Electrocardiogram monitoring equipment and electrocardiogram data processing method

Abstract

The invention relates to electrocardiogram monitoring equipment and an electrocardiogram data processing method. The device comprises six patch electrodes which are used for being pasted in a preset range of the chest of a monitored object to obtain six precordial lead signals; the signal collector is used for collecting the six precordial lead signals and sending the six precordial lead signals to the signal processor; the signal processor is used for obtaining five groups of chest lead difference vectors according to the six precordial lead signals and obtaining current standard 12-lead electrocardiosignals of the monitored object according to the five groups of chest lead difference vectors and a trained lead conversion model; wherein the trained lead conversion model is obtained by training a 12-lead electrocardiosignal sample. By adopting the equipment, an accurate electrocardiogram monitoring result can still be obtained under the condition that a pasting position deviates when a user pastes an electrode by himself / herself, and independent use of the user is facilitated.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to an electrocardiogram (ECG) monitoring device and a method for processing ECG data. Background Technology

[0002] Electrocardiography (ECG) has become one of the main methods used in clinical diagnosis of heart diseases. An ECG measuring device acquires cardiac function data by measuring potential differences.

[0003] Currently, the most commonly used clinical method for simultaneous 12-lead electrocardiogram (ECG) acquisition amplifies and synchronously labels the ECG signal, typically using the Mason-Likar lead system electrode placement for ECG recording. The Mason-Likar lead system electrode placement on the human body is as follows... Figure 1 As shown, the system includes six precordial leads (C1-C6) and four limb leads (R, L, F, N). The upper limb leads are placed between the middle of the clavicle and the acromion, while the left lower limb lead is placed in the middle between the costal margin and the ileocecal junction at the left anterior axillary line. The Mason-Likar lead system has also recently been used for ECG monitoring, with the upper limb leads positioned outside the clavicle. Of the six precordial leads, C1 is located at the right sternal border in the 4th intercostal space, C2 at the left sternal border in the 4th intercostal space, C3 at the midpoint of the line connecting C2 and C4, C4 at the intersection of the left midclavicular line and the 5th intercostal space, C5 at the left anterior axillary line at the same level as C4, and C6 at the left midaxillary line at the same level as C4.

[0004] However, using the Mason-Likar lead system electrodes for ECG monitoring requires professional installation guidance, is limited to hospital use, and is not conducive to users' long-term home monitoring to obtain accurate ECG monitoring results, thus failing to meet users' needs for independent use of the product. Summary of the Invention

[0005] Based on this, it is necessary to provide an ECG monitoring device, ECG data processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can meet the needs of independent user use and can provide monitoring results as a standard 12-lead ECG.

[0006] In a first aspect, this application provides an electrocardiogram monitoring device, comprising: a flexible substrate, a signal collector, and a signal processor, wherein the flexible substrate is provided with six patch electrodes;

[0007] The six patch electrodes are used to be attached to a preset area on the chest of the monitored object to obtain six precordial lead signals.

[0008] The signal collector is used to collect the signals from the six precordial leads and send them to the signal processor;

[0009] The signal processor is configured to obtain five sets of precordial lead difference vectors based on the six precordial lead signals, and to obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of precordial lead difference vectors and a trained lead conversion model; wherein the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0010] Secondly, this application also provides a method for processing electrocardiogram (ECG) data, including:

[0011] Acquire the six precordial leads of the monitored object by six patch electrodes;

[0012] Based on the six precordial lead signals, five sets of precordial lead difference vectors are obtained;

[0013] The current standard 12-lead ECG signal of the monitored object is obtained based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0014] Thirdly, this application also provides an apparatus for processing electrocardiogram data, comprising:

[0015] The signal acquisition module is used to acquire the six precordial lead signals of the monitored object collected by the six patch electrodes.

[0016] The difference calculation module is used to obtain five sets of chest lead difference vectors based on the six precordial lead signals.

[0017] The lead conversion module is used to obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0018] Fourthly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0019] Acquire the six precordial leads of the monitored object by six patch electrodes;

[0020] Based on the six precordial lead signals, five sets of precordial lead difference vectors are obtained;

[0021] The current standard 12-lead ECG signal of the monitored object is obtained based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0022] Fifthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0023] Acquire the six precordial leads of the monitored object by six patch electrodes;

[0024] Based on the six precordial lead signals, five sets of precordial lead difference vectors are obtained;

[0025] The current standard 12-lead ECG signal of the monitored object is obtained based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0026] Sixthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0027] Acquire the six precordial leads of the monitored object by six patch electrodes;

[0028] Based on the six precordial lead signals, five sets of precordial lead difference vectors are obtained;

[0029] The current standard 12-lead ECG signal of the monitored object is obtained based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0030] The aforementioned ECG monitoring equipment, ECG data processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product acquire six precordial lead signals through six patch electrodes within their respective preset chest areas. A signal collector collects the six precordial lead signals and sends them to a signal processor. The signal processor obtains five sets of precordial lead difference vectors based on the six precordial lead signals. Based on the five sets of precordial lead difference vectors and a trained lead conversion model, it obtains the current standard 12-lead ECG signal of the monitored subject. This reduces the number of patch electrodes and simplifies usage, while allowing patch electrodes to be placed within the preset chest area. Even if the user places the electrodes themselves, resulting in placement deviations, accurate ECG monitoring results can still be obtained, facilitating independent use by the user. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 A schematic diagram showing the electrode positions of the Mason-Likar lead system;

[0033] Figure 2 This is a schematic diagram of an electrocardiogram monitoring device in one embodiment;

[0034] Figure 3 This is a schematic diagram of the target location where the patch electrode is attached in one embodiment;

[0035] Figure 4 This is a schematic diagram of an alignment aid in one embodiment;

[0036] Figure 5 This is a schematic diagram of an electrocardiogram monitoring device with an alignment aid installed in one embodiment;

[0037] Figure 6 This is a flowchart illustrating a method for processing electrocardiogram (ECG) data in one embodiment.

