode detection in the application embodiment can be applied to electrocardio lead mode detection in various application scenarios, such as electrocardio lead mode switching and detection based on a wearable electrocardio device, switching and detection of an integrated monitoring device lead system, and electrocardio lead mode detection for local electrocardio data.

[0066] For the obtained original electrocardio data, the original electrocardio data meeting the full connection state are taken as the current electrocardio data after the lead off state is determined. The full connection state means that each signal channel in the original electrocardio data is in a connection state.

[0067] The current electrocardio data is preprocessed to obtain first and second quasi electrocardio data; the first and second quasi electrocardio data are respectively subjected to feature extraction to generate first and second target signal features. After the feature extraction of the current electrocardio data is completed, the quality of the current electrocardio data is detected. If the quality of the current electrocardio data does not meet a preset condition, the electrocardio data needs to be reacquired until the electrocardio data meets the preset condition. The first and second target signal features corresponding to the electrocardio data whose quality detection result meets the preset condition are determined as target signal features, and lead mode detection is further performed.

[0068] In step S220, lead channel analysis is performed on the current electrocardio data based on the target signal features and corresponding original electrocardio data to obtain a to-be-output lead mode.

[0069] In different lead modes, the connection mode of the electrode assembly is different, and the performance of the collected electrocardio data on each lead signal channel is also different. Figure 3 is a schematic diagram of the signal channel in different lead modes in this embodiment, Figure 3 In the diagram, there are 8 signal channels of lead mode A, lead mode B and lead mode C. The signal channel collected by Patch one-lead (patch type dynamic electrocardio Figure 1 lead) is shown in lead mode A; the signal channel collected by Patch three-lead (patch type dynamic electrocardio Figure 3 lead) is shown in lead mode B; and 8 signal channels collected by Holter twelve-lead (dynamic electrocardiogram 12-lead) are shown in lead mode C.

[0070] When the quality detection result of the current electrocardio data in the above steps meets the preset condition, lead mode analysis is performed on a plurality of signal channels corresponding to the original electrocardio data based on the target signal features and the corresponding original electrocardio data.

[0071] Specifically, the signal channels can be compared and analyzed according to specific parameters (such as wave peak, wave trough and RT wave) of the signal channels to obtain a to-be-output lead mode.

[0072] In addition, the signal channels can also be analyzed one-to-one from the perspective of similarity between the signal channels to obtain a single-channel similarity result. Then, based on the single-channel similarity result, the signal channels are analyzed one-to-many from the perspective of difference of any signal channel relative to all other signal channels, and the channel analysis results of all the signal channels are integrated to output a to-be-output lead mode corresponding to the current electrocardio data.

[0073] In step S230, the first lead mode is output when the target signal features and the to-be-output lead mode meet a preset output condition.

[0074] The obtained to-be-output lead mode is taken as a preliminary detection result, and whether to output the lead mode is judged according to the target signal feature and the to-be-output lead mode, and the to-be-output lead mode is output limited.

[0075] Specifically, for different time periods in the lead mode detection process, such as a first preset time period of unstable electrocardio data in the early detection stage, a second preset time period in the middle detection stage, and a third preset time period in the late detection stage, corresponding output strategies are respectively adopted, the to-be-output lead mode is taken as a first lead mode, or the to-be-output lead mode is processed and output, and the host computer system is used for data viewing and subsequent data analysis. The host computer system can be a terminal in the above embodiments, including but not limited to a personal computer, a notebook computer, a smart phone, a tablet computer and the like.

[0076] Figure 4 is a schematic diagram of an electrocardio detection scene to which the electrocardio lead mode detection method of the embodiment is applied, as shown in Figure 4 The signal source is a human electrocardio signal; the lower computer system is an electrocardio acquisition device for obtaining electrocardio data from the human electrocardio signal, which can be a wearable electrocardio device and an integrated monitoring device, etc., capable of supporting Holter 12-lead, Holter 3-lead, Patch 1-lead, Patch 3-lead and other lead systems; the host computer system covers lead mode detection, data viewing and data analysis functions, and the function carriers include but are not limited to Android system, IOS system, Windows system and Linux system, etc. In this scene, through the lead mode detection method, the host computer system can automatically identify different lead systems and perform electrocardio data viewing and data analysis.

[0077] The above steps extract the target signal feature in the current electrocardio data, analyze the lead channel according to the target signal feature, and limit the output of the lead mode. Compared with the prior art, the user does not need to configure the lead mode in advance, can analyze the lead channel in the full connection state according to the electrocardio data when the lead mode is switched, automatically detects and identifies the lead mode, reduces the operation difficulty, and can be applied to electrocardio lead mode detection in various application scenes.