[0038] Figure 7 This is a flowchart illustrating the training process of a standard 12-lead ECG model in one embodiment.

[0039] Figure 8 This is a schematic diagram illustrating the construction of a lead conversion model in one embodiment;

[0040] Figure 9 This is a schematic diagram illustrating an application scenario of the lead conversion model in one embodiment;

[0041] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0042] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0043] In the description of this application, it should be understood that if terms such as "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" appear, these terms indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0044] Furthermore, where the terms "first" and "second" appear, these terms are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, where the term "multiple" appears, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0045] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "joining," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0046] In this application, unless otherwise expressly specified and limited, the use of descriptions such as "above" or "below" the second feature indicates that the first and second features are in direct contact or indirect contact via an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. Similarly, "below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0047] It should be noted that if an element is referred to as being "fixed to" or "set on" another element, it can be directly on the other element or there may be an intervening element. If an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intervening element. If so, the terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used in this application are for illustrative purposes only and do not represent the only possible implementation.

[0048] refer to Figure 2 In one embodiment, an electrocardiogram (ECG) monitoring device is provided, including: a flexible substrate 102, a signal collector 104, and a signal processor 106. The flexible substrate 102 is provided with six patch electrodes 108. The six patch electrodes 108 are used to be attached to a predetermined area on the chest of the monitored subject to obtain six precordial lead signals. The signal collector 104 is used to collect the six precordial lead signals and send them to the signal processor 106. The signal processor 106 is used to obtain five sets of precordial lead difference vectors based on the six precordial lead signals, and to obtain the current standard 12-lead ECG signal of the monitored subject based on the five sets of precordial lead difference vectors and a trained lead conversion model. The trained lead conversion model is trained based on 12-lead ECG signal samples.

[0049] The flexible substrate can be a base formed of a flexible material, such as polyester film or polyvinyl alcohol film. The signal collector can be, but is not limited to, various devices, equipment, or unit modules with signal receiving and transmitting functions. The signal processor can be, but is not limited to, various devices, equipment, or unit modules with signal processing functions; it can be integrated with the signal processor and carried by the user, or it can be a terminal or server that communicates wirelessly with the signal processor. The patch electrode can be a conductive electrode with adhesive or magnetic adhesion functions.

[0050] The monitored object can be any object subject to electrocardiogram (ECG) monitoring. The preset chest area can be a region determined based on the target location, such as a designated area surrounding the target location, where the target location can be the placement location of the six precordial leads in a standard 12-lead ECG. The six precordial lead signals can be the precordial potential signals acquired by the six patch electrodes. The precordial lead difference vector can be the vector obtained by subtracting any two precordial lead signals from the six precordial lead signals. The trained lead conversion model can be a pre-trained model capable of reconstructing the five sets of precordial lead difference vectors into a standard 12-lead ECG signal, including but not limited to deep learning models, neural network models, etc. The 12-lead ECG signal samples can be the sample data used to train the lead conversion model.

[0051] Specifically, six patch electrodes can be attached to a preset area on the chest of the monitored subject, and each patch electrode can collect its own precordial lead signal to obtain six precordial lead signals. Each patch electrode can send the collected precordial lead signals to a signal collector. The signal collector collects the six precordial lead signals and forwards them to a signal processor. The signal processor can store a lead conversion model that has been trained in advance based on 12-lead ECG signal samples. After receiving the six precordial lead signals, the signal processor obtains five sets of precordial lead difference vectors based on the six precordial lead signals. The five sets of precordial lead difference vectors are input into the trained lead conversion model to obtain the current standard 12-lead ECG signal of the monitored subject.

[0052] Figure 3 A schematic diagram of the target location for patch electrode placement is provided. (Reference) Figure 3 The six patch electrodes can be placed at the target locations: right fourth intercostal space 201, left fourth intercostal space 202, left fifth intercostal space 203, the intersection of the left midclavicular line and the fifth intercostal space 204, left anterior axillary line 205, and left midaxillary line 206. During use, the patch electrodes can be placed within a preset range at each target location. For example, it is permissible to place the corresponding patch electrode at any location within a circular area centered on the target location and with a specified distance as the radius. It is understood that the preset range on the chest is not limited to a circle and can be of any shape. Thus, the six patch electrodes can acquire six precordial lead signals UV1, UV2, UV3, UV4, UV5, and UV6, which are collected by the signal collector and sent to the signal processor. The signal processor determines five sets of precordial lead difference vectors based on the received UV1, UV2, UV3, UV4, UV5, and UV6. Specifically, a precordial lead signal can be selected, and the differences between other precordial lead signals and this precordial lead signal can be calculated. For example, if UV1 is selected, the differences between other precordial lead signals and this precordial lead signal can be calculated to obtain five sets of precordial lead difference vectors: UV2-UV1, UV3-UV1, UV4-UV1, UV5-UV1, and UV6-UV1. In addition, the differences between any two adjacent precordial lead signals can also be calculated. For example, five sets of precordial lead difference vectors can be obtained based on UV1-UV2, UV2-UV3, UV3-UV4, UV4-UV5, and UV5-UV6. This application does not limit the method for determining the five sets of precordial lead difference vectors. The signal processor can also store a pre-trained lead conversion model. The obtained five sets of precordial lead difference vectors are input into the lead conversion model, and the five sets of precordial lead difference vectors are reconstructed into the current standard 12-lead ECG signal through the lead conversion model.

[0053] It should be noted that the six precordial lead signals may not have been acquired at the target location, thus introducing bias, which in turn introduces bias into the five sets of precordial lead difference vectors. In order to accurately reconstruct the current standard 12-lead ECG signal from the five sets of precordial lead difference vectors, 12-lead ECG signal samples with bias can be used during the training of the lead conversion model. For example, the six precordial lead signal samples in the 12-lead ECG signal sample can be acquired within a preset range of the monitored subject's chest.

[0054] The aforementioned ECG monitoring device acquires six precordial lead signals through six patch electrodes placed within their respective preset chest areas. The signal collector collects these six precordial lead signals and sends them to the signal processor. The signal processor obtains five sets of precordial lead difference vectors based on the six precordial lead signals. Based on these five sets of precordial lead difference vectors and a trained lead conversion model, it obtains the current standard 12-lead ECG signal of the monitored subject. This device can reduce the number of patch electrodes and simplify usage, while allowing patch electrodes to be placed within the preset chest area. Even if the user places the electrodes themselves and causes deviations in placement, accurate ECG monitoring results can still be obtained, making it convenient for independent use.

[0055] In one embodiment, the ECG monitoring device further includes an alignment aid for covering a flexible substrate; the alignment aid has holes equal in number to the number of patch electrodes, the holes corresponding to a preset range on the chest, and the holes are used to guide the user to move the patch electrodes to the corresponding preset range on the chest.

[0056] The alignment aid can be a device that assists in attaching the patch electrode, such as a paper or plastic template.

[0057] Specifically, alignment aids can be installed for electrocardiogram (ECG) monitoring equipment. Figure 4 A schematic diagram of an alignment aid is provided for reference. Figure 4 The alignment aid 302 can be provided with the same number of holes 304 as the patch electrodes. Each hole corresponds to a preset area on the chest. For example, the size of the hole can be the same as the preset chest area. The holes can be transparent or hollow, and their shape is not limited, as long as it allows for easy observation of the patch electrodes. Refer to [reference needed] when using. Figure 5 Alignment aids can be placed over the flexible substrate of the ECG monitoring device, and patch electrodes can be attached to each hole to ensure that the placement of the patch electrodes does not exceed the corresponding preset range on the chest. By setting alignment aids, the acquisition error of the six precordial leads can be reduced, improving the accuracy and reliability of ECG monitoring.

[0058] In one embodiment, the flexible substrate has at least one opening, and the patch electrode is slidably mounted in the opening to move to a predetermined range on the chest according to the hole position.

[0059] Specifically, one or more openings can be provided on the flexible substrate, and six patch electrodes can be slidably installed within the openings. In use, the patch electrodes are slid within the openings and moved to a preset range on the chest according to the hole positions. By providing openings on the flexible substrate, it can be ensured that the patch electrodes are placed within the preset range on the chest, improving the accuracy and reliability of ECG monitoring.

[0060] In one embodiment, a conductive layer is also provided on the flexible substrate. The conductive layer is electrically connected to the patch electrode. By designing conductive layers of different lengths, the patch electrode can be attached to a predetermined area on the chest.

[0061] Specifically, a conductive layer can be disposed on a flexible substrate. This conductive layer contains a conductive material and can be electrically connected to the patch electrode. The length of the conductive layer can be set according to the size of a preset chest area. A larger preset chest area allows for a longer conductive layer, while a smaller preset chest area allows for a shorter conductive layer, thus facilitating the movement of the patch electrode. By also providing a conductive layer on the flexible substrate, the precordial lead signals acquired by the patch electrode can be transmitted to the signal collector, ensuring reliable transmission of the precordial lead signals.

[0062] In one embodiment, the flexible substrate is further provided with a connection layer, and the conductive layer is electrically connected to the patch electrode through the connection layer.

[0063] Specifically, a connection layer can be disposed on a flexible substrate. This connection layer contains a conductive material, with one end of the conductive layer electrically connected to six patch electrodes via the connection layer, and the other end electrically connected to a signal collector. In use, each patch electrode can transmit the acquired precordial lead signal through the connection layer to the conductive layer, which then sends it to the signal collector. By using the connection layer, reliable transmission of the precordial lead signal can be ensured, preventing monitoring interruptions.

[0064] In practical applications, a standard 12-lead system has ten different electrode sites, including three limb electrodes: LA (Leftarm), RA (Right Arm), and LL (Left Leg); six precordial electrodes: UV1, UV2, UV3, UV4, UV5, and UV6; and one RLD (Right Leg Drive). The RLD only enhances signal quality and is not involved in lead calculation; therefore, the 12 leads are derived from the remaining nine electrodes. The six limb leads are obtained from the LA, RA, and LL electrodes; the six precordial leads are obtained by subtracting the Wilson's Central Terminal (WCT) from the six precordial electrodes. The WCT is a theoretical reference point and can be represented by the average of the RA, LA, and LL electrode sites. Therefore, existing technologies require first determining the WCT based on the RA, LA, and LL electrode sites, and then determining the six precordial leads based on the WCT and the six precordial electrodes; it is not possible to directly obtain the six precordial leads from the six precordial electrodes. In this application, in order to avoid WCT measurement, five sets of precordial lead difference vectors are calculated based on the six precordial lead signals acquired by the six precordial electrodes, and the standard 12-lead ECG signal is reconstructed based on the five sets of precordial lead difference vectors.