[0078] With the progress of dynamic electrocardio technology, wearable electrocardio devices gradually develop towards small size and easy to wear. At present, there are many types of wearable electrocardio products, and the data acquisition box of the product corresponds to the electrode assembly one by one. Users often need to choose multiple types of products for use according to their needs, which greatly increases the operation difficulty of the users. In addition, the one-to-one structure also increases the design cost of wearable electrocardio products.

[0079] In view of the application scenario of the above-mentioned wearable ECG device, this embodiment provides a wearable ECG device, including: an electrode assembly and a data acquisition box; the electrode assembly and the data acquisition box are detachably connected through a unified interface.

[0080] Electrode assemblies include, but are not limited to, electrode pads and lead wire sets corresponding to lead systems such as Holter 12-lead, Holter 3-lead, Patch 1-lead, and Patch 3-lead. The data acquisition box is the hardware and software system for collecting human ECG signals. The data acquisition box and electrode assemblies operate in a one-to-many configuration, using a unified interface. A single data acquisition box can connect to multiple electrode assemblies.

[0081] By adopting the wearable ECG device in this embodiment, on the one hand, there is no need to design additional lead mode interfaces for the data acquisition box and the electrode assembly. Different types of electrode assemblies can be designed as a unified interface, which improves the adaptability of each component, reduces the design cost of each component, and broadens the usage scenarios of the product, thereby improving the applicability and application value of the product; on the other hand, it can automatically and quickly identify the lead mode when using the wearable ECG device, greatly improving the user experience.

[0082] In one embodiment, the above-mentioned current ECG data can be obtained by the following steps:

[0083] Collect the original ECG data of each signal channel within the current preset time; determine the lead-off status of the signal channel based on the original ECG data; and determine the original ECG data in the full connection state based on the lead-off status of the signal channel.

[0084] Here, dynamic raw ECG data is collected by an ECG acquisition device, or local ECG data is obtained as offline raw ECG data. ECG acquisition devices include but are not limited to wearable ECG devices and integrated monitoring devices.

[0085] The fully connected state means that all signal channels in the original ECG data are in a connected state, and the original ECG data of each signal channel within the current preset time is collected, wherein the current preset time can be determined by the sampling rate of the ECG data. For the obtained original ECG data, after the lead-off state is judged, the original ECG data that meets the fully connected state is used as the current ECG data. Since the signal channel in the lead-off state lacks ECG characteristics, the lead-off state can be judged by the lower computer such as the acquisition device based on the flag bit. For example, when acquiring the original ECG data, the acquisition device adds a signal channel for feedback of the lead-off state, and the lead-off state of the current ECG data can be directly obtained in the upper computer system.

[0086] If the lead is off, an alarm will be output and the user will be prompted to adjust and re-collect ECG data until the electrode assembly is in normal contact with the human skin and the original ECG data in a fully connected state is obtained.

[0087] By determining the lead-off status of the signal channel based on the raw ECG data in this embodiment and determining the raw ECG data in the fully connected state, the current ECG data within the preset time period can be obtained. This can provide the current ECG data in the fully connected state for subsequent lead channel analysis. It can also standardize lead mode detection scenarios and reduce the probability of lead mode judgment errors due to improper operation or electrode material problems. Compared to the method of using the lead-off status to determine the lead mode in related technologies, this method separates the determination of the lead-off status from the lead mode judgment logic, making the detection results unaffected by the user's wearing method and improving the consistency of the detection results.

[0088] In one embodiment, Figure 5 is a schematic diagram of ECG feature extraction in this embodiment, as shown in Figure 5 As shown, the upstream input of ECG feature extraction is the raw ECG data collected by the lower-level system. After determining the lead-off status, the current ECG data in the fully connected state is obtained. The output information of ECG feature extraction is the target signal feature, and the downstream output is the lead mode detection. During ECG feature extraction, the ECG data quality is also judged. If the quality meets the preset conditions, the target signal feature is output and transmitted downstream for lead mode detection. If the quality does not meet the preset conditions, the lower-level system will wait for new ECG data to be transmitted.

[0089] In this embodiment, the specific steps of extracting ECG features and obtaining target signal features are as follows:

[0090] Step S211 , preprocessing the current ECG data to obtain first quasi-ECG data and second quasi-ECG data.

[0091] When the current ECG data is acquired in a fully connected state and the length of the current ECG data meets the requirements, the current ECG data can be preprocessed and ECG features extracted. Specifically, the data length requirement is a requirement for communication with the lower computer.

[0092] In one embodiment, a multi-feature fusion method is designed to pre-process the ECG data, which can maximize the extraction of effective information in the ECG data. Figure 5 As shown, the current ECG signal data is paced to generate paced data; the paced data is baseline offset to obtain first quasi-ECG data and second pre-processed ECG data; the second pre-processed ECG data is noise-reduction processed and data amplification processed in turn to obtain second quasi-ECG data.