[0065] In one embodiment, the signal processor includes a precordial lead vector calculation module and a lead conversion module; the precordial lead vector calculation module is used to subtract any two precordial lead vectors from the six precordial lead signals to obtain five sets of precordial lead difference vectors; the lead conversion module is used to obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of precordial lead difference vectors and the trained lead conversion model.

[0066] The chest lead vector calculation module and lead conversion module can both be physical modules in a signal processor, including but not limited to general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence processors, etc.

[0067] Among them, the precordial lead vector can be the vector formed by connecting the center point of the heart to the precordial lead signals.

[0068] Specifically, the target positions for attaching the six patch electrodes can be set at the right fourth intercostal space, the left fourth intercostal space, the left fifth intercostal space, the intersection of the left midclavicular line and the fifth intercostal space, the left anterior axillary line, and the left mid-axillary line, respectively. In use, the patch electrodes are attached within a preset range at each target position, and six precordial lead signals UV1, UV2, UV3, UV4, UV5, and UV6 are collected. These signals are collected by the signal collector and sent to the signal processor. The precordial lead vector calculation module of the signal processor can perform vector difference calculation on the precordial lead vectors corresponding to any two of the six received precordial lead signals to obtain five sets of precordial lead difference vectors. For example, a precordial lead signal can be selected, and the vector difference between the other precordial lead signals and the precordial lead signal can be calculated to obtain five sets of precordial lead difference vectors. Alternatively, the vector difference between any two adjacent lead signals can be calculated to obtain five sets of precordial lead difference vectors. This application does not limit this. Five sets of precordial lead difference vectors are transmitted to the lead conversion module, which stores a trained lead conversion model. The five sets of precordial lead difference vectors are input into the trained model to obtain the current standard 12-lead ECG signal of the monitored subject. By calculating the five sets of precordial lead difference vectors, the WCT measurement can be avoided; the standard 12-lead ECG signal can be obtained directly by measuring the signals of six precordial leads. This reduces the complexity of the ECG monitoring equipment, enriches the information in the ECG signal, and increases the portability of the ECG monitoring equipment.

[0069] In one embodiment, such as Figure 6 As shown, a method for processing electrocardiogram (ECG) data is provided, which can be applied to... Figure 2 Taking the signal processor 106 of the central electrical monitoring equipment as an example, the method includes the following steps:

[0070] Step S510: Acquire the six precordial leads of the monitored object collected by the six patch electrodes;

[0071] Step S520: Based on the six precordial lead signals, obtain five sets of precordial lead difference vectors;

[0072] Step S530: Obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on the 12-lead ECG signal samples.

[0073] In practice, six patch electrodes can collect six precordial lead signals from the monitored subject. The signal collector collects these signals and sends them to the signal processor. The signal processor determines five sets of precordial lead difference vectors based on the received signals and inputs these vectors into a trained lead conversion model to reconstruct the subject's current standard 12-lead ECG signal. The signal processor can be pre-trained using 12-lead ECG signal samples to obtain the trained lead conversion model. To allow for accurate reconstruction of the standard 12-lead ECG signal even with deviations in the patch electrode placement, six precordial lead signal samples from the 12-lead ECG signal sample can be collected within a preset area on the chest.

[0074] In this embodiment, six precordial lead signals of the monitored subject are acquired by six patch electrodes. Based on the six precordial lead signals, five sets of precordial lead difference vectors are obtained. The current standard 12-lead ECG signal of the monitored subject is obtained based on the five sets of precordial lead difference vectors and the trained lead conversion model. This avoids the need to calculate WCT, only requires the acquisition of six precordial lead signals, and reconstructs the standard 12-lead ECG signal based on the five sets of precordial lead difference vectors corresponding to the six precordial lead signals. This enriches the information of the ECG signal and enables long-term accurate monitoring of ECG data.

[0075] In one embodiment, such as Figure 7 As shown, the training process of the lead conversion model in the above-mentioned ECG data processing method can specifically include:

[0076] Step S501: Acquire 12-lead ECG signal samples; the 12-lead ECG signal samples include six precordial lead signal samples; the six precordial lead signal samples are collected within a preset range on the chest of the monitored subject;

[0077] Step S502: Obtain the ECG label dataset based on the 12-lead ECG signal samples;

[0078] Step S503: Subtract any two precordial lead signal samples from the six precordial lead signal samples to obtain five sets of precordial lead difference vector samples.

[0079] Step S504: Obtain the ECG sample dataset based on the five sets of chest lead difference vector samples;

[0080] Step S505: Train the lead conversion model to be trained based on the ECG label dataset and the ECG sample dataset to obtain the trained lead conversion model.

[0081] The ECG label dataset can be a set of labels used to train the lead conversion model. The ECG sample dataset can be a set of samples used to train the lead conversion model.

[0082] Specifically, multiple sets of 12-lead ECG signals can be collected from the monitored subject as 12-lead ECG signal samples. Among them, six precordial lead signals from each set of 12-lead ECG signals, i.e., six precordial lead signal samples, can be collected within a preset range on the chest of the monitored subject. The collected 12-lead ECG signal samples are sent to the signal processor through a signal collector. The signal processor combines the 12-lead ECG signal samples into an ECG label dataset and subtracts any two precordial lead signal samples from the six precordial lead signal samples to obtain five sets of precordial lead difference vector samples. These five sets of precordial lead difference vector samples are combined into an ECG sample dataset. During model training, the five sets of precordial lead difference vector samples from the ECG sample dataset can be input into the lead conversion model to be trained to obtain the predicted value of the 12-lead ECG signal. The model parameters are adjusted according to the difference between the predicted value and the 12-lead ECG signal samples in the ECG label dataset until convergence, resulting in the trained lead conversion model.

[0083] In this embodiment, a 12-lead ECG signal sample is acquired, including six precordial lead signal samples. These six precordial lead signal samples are collected within a preset range on the chest of the monitored subject. Based on the 12-lead ECG signal samples, an ECG label dataset is obtained. Any two precordial lead signal samples are subtracted to obtain five sets of precordial lead difference vector samples. Based on these five sets of precordial lead difference vector samples, an ECG sample dataset is obtained. The lead conversion model to be trained is then trained using the ECG label dataset and the ECG sample dataset. This results in a trained lead conversion model. During model training, biases can be introduced into the six precordial lead signal samples, allowing the trained lead conversion model to accurately reconstruct the standard 12-lead ECG data of the monitored subject even when there are deviations in the acquisition positions of the six precordial lead signals, thus improving the accuracy and reliability of ECG monitoring.

[0084] In one embodiment, step S503 may specifically include: determining six precordial lead signal samples acquired at six target locations, and defining the six precordial lead signal samples as six precordial lead signal standard samples; subtracting any two precordial lead signal standard samples from the six precordial lead signal standard samples to obtain a precordial lead difference vector standard sample group; the precordial lead difference vector standard sample group includes five precordial lead difference vector standard samples; acquiring six precordial lead signal deviation samples according to the six target locations; the six precordial lead signal deviation samples include at least one precordial lead signal sample acquired at a deviation location; the deviation location is determined based on the target location and the deviation amount of a preset location; subtracting any two precordial lead signal deviation samples from the six precordial lead signal deviation samples to obtain a precordial lead difference vector deviation sample group; the precordial lead difference vector deviation sample group includes five precordial lead difference vector deviation samples; step S04 may specifically include: obtaining an electrocardiogram sample dataset based on the precordial lead difference vector standard sample group and at least one precordial lead difference vector deviation sample group.

[0085] The six standard samples of precordial lead signals can be any six standard precordial lead signal samples. The standard sample group of precordial lead difference vectors can be the differences between these six standard precordial lead signal samples. The five standard sample groups of precordial lead difference vectors can be any five standard precordial lead difference vector samples. The six precordial lead signal deviation samples can be any six precordial lead signal samples carrying deviations. The precordial lead difference vector deviation sample group can be the differences between these six precordial lead signal samples carrying deviations. The five precordial lead difference vector deviation samples can be any five precordial lead difference vector samples carrying deviations.

[0086] The preset position deviation can be the distance between the electrode patch's placement position and the target position. The six target positions can be located at the right fourth intercostal space, left fourth intercostal space, left fifth intercostal space, the intersection of the left midclavicular line and the fifth intercostal space, the left anterior axillary line, and the left midaxillary line of the monitored object.

[0087] Specifically, the signal processor can identify samples acquired at six target locations from six precordial lead signal samples, designating these samples as six standard precordial lead signal samples. Subtracting any two standard precordial lead signal samples from these six samples yields a set of standard precordial lead difference vector samples, comprising five sets of standard precordial lead difference vector samples. The signal processor can also identify six precordial lead signal samples containing one or more precordial lead signal samples acquired at deviation locations, serving as six precordial lead signal deviation samples. Subtracting any two deviation precordial lead signal samples from these six deviation samples yields a set of precordial lead difference vector deviation samples, comprising five sets of precordial lead difference vector deviation samples. These precordial lead difference vector standard sample sets and one or more precordial lead difference vector deviation sample sets can then be combined to form an ECG sample dataset.

[0088] For example, the signal processor can filter six precordial lead signal samples. If the six precordial lead signal samples were acquired at six target locations, they are identified as six standard precordial lead signal samples. Five sets of standard precordial lead difference vector samples are then obtained based on the difference between any two of these standard precordial lead signal samples, resulting in a standard precordial lead difference vector sample set. Otherwise, if one or more precordial lead signal samples were acquired within a preset range on the chest corresponding to the target location, they are identified as six precordial lead signal deviation samples. Five sets of precordial lead difference vector deviation samples are then obtained based on the difference between any two of these deviation samples, resulting in a precordial lead difference vector deviation sample set. The standard precordial lead difference vector sample set and the precordial lead difference vector deviation sample set can then be combined to form an ECG sample dataset.

[0089] In this embodiment, by determining the standard sample group and the deviation sample group of the chest lead difference vector, the ECG sample dataset is obtained based on the standard sample group and at least one deviation sample group of the chest lead difference vector. Chest lead difference vector deviation can be introduced into the ECG sample dataset to improve the anti-bias capability of the lead conversion model and improve the accuracy of ECG monitoring.

[0090] In one embodiment, step S505 may specifically include: obtaining standard 12-lead ECG signal prediction data based on the ECG sample dataset and the lead conversion model to be trained; determining the maximum preset position deviation for the target location based on the standard 12-lead ECG signal prediction data and the ECG label dataset; and obtaining the preset chest range corresponding to the target location based on the maximum preset position deviation.

[0091] The standard 12-lead ECG signal prediction data can be the standard 12-lead ECG signal predicted by the lead conversion model to be trained. The preset position deviation can be the deviation value of the actual sampling position of the six precordial leads relative to the target position.