[0093] The pacing processing includes a pacing spike detection and filling operation, and generates pacing processing data. The pacing processing avoids the influence of high pulses on the current electrocardio data. The pacing processing data is subjected to baseline offset processing. The baseline offset processing is used for correcting the up and down fluctuations of the electrocardio data signal caused by the chest breathing movement and the slight movement of the electrode, and ensures the consistency of the data of each signal channel, so as to obtain first quasi-electrocardio data and second preprocessed electrocardio data.

[0094] Further, the second preprocessed electrocardio data is subjected to noise reduction processing and data augmentation processing to obtain second quasi-electrocardio data. The noise reduction processing is used for identifying and processing various noises in the electrocardio data. The data augmentation is used for multiplying the data by a plurality of times when the first data is input. Since the electrocardio data in a current preset time is insufficient for feature extraction, some historical database needs to be matched. The database is filled in advance by the augmentation, so that the data length required for subsequent feature extraction of the second quasi-electrocardio data can be met, and the feature extraction can be quickly performed. The electrocardio data in the database is updated each time the feature extraction is performed.

[0095] In step S212, the first quasi-electrocardio data is subjected to feature extraction to generate first target signal features.

[0096] As shown in Figure 5 The time domain signal processing method is used to obtain mean value and amplitude and other data features of the first quasi-electrocardio data to generate the first target signal features.

[0097] In step S213, the second quasi-electrocardio data is subjected to feature extraction based on the historical target signal features to generate second target signal features.

[0098] As shown in Figure 5 The peak detection method is used to obtain P wave (waveform generated when electrocardio activity is conducted to atrium), QRS wave (process of ventricular depolarization), T wave (reflects the change of potential and time in the process of ventricular late repolarization), RR interval (time from atrial activation start to ventricular start activation), QT (from QRS wave starting point to T wave ending point, reflects the start of ventricular repolarization to the end of repolarization), ST (period from QRS terminal to T wave starting point) and other characteristic information of the current electrocardio data as the second target signal features. Since the electrocardio data in a current preset time is insufficient for feature extraction, the historical target signal features in the historical database need to be referred to for feature extraction of the current electrocardio data.

[0099] In step S214, the quality of the current electrocardio data is detected based on the first target signal features and the second target signal features.

[0100] As shown in Figure 5As shown, based on the first target signal feature and the second target signal feature, the quality of each signal channel of the current electrocardio data is detected, and finally the quality of the current electrocardio data is determined according to the quality of each signal channel.

[0101] The following gives a method for representing the quality detection result of the current electrocardio data:

[0102] Q(A) = f Q (Feature1, Feature2, …, FeatureM);

[0103] Q(All) = f All (Q(A), Q(B), …, Q(N));

[0104] Wherein, f Q represents the operation of the signal channel quality; f All represents the operation of the current electrocardio data quality; Feature represents the first target signal feature and the second target signal feature, and the number is M; Q(A), Q(B), …, Q(N) respectively represent the quality of signal channels A, B … in the current electrocardio data, and there are N signal channels; Q(All) represents the final quality of the entire current electrocardio data.

[0105] When detecting the quality of each signal channel of the current electrocardio data based on the first target signal feature and the second target signal feature, the first target signal feature and the second target signal feature can be compared with the corresponding threshold value to determine whether the quality of the signal channel meets the requirements, and finally the number of signal channels meeting the requirements is obtained. Among them, the comparison of parameters such as QRS wave, RR interval, heart rate, amplitude and signal-to-noise ratio can be specifically selected. Further determine whether the number of signal channels meeting the requirements in the current electrocardio data meets the quality requirements, if it does, then Q(All) value is 1; otherwise, Q(All) value is 0.

[0106] Exemplarily, the number of features M is set to 2, Feature1 represents the number of R peaks in the current electrocardio data, and Feature2 represents the RR interval. If the number of R peaks is greater than the corresponding threshold value, and the difference between each RR interval and the average RR interval is less than the corresponding threshold value, then Q(A) is equal to 1, and the quality of signal channel A meets the requirements, otherwise Q(A) is equal to 0, and the quality of signal channel A does not meet the requirements.

[0107] In step S215, the first target signal feature and the second target signal feature corresponding to the current electrocardio data meeting the preset condition of the quality detection result are determined as the target signal feature.

[0108] The first target signal feature and the second target signal feature corresponding to the current electrocardio data satisfying the preset condition in the quality detection result are determined as the target signal feature.

[0109] The output target signal feature is transmitted to the downstream for lead mode detection. If the quality does not satisfy the preset condition, new electrocardio data is re-transmitted from the lower computer system for waiting.