[0092] Specifically, the signal processor can input five sets of chest lead difference vector samples from the ECG sample dataset into the lead conversion model to be trained, obtaining multiple sets of standard 12-lead ECG signal prediction data. The multiple sets of standard 12-lead ECG signal prediction data are compared with the 12-lead ECG signal samples in the ECG label dataset to obtain the maximum preset position deviation for each target position. Then, the preset chest range corresponding to each target position can be determined based on the maximum preset position deviation. For example, the preset chest range for the target position can be defined with the target position as the center and the maximum preset position deviation as the radius. Alternatively, the preset chest range can be set to an arbitrary shape region containing the target position and the location of the standard 12-lead ECG signal prediction data with the maximum deviation. This application does not limit the specific method for setting the preset chest range.

[0093] In this embodiment, standard 12-lead ECG signal prediction data is obtained based on the ECG sample dataset and the lead conversion model to be trained. Based on the standard 12-lead ECG signal prediction data and the ECG label dataset, the maximum preset position deviation for the target position is determined. Based on the maximum preset position deviation, the preset chest range corresponding to the target position is obtained. This allows for a preset chest range that meets the needs of actual clinical applications, enabling ECG signal monitoring to meet the requirements of actual clinical applications.

[0094] In one embodiment, step S520 may specifically include: subtracting any two of the six precordial lead signals to obtain five sets of precordial lead difference vectors; the six target locations are respectively located at the right fourth intercostal space, left fourth intercostal space, left fifth intercostal space, the intersection of the left midclavicular line and the fifth intercostal space, the left anterior axillary line, and the left midaxillary line of the monitored object.

[0095] Specifically, the target locations for attaching the six patch electrodes can be the right fourth intercostal space, the left fourth intercostal space, the left fifth intercostal space, the intersection of the left midclavicular line and the fifth intercostal space, the left anterior axillary line, and the left mid-axillary line of the monitored subject. After receiving the signals from the six precordial leads, the signal processor can subtract any two of the precordial lead signals to obtain five sets of precordial lead difference vectors. Then, based on the five sets of precordial lead difference vectors, lead reconstruction is performed to obtain the standard 12-lead ECG signal of the monitored subject.

[0096] In this embodiment, by subtracting any two precordial lead signals from the six precordial lead signals to obtain five sets of precordial lead difference vectors, and setting six target positions, the WCT measurement can be avoided. Standard 12-lead ECG data can be reconstructed using only six precordial leads, enriching the information of the ECG signal and enabling long-term accurate monitoring of ECG data.

[0097] To facilitate a deeper understanding of the embodiments of this application by those skilled in the art, a specific example will be used for illustration below.

[0098] This application uses precordial leads as optimized electrode acquisition sites for ECG signal acquisition, and then designs a lead conversion method to reconstruct a standard 12-lead ECG, thereby significantly reducing the number of electrodes acquired in the standard 12-lead ECG while reconstructing a high-quality 12-lead signal.

[0099] In electrocardiogram (ECG) leads, I, II, III, aVF, aVL, and aVR are obtained using three electrodes placed on RA, LA, and LL, and are called limb leads. I, II, and III are also called bipolar limb leads, reflecting the potential difference between two limbs; aVL, aVR, and aVF are also called pressurized unipolar limb leads, reflecting the potential difference between two points on the body surface. Relative to the heart, these leads primarily detect planar electrical activity in the front of the body. The recording electrode sites (UV1-UV6) of the precordial (precordial) leads (V1-V6) are located anterior to the chest wall, while the reference electrode site WCT is located within the thoracic cavity. Therefore, precordial leads are suitable for detecting electrical activity in the horizontal direction. Precordial leads are unipolar leads; by placing the recording electrodes at specific locations on the chest, ECG signals can be obtained. This electrode placement method, with the electrodes very close to the heart and separated only by a layer of chest wall, results in a larger amplitude waveform in the obtained ECG lead signals.

[0100] To ensure the accuracy of lead conversion and considering the convenience of patch wearing, this application uses the six precordial leads from the standard 12-lead ECG for lead conversion. Existing lead conversion methods often use two limb leads and one precordial lead for 12-lead conversion. Since the two limb leads are known, all other limb leads can be derived from the formula, thus effectively using all limb leads and one precordial lead, for a total of seven leads. However, this contains sufficient information, and any two limb leads require connection to at least three limbs, failing to achieve sufficient simplification in ECG patch design. While using precordial leads alone simplifies ECG patch design, precordial leads require WCT assistance and cannot be used directly, significantly limiting the simplification of standard 12-lead ECG acquisition sites.

[0101] In response, this application proposes using the difference between the six precordial leads to reconstruct the standard 12-lead network, thereby overcoming the limitation of requiring WCT in the precordial leads. The specific process may include:

[0102] Step S601: The user first wears a standard 12-lead device for a period of time to collect 12-lead electrocardiogram signal samples;

[0103] Step S602: Construct the difference between the six precordial lead signal samples in the 12-lead ECG signal sample, and determine the mapping relationship between the difference and the 12-lead ECG signal sample. Specifically, a linear method or a nonlinear method can be used. The linear method can be a linear equation, and the nonlinear method includes support vector machines, artificial neural networks, etc.

[0104] Step S603: The user switches the device to an ECG monitoring device that only collects data from the precordial leads;

[0105] Step S604: Acquire six precordial lead signals using precordial electrodes, calculate the differences between the six precordial lead signals, and obtain five sets of precordial lead difference vectors.