[0110] In the embodiment, after the current electrocardio data is preprocessed, the feature extraction is respectively performed to generate the first target signal feature and the second target signal feature. After the feature extraction of the electrocardio signal is completed, the quality of the current electrocardio data is judged. If the quality of the current electrocardio data does not satisfy the preset condition, the electrocardio data is re-acquired until the electrocardio data satisfies the preset condition. The first target signal feature and the second target signal feature corresponding to the electrocardio data satisfying the preset condition in the quality detection result are determined as the target signal feature, and the lead mode detection is further performed. The data fluctuation caused by the action or other interference can be effectively reduced, and the detection accuracy is improved.

[0111] In one embodiment, the lead channel analysis on the current electrocardio data based on the target signal feature and the corresponding original electrocardio data to obtain the to-be-output lead mode in the step S220 can be realized by the following steps.

[0112] In the step S221, the number of signal channels corresponding to the original electrocardio data is determined.

[0113] In the step S221, the number of signal channels corresponding to the original electrocardio data is determined.

[0114] In the step S222, the similarity between each two signal channels is analyzed based on the target signal feature and the corresponding original electrocardio data to obtain a single-channel similarity result.

[0115] In the step S222, the similarity between each two signal channels is analyzed based on the target signal feature and the corresponding original electrocardio data to obtain a single-channel similarity result. Specifically, the similarity between each two signal channels is analyzed according to the data of the original electrocardio data in each signal channel and the feature information of the target signal feature in each signal channel. For example, for the signal channels A, B, C and D, the similarity analysis between each two signal channels means that the similarity between the signal channels AB, AC and AD is analyzed respectively.

[0116] In practical applications, some target signal features can be selected for similarity analysis between every two signal channels, such as R peak, RR interval, heart rate, and signal amplitude.

[0117] For example, the following gives a single channel similarity result eChan(A) between two signal channels A and B. B ) is calculated as follows:

[0118] eChan(A B )=E S (ichan(A),ichan(B),ibeat(A),iarrRpeak(A));

[0119] Where ichan(A) and ichan(B) represent the original ECG data in signal channels A and B, respectively; ibeat(A) represents the number of heartbeats in signal channel A; iarrRpeak(A) represents the R peak position in signal channel A; E S Represents the operation of the similarity between signal channels A and B. It should be noted that ibeat(A) and iarrRpeak(A) in the above calculation method can also be replaced by other target signal features.

[0120] The following continues to give the above E S A specific calculation method:

[0121] When ibeat(A)>0, that is, when a heartbeat is detected in signal channel A:

[0122]

[0123] Among them, N represents the range of sites before and after the selected R peak, and M The operation selects the average difference in the raw ECG data within the current preset time period; [-1] represents the horizontal coordinate of the R-peak location. Since ECG data can vary significantly near the R-peak when a heartbeat is detected, the average of this range is taken. When the average difference exceeds a threshold, a single-channel similarity result is obtained, indicating that signal channel A is dissimilar to signal channel B.

[0124] When ibeat(A)=0, that is, no heartbeat is detected in signal channel A:

[0125]

[0126]

[0127]

[0128] wherein n represents the abscissa of the original electrocardio data, i.e. the sampling point, f A ichan(A)(n) represents the Sigmoid mapping of the electrocardio signal data of signal channel A, f B ichan(B)(n) represents the Sigmoid mapping of the electrocardio signal data of signal channel B; m represents the amount of sliding, i.e. the deviation of the sampling point movement; represents the mapped cross-correlation coefficient. The Sigmoid function has the characteristics of smoothness and monotonic boundedness, and the Sigmoid mapping of the electrocardio signal can suppress impulse noise and preserve the original information of the signal to be detected. Then, the correlation function is calculated for the mapped signal, and the similarity information of the electrocardio data signal can be obtained from the peak value information of the correlation function.

[0129] The range of m is set to [-d, d], and the resolution is set to 1. The value of R AB (m) corresponding to each value of m in the set range is calculated, and the maximum value in the R AB (m) array is calculated. If there is an obvious maximum value, it is defined that m=D at this time, i.e. when m=D, R AB (m) has a maximum value, which means that ichan(A) and ichan(B) have strong similarity at this time, i.e. it is considered that signal channel A is similar to signal channel B. If there is no obvious maximum value, it means that ichan(A) and ichan(B) do not have strong similarity, and the single-channel similarity result that signal channel A is not similar to signal channel B is obtained.

[0130] Step S223, for any signal channel in the plurality of signal channels: based on the single-channel similarity result, determining the difference of the signal channel with respect to all other signal channels, to obtain a difference degree score corresponding to the signal channel.

[0131] wherein based on the single-channel similarity result, a single-to-multiple analysis is performed on the signal channel from the perspective of the difference of any signal channel with respect to all other signal channels. For example, for signal channels A, B, C, and D, the difference degree score of any signal channel in the plurality of signal channels with respect to all other signal channels is obtained. Specifically, different analysis weights are applied to each single-channel similarity result to obtain a total difference degree score of the current electrocardio data.

[0132] For example, the following gives a representation method of the difference degree score EChan(A) of signal channel A with respect to all other signal channels:

[0133] EChan(A) = E N (eChan(A B ), eChan(AC ),…,eChan(A N ),iarrWeight(A));

[0134] where iarrWeight(A) represents the analysis weight between signal channel A and other channels B, C, …; eChan(A) represents the difference score of signal channel A relative to all channels; e N represents the operation of the difference score. The analysis weight can be determined by different electrocardio detection scene logic and channel correlation known by priori knowledge.

[0135] The following gives a specific calculation method of e N for eight selected signal channels (I, II, V1-V6) in 12-lead:

[0136]

[0137] where iarrWeight(I II ) represents the analysis weight between I channel and II channel; the analysis weight between I channel and II, V1-V6 channels is [1, 0, 0, 0, 0, 0, 0], and S is the total score e N of I channel relative to all channels. When the score is less than a threshold value, it can be considered that there is a difference between I channel and other related channels.

[0138] In step S224, the lead mode to be output corresponding to the current electrocardio data is determined based on the difference score corresponding to each signal channel.

[0139] where the total difference score corresponding to all signal channels is determined based on the difference score corresponding to each signal channel; the lead mode to be output corresponding to the current electrocardio data is determined based on the preset threshold value for judging the total difference score.

[0140] Specifically, the difference score corresponding to each signal channel is counted to determine the total difference score corresponding to all signal channels. The lead mode to be output is finally determined based on the preset threshold value for judging the total difference score. The following gives a representation method of the lead mode LeadMode:

[0141] LeadMode=f(EChan(A),EChan(B),EChan(C),…,EChan(N));

[0142] where f represents the operation of determining the total difference score according to the difference score corresponding to each signal channel.

[0143] In actual application, multiple lead mode categories can be detected according to all signal channels in full connection state.

[0144] Further, in order to avoid the complexity or redundancy of the judgment logic, the judgment logic of the to-be-output lead mode is determined according to the category of the to-be-detected lead mode. For example, when the lead mode detection category includes Holter 12 leads and Patch 3 leads, the following is a kind of lead mode judgment logic:

[0145]

[0146] Wherein, EChan(v1)~EChan(v6) represent the difference degree scores of V1~V6 channels obtained in the above steps. By counting the difference degree scores and outputting the total difference degree score, if it is greater than a threshold value, it is considered that the signal channels in the range have differences, and the to-be-output lead mode is determined as Holter 12 leads. If it is less than the threshold value, it is considered that the signal channels in the range have no differences, and the to-be-output lead mode is determined as Patch 3 leads.

[0147] In the present example, the adopted basic system is a conventional 12-lead system, including I, II, III, avL, avF, avR, V1, V2, V3, V4, V5, V6, a total of 12 lead channels. I, II, III leads are standard bipolar limb leads, I lead is the potential difference between the left and right hands, II lead is the potential difference between the left leg and the right hand, III lead is the potential difference between the left leg and the left hand, and III lead can be calculated from I and II leads. avL, avF, avR leads are augmented unipolar limb leads, which are used to directly record the voltage at the lead site. V1, V2, V3, V4, V5, V6 leads are chest leads, which are used to record the actual potential of each part of the heart under the electrode. Limb leads observe the changes of electrocardiogram from the front (up, down, left and right directions), and chest leads observe the changes of electrocardiogram from the transverse section (assuming that the heart is cut into a transverse section at a certain horizontal plane). Based on 12-lead channels, the signals collected by different electrode assemblies and lead modes have different performances on 12-lead channels. In order to simplify the judgment logic, eight channels of I, II, V1, V2, V3, V4, V5, V6 are used for lead mode analysis and judgment in the present example.

[0148] Similarly, if the Patch 1 lead and the Patch 3 lead are classified, only the difference degree scores of I and II channels need to be statistically analyzed.

[0149] In this embodiment, the single-to-single analysis of the signal channels is performed from the perspective of the similarity between the signal channels, and the single-channel similarity result is obtained. Then, based on the single-channel similarity result, the single-to-multiple analysis of the signal channels is performed from the perspective of the difference of any signal channel relative to all other signal channels, and the channel analysis results of all signal channels are integrated to output the lead mode to be output corresponding to the current electrocardio data. Different judgment logics are adopted in different application scenarios, which can simplify the algorithm steps, improve the algorithm efficiency, and enable accurate lead mode detection and judgment.