[0106] In step S605, the standard 12-lead ECG signal is calculated by using the five sets of chest lead difference vectors and combining them with the mapping relationship (i.e., lead conversion model) determined in step S602.

[0107] Figure 8 A schematic diagram illustrating the construction of a lead conversion model is provided. (Reference) Figure 8 The process of constructing a lead conversion model may include the following steps:

[0108] Step S710: Collect raw ECG signals. The raw ECG signals are mainly used for training the lead conversion model. The collected raw ECG signals are used as lead data. First, the chest lead data is divided into training set, validation set and test set. The training set is used to determine the model's weights and biases, the validation set is used to determine the model's hyperparameters, and the test set is used to evaluate the final model.

[0109] Step S720, Data Preprocessing. First, the lead data is resampled to ensure that data from different sources have the same frequency. In this embodiment, the resampling frequency is set to 200Hz. Then, each lead data is filtered sequentially to remove noise interference. The filter cutoff frequency can be a low-pass filter or a band-pass filter, and the filter can be a Butterworth filter, a Chebyshev filter, or other filters with equivalent effects. Then, normalization is performed according to the following formula.

[0110]

[0111] in, , These are the maximum and minimum values ​​in the training set, respectively. For the initial lead data, These are the normalized lead data.

[0112] In step S730, the 12-lead data is used as labels. The preprocessed 12-lead data is used as labels for subsequent model training.

[0113] Step S740, Chest Lead Data. Remove the six limb lead data from the 12-lead data, retaining the six chest lead data as the input to the model.

[0114] Step S741: The chest leads are directly used as input data. The chest leads are obtained by subtracting the WCT from the six precordial electrodes. If the... The electrode potentials on the chest are represented as follows: Then the calculation of this precordial lead can be expressed as:

[0115]

[0116] Here, WCT is a theoretical reference point that is close to zero and stable during the cardiac cycle, and is expressed as the average value of the three electrode sites RA, LA, and LL. This method is relatively common, but it requires the introduction of WCT.

[0117] Step S742: The chest lead difference vectors are used as input data. Chest leads cannot be obtained solely from chest electrodes; the Western CT (WCT) is required. However, WCT requires calculation using three limb leads. By subtracting the chest leads, WCT can be avoided. In this application, V1 is used as the minuend, and the other five chest leads (V2, V3, V4, V5, V6) are used as the subtrahends. Thus, it is equivalent to using the five chest electrode sites (UV2, UV3, UV4, UV5, UV6) as the minuend and UV1 as the subtrahend, resulting in five sets of chest lead difference vectors. The 12 leads can then be calculated from these chest lead difference vectors.

[0118] Step S750: Determine the structure of the neural network. Lead conversion model prediction is a type of regression prediction, including both linear and nonlinear implementation methods. Since nonlinear methods are not limited by the XOR problem, they are generally superior to linear methods. This application uses the ANN (Artificial Neural Networks) method to construct the lead conversion model. The five sets of chest lead difference vectors mentioned above are used as the model input, and the standard 12 leads are used as the model output. The ANN has a three-layer structure: the first layer is the input layer with 5 neurons, the second layer is the hidden layer with 10 neurons, and the third layer is the output layer with 12 neurons. The sigmoid function is used as the activation function. Before ANN training, the weights and biases of the hidden layer neurons are set using random initialization. During backpropagation, the network parameters are updated using batch gradient descent.

[0119] ANN models typically perform well in lead conversion, but they have limitations in certain situations, such as when the user has conditions like ventricular tachycardia, congestive heart failure, or ST-segment elevation myocardial infarction. In these cases, constructing an LSTM (Long Short-Term Memory Network) model can improve lead conversion accuracy. The LSTM model structure is similar to ANN, except that the second hidden layer is set to an LSTM structure with 32 neurons.

[0120] Step S760, Neural Network Parameter Initialization. Network initialization mainly involves initializing the weights and biases of each neuron. There are three common initialization methods: random initialization, fixed-value initialization, and pre-training initialization. For models that have not yet been trained, random initialization or fixed-value initialization is usually chosen; while for models that have already been trained, pre-training initialization can be chosen if further training is required.

[0121] Step S770, Forward Propagation. Forward propagation refers to the process by which data enters from the input layer, passes through the hidden layers, and finally reaches the output layer. The result of forward propagation is determined by the parameters of the neural network and the input data. One forward propagation is essentially one fitting result of the neural network on the data.

[0122] Step S780, Model Training. Model training mainly includes constructing the loss function, minimizing the loss value, and backpropagation. The main process includes: (1) calculating the state and activation value of each layer of the network until the last layer; (2) calculating the error of each layer, and the error calculation process proceeds from the last layer forward; (3) updating the parameters.

[0123] Step S781: Construct the loss function. In the lead conversion model, the predicted result is the value of 12 leads, which is a regression task. In regression tasks, the calculation of the loss function mainly involves the calculation of errors, including the mean squared error loss function, the absolute value loss function, and the mean absolute value error loss function.

[0124] Step S782: Minimize the loss value. The error obtained from batch calculation is called the loss value. A common method to reduce the loss value is the gradient descent algorithm. Gradient descent algorithms can include stochastic gradient descent, mini-batch gradient descent, RMSprop (Root Mean Square propagation) optimization algorithm, adaptive moment estimation algorithm, etc.

[0125] Step S783, backpropagation. The error backpropagation algorithm is also known as the backpropagation algorithm. Backpropagation is mainly used to update the parameters of a neural network, updating the parameters layer by layer using the chain rule.