[0150] The obtained lead mode to be output is taken as the preliminary detection result, and the output of the lead mode is judged based on the target signal feature and the lead mode to be output, and the output of the lead mode to be output is limited. In one embodiment, when the target signal feature and the lead mode to be output satisfy the preset output condition in step S230, the first lead mode is output, and the corresponding output strategy can be adopted based on the preset time period in which the current preset time is located.

[0151] In the first preset time period, if the target signal feature satisfies the first preset output condition and the lead mode to be output satisfies the second preset output condition, the first lead mode is output based on the lead mode to be output.

[0152] Alternatively, in the second preset time period, if the lead mode to be output satisfies the second preset output condition, the first lead mode is output based on the lead mode to be output.

[0153] In addition, when the target signal feature and the lead mode to be output do not satisfy the preset output condition, if the current preset time satisfies the third preset time period, the lead mode of the historical electrocardio data is weighted and processed, and the second lead mode is output.

[0154] Wherein, Figure 6 is a schematic diagram of different time periods in the lead mode detection process in this embodiment, as shown in Figure 6 Different output strategies are adopted for different time periods in the lead mode detection process, and the lead mode (first lead mode or second lead mode) is output according to the lead mode to be output. For example, the first preset time period for detecting the unstable electrocardio data in the early stage, the second preset time period in the middle stage of detection, and the third preset time period in the late stage of detection.

[0155] In the first preset time period, it is judged whether the target signal feature meets the first preset output condition and whether the to-be-output lead mode meets the second preset output condition. According to the comparison result of the target signal feature and the corresponding threshold, it is judged whether the target signal feature meets the first preset output condition, for example, the number of detected R peaks in the current electrocardio data is less than 1 or the amplitude of the detected R peak is less than a threshold, and then it is judged that the target signal feature does not meet the first preset output condition.

[0156] The consistency principle is used to judge whether the to-be-output lead mode meets the second preset output condition. Specifically, after obtaining the to-be-output lead mode, the to-be-output lead mode results of multiple pieces of electrocardio data are continuously analyzed. If the to-be-output lead mode is consistent within a specified time range, the second preset output condition is met; otherwise, the second preset output condition is not met.

[0157] If the target signal feature meets the first preset output condition and the to-be-output lead mode meets the second preset output condition, the to-be-output lead mode is output as the first lead mode; otherwise, no result is output.

[0158] Since the electrocardio data is unstable in the first preset time period at the beginning of the detection period, some interference will be caused. By using the first preset output condition and the second preset output condition, the interference influence can be reduced and the accuracy of the lead mode detection can be improved.

[0159] In the second preset time period, the to-be-output lead mode is judged according to the second preset output condition. If the second preset output condition is met, the to-be-output lead mode is output as the first lead mode; otherwise, no result is output. In this way, the influence of excessive signal fluctuation on the lead mode detection can be avoided.

[0160] In the third preset time period, the second lead mode is output based on the timeliness principle. Specifically, if an effective lead mode cannot be output within a specified time limit, the historical electrocardio data is weighted based on the proximity principle after a certain time range is exceeded (for example, after the first preset time period and the second preset time period are exceeded), and the second lead mode is output after the weighted processing.

[0161] The following lead mode detection categories include Holter 12 leads and Patch 3 leads as examples, and a calculation method of the second lead mode is given:

[0162] Suppose that after a set time T, the to-be-output lead mode is Holter 12 leads or Patch 3 leads. First, N to-be-output lead modes before the time point T are obtained, which include a default corresponding value -1, a Holter 12 lead corresponding value 12, and a Patch 3 lead corresponding value 3. Then, the N to-be-output lead modes L1~L NSet the weight, based on the principle of proximity, the closer to the time point T, the greater the weight corresponding to the result, L1 is the farthest in time, set the weight as W1 = 1 + 1 / N, L N is the closest in time, set the weight as W N = 1 + N / N; Next, according to the N output lead mode and the corresponding weight, the scores of Holter 12 lead and Patch 3 lead are calculated, the score of Holter 12 lead is Where a~g represents the result Holter 12 lead corresponding to the result subscript, the score of Patch 3 lead is Where h~n represents the result Patch 3 lead corresponding to the result subscript. If Q 12 ≥ Q3, the output result is Holter 12 lead, and a prompt signal is output to indicate that the lead mode needs to be confirmed, but it does not affect the subsequent operation of the host computer system; if Q 12 < Q3, the output result is Patch 3 lead, and a prompt signal is output to indicate that the lead mode needs to be confirmed, which also does not affect the subsequent operation of the host computer system.

[0163] The above-mentioned time-based principle outputs the second lead mode when the detection time exceeds the appropriate detection time, which can avoid infinite loop of logic, ensure output of results within a certain time, and output a prompt signal, which can improve the stability of lead mode detection.

[0164] Figure 7 is the flow chart of the central lead mode detection method of the embodiment, as Figure 7 shown, the method comprises the following steps:

[0165] Step S710, collect the original electrocardio data of each signal channel in the current preset time, and judge the lead off state of the signal channel to determine the original electrocardio data in the full connection state.

[0166] Step S720, pre-process the current electrocardio data to obtain first and second quasi-electrocardio data; extract features from the first quasi-electrocardio data to generate first target signal features; extract features from the second quasi-electrocardio data based on historical target signal features to generate second target signal features.

[0167] Step S730, determine the first target signal features and the second target signal features corresponding to the current electrocardio data whose quality detection result meets the preset condition as target signal features.

[0168] Step S740, determine a plurality of signal channels corresponding to the original electrocardio data; analyze the similarity between each two signal channels based on the target signal features and the corresponding original electrocardio data to obtain single-channel similarity results.

[0169] In step S750, for any one of the plurality of signal channels, based on the single-channel similarity result, a difference of the signal channel relative to all other signal channels is determined to obtain a difference degree score corresponding to the signal channel.

[0170] In step S760, based on the difference degree score corresponding to each signal channel, a lead mode to be output corresponding to the current electrocardio data is determined.

[0171] In step S770, based on a preset time period in which the current preset time is located, a corresponding output strategy is adopted to output the lead mode.

[0172] It should be understood that, although each step in the flowchart involved in each of the above-described embodiments is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each of the above-described embodiments can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0173] Based on the same inventive concept, the embodiments of the present application also provide an electrocardio lead mode detection system for implementing the above-mentioned electrocardio lead mode detection method. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme described in the above method, and therefore the specific limitations in one or more electrocardio lead mode detection system embodiments provided below can refer to the limitations of the electrocardio lead mode detection method described above, which will not be described here again.

[0174] Figure 8 is a structural block diagram of the electrocardio lead mode detection system of the present embodiment, as shown in the figure, the system includes a feature extraction module 10, a lead detection module 20, and a lead output module 30. Figure 8

[0175] The feature extraction module 10 is configured to perform electrocardio feature extraction on the current electrocardio data to obtain target signal features; the current electrocardio data is used to indicate original electrocardio data satisfying the full connection state within the current preset time.

[0176] The lead detection module 20 is configured to perform lead channel analysis on the current electrocardio data based on the target signal features and the corresponding original electrocardio data to obtain a lead mode to be output.

[0177] ​The lead output module 30 is configured to output the first lead mode when the target signal feature and the to-be-output lead mode meet a preset output condition.

[0178] It should be noted that each of the above modules can be a functional module or a program module, and can be implemented by software or hardware. For the modules implemented by hardware, each of the above modules can be located in the same processor of the computer device, or each of the above modules can be located in different processors of the computer device in any combination.

[0179] Through the system provided in the embodiment, the target signal feature in the current electrocardio data is extracted, the lead channel is analyzed according to the target signal feature, and the output of the lead mode is limited. Compared with the prior art, the user does not need to configure the lead mode in advance, the lead channel in the full connection state can be analyzed according to the electrocardio data when the lead mode is switched, the lead mode is automatically detected and recognized, the operation difficulty is reduced, and the electrocardio lead mode detection in various application scenarios can be applied.

[0180] In the embodiment, a computer device is also provided, which includes a memory and a processor. The memory stores a computer program. The processor is configured to execute the computer program to perform the steps in any of the above method embodiments.

[0181] Optionally, the computer device can further include a transmission device and an input / output device, wherein the transmission device is connected with the processor, and the input / output device is connected with the processor.

[0182] It should be noted that the specific examples in the embodiment can refer to the examples described in the above embodiments and optional implementation manners, which will not be described herein again.

[0183] In addition, in combination with the electrocardio lead mode detection method provided in the above embodiments, a storage medium can also be provided to implement the electrocardio lead mode detection method in the embodiment. The storage medium stores a computer program. When the computer program is executed by a processor, any of the electrocardio lead mode detection methods in the above embodiments is implemented.

[0184] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties.

[0185] It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not to be taken in a limiting sense. All other embodiments falling within the scope of the application are contemplated and are within the scope of the application.

[0186] It is apparent that the drawings depicted are only a few examples of the application and that many other embodiments of the application can be made without departing from the scope of the application disclosed herein. Furthermore, it should be understood that the drawings and detailed description thereto are not indicative of every possible embodiment of the application. In fact, many modifications and variations to the application disclosed herein will occur to those skilled in the art, once advised of the application disclosed herein. All such modifications and variations are believed to be within the scope of the application and are intended to be within the scope of the application.

[0187] The word "example" is used herein to mean serving as an example, instance, or illustration. Any aspect or embodiment described herein as "example" is not necessarily to be construed as preferred or advantageous over other aspects or embodiments. The disclosure herein using "example" terminology is to be understood that at least one aspect or embodiment described herein is in a way an example, and is used to elucidate a specific implementation of the application. A single feature of a example implementation cannot, therefore, be necessarily interpreted as an example of all, but rather an example of at least one of the aspects or embodiments described herein.

[0188] The above-described embodiments are merely illustrative of several embodiments of the application and do not limit the scope of the application. It is apparent that many modifications and improvements can be made to the embodiments without departing from the scope of the application. Accordingly, the scope of the application should be determined by the appended claims.

Claims

1. A method for detecting an electrocardiogram lead pattern, characterized in that: include: Extracting ECG features from the current ECG data to obtain target signal features; the current ECG data is used to indicate the original ECG data that meets the full connection state within the current preset time; Determining a number of signal channels corresponding to the original electrocardiogram data; Based on the target signal features and the corresponding original ECG data, the similarity between each two signal channels is analyzed to obtain a single-channel similarity result; For any one of the plurality of signal channels: determining, based on the single-channel similarity result, the difference of the signal channel relative to all other signal channels, and obtaining a difference score corresponding to the signal channel; Determining a lead mode to be output corresponding to the current electrocardiogram data based on the difference score corresponding to each of the signal channels; When the target signal characteristics and the lead pattern to be output meet a preset output condition, a first lead pattern is output.

2. The method according to claim 1, characterized in that The step of determining the lead mode to be output corresponding to the current electrocardiogram data based on the difference score corresponding to each signal channel comprises: Determining a total difference score corresponding to all signal channels based on the difference score corresponding to each signal channel; The total difference score is judged based on a preset threshold to determine the lead mode to be output corresponding to the current electrocardiogram data.

3. The method according to claim 1, characterized in that The method of outputting a first lead pattern when the target signal characteristic and the lead pattern to be output meet a preset output condition comprises: Based on the preset time period where the current preset time is located, a corresponding output strategy is adopted; wherein, Within a first preset time period, if the target signal characteristic satisfies a first preset output condition and the lead pattern to be output satisfies a second preset output condition, then outputting the first lead pattern based on the lead pattern to be output; or Within a second preset time period, if the lead pattern to be output meets the second preset output condition, the first lead pattern is output based on the lead pattern to be output.

4. The method according to claim 1, wherein Also includes: When the target signal feature and the lead pattern to be output do not meet the preset output condition, if the current preset time meets the third preset time period, the lead pattern of the historical ECG data is weighted and a second lead pattern is output.

5. The method according to claim 1, wherein Extracting ECG features from the current ECG data to obtain target signal features includes: Preprocessing the current ECG data to obtain first quasi-ECG data and second quasi-ECG data; performing feature extraction on the first quasi-ECG data to generate a first target signal feature; performing feature extraction on the second quasi-ECG data based on historical target signal features to generate second target signal features; performing a quality check on the current ECG data based on the first target signal feature and the second target signal feature; The first target signal feature and the second target signal feature corresponding to the current electrocardiogram data whose quality detection result meets the preset conditions are determined as the target signal features.

6. The method according to claim 5, characterized in that The preprocessing of the current ECG signal data to obtain first quasi-ECG data and second quasi-ECG data comprises: performing pacing processing on the current electrocardiogram signal data to generate pacing processing data; Performing baseline shift processing on the pacing processed data to obtain first quasi-ECG data and second pre-processed ECG data; The second preprocessed ECG data is subjected to noise reduction processing and data amplification processing in sequence to obtain second quasi-ECG data.

7. The method according to claim 1, characterized in that Also includes: Collecting the original ECG data of each signal channel within the current preset time; Determining the lead-off state of the signal channel according to the original electrocardiogram data; The original electrocardiogram data in a fully connected state is determined based on the lead-off state of the signal channel.

8. An electrocardiogram lead pattern detection system, characterized in that: include: Feature extraction module, lead detection module and lead output module; The feature extraction module is used to extract ECG features from the current ECG data to obtain target signal features; the current ECG data is used to indicate the original ECG data that meets the full connection state within the current preset time; The lead detection module is configured to determine a number of signal channels corresponding to the original ECG data; based on the target signal characteristics and the corresponding original ECG data, analyze the similarity between each two signal channels to obtain a single-channel similarity result; for any one of the several signal channels: based on the single-channel similarity result, determine the difference of the signal channel relative to all other signal channels to obtain a difference score corresponding to the signal channel; based on the difference score corresponding to each signal channel, determine a lead mode to be output corresponding to the current ECG data; The lead output module is configured to output a first lead pattern when the target signal characteristics and the lead pattern to be output meet a preset output condition.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the electrocardiogram lead pattern detection method according to any one of claims 1 to 7 are implemented.

Images