[0126] Step S784: Update parameters. The parameter update mainly involves the weights and biases. During the process of updating the weights and biases, the parameters of the gradient descent algorithm are also updated.

[0127] Step S790: Model construction complete. Once the model meets the set number of iterations or the calculated loss value is minimized, the model construction is complete, resulting in a neural network-based lead conversion model.

[0128] The lead conversion model proposed in this application can be used in the following two scenarios: (1) Constructing a general model that can be used to convert standard 12-lead ECG signals. The model constructed in this way is an offline model, and the model parameters are not updated; (2) Constructing a personalized model based on user data to convert standard 12-lead ECG signals. The model constructed in this way can accept new data collected by the ECG patch worn by the user, and the model parameters can be updated according to the data, thereby accurately converting the user's 12-lead ECG.

[0129] The lead conversion model proposed in this application can solve the following problems:

[0130] (1) Lead detachment: In the standard 12-lead system, RA and LA are highly important. If lead RA is detached, leads I and II cannot be calculated normally, nor can leads aVR, aVL, and aVF be calculated, and WCT cannot be calculated normally either, meaning that the six precordial leads cannot be calculated normally. In other words, even if only lead RA is detached, 11 out of the 12 leads cannot be displayed normally. Since this application only includes precordial electrodes, only the six precordial leads need to be collected, thus avoiding the above phenomenon.

[0131] (2) Poor quality of individual leads: In the standard 12 leads, when there is poor contact of individual leads or the lead wire is broken, the same situation as lead detachment will occur. Using the method in this application can also avoid this phenomenon.

[0132] Figure 9 A schematic diagram illustrating an application scenario of a lead conversion model is provided. (Reference) Figure 9 A general offline model can be built based on Path One. Public databases of 12-lead ECG (such as the PTB database) contain a large amount of data with good signal quality and high acceptance in the field. Using public datasets as input data helps to build models with good generalization performance and strong robustness. Model evaluation uses CC (Correlation Coefficient) and RMSE (Root Mean Square Error).

[0133] A personalized online model can also be constructed based on path two. For personalized models, the user's individual circumstances need to be considered. Users with different physical conditions exhibit significant differences in their electrocardiogram (ECG) signals. Therefore, it is necessary to collect 12-lead ECG data by having the user wear a 12-lead ECG acquisition device to initially construct the model. Subsequently, by having the user wear the ECG monitoring device in this embodiment of the application, six precordial lead signals are collected, and the lead conversion model is updated with parameters based on the six precordial lead signals. This allows the model to better learn the characteristics of the user's ECG data, thereby accurately converting the user's 12-lead ECG data based on the user's six precordial lead signals.

[0134] The aforementioned personalized online model can achieve differentiated detection and recognition of electrocardiogram signals from different patients, with the following specific features:

[0135] (1) Nonlinear methods based on neural networks. ANN and LSTM models based on neural networks are proposed. These models can effectively reconstruct a standard 12-lead ECG using precordial leads. Compared to traditional linear or machine learning methods, neural network-based methods eliminate the need for manual ECG feature extraction. Furthermore, by modifying parameters during model training, the model avoids getting trapped in local optima, improving its ability to fit ECG reconstructions. For general models, multiple public databases such as the PTB and WCT-ECG databases are used as input, and the reconstruction results are better than those obtained using other methods.

[0136] (2) Building a model based on the user's own data. Currently, most ECG lead reconstruction models do not consider individual differences between different users. In fact, this difference is quite significant and cannot be generalized. Using a general model to reconstruct ECG data from different users has obvious limitations. To address this issue, this application proposes a method for building personalized models based on the user's own data. For users with different heart diseases, the personalized model shows better lead reconstruction results than the general model.

[0137] (3) Effectively applicable to users with different heart diseases. There are huge differences in ECG signals among users with different heart diseases. By collecting data from the user, the personalized model can learn the characteristics of the user's ECG signal, thereby effectively realizing the reconstruction of ECG leads.

[0138] (4) It can be used to monitor the recovery of the condition. For users with heart disease, their electrocardiogram can reflect the severity of the condition to a certain extent. Through continuous monitoring of the user, the lead signals reconstructed by the personalized model can show the user's recovery status.

[0139] (5) Protect user privacy and avoid data leakage. The training data for personalized models comes from the users themselves, and the collected data can be stored locally, thereby avoiding the transfer and flow of data between different entities and effectively preventing user data leakage.

[0140] (6) Personalized placement can be customized for each user. Different users have different wearing habits and placement when wearing ECG patches, which can lead to signal differences. By learning the user's own ECG characteristics through a personalized model, such problems can be effectively avoided. In addition, for users whose own reasons make it inconvenient to place the electrodes in the conventional wearing position, personalized placement can be customized for them.

[0141] The aforementioned general offline model uses two public databases (PTB database and WCT-ECG database). The five precordial leads (excluding V1) are subtracted from lead V1: V2 minus V1, V3 minus V1, V4 minus V1, and V5 minus V1. The resulting ECG signals representing the differences in these five precordial leads are then used as inputs to construct the ANN model. CC and RMSE are used as evaluation metrics.

[0142] Table 1 shows the ANN model constructed using the PTB database. Table 2 shows the ANN model constructed using the WCT-ECG database. The results show the maximum, minimum, average, and median values ​​of the model evaluation metrics. The average and median values ​​indicate that the reconstructed model exhibits good lead reconstruction results. However, since both datasets contain subjects with various cardiac diseases and significant signal fluctuations among different subjects, the minimum value shows some limitation. This limitation can be addressed by constructing a personalized model.

[0143]

[0144]

[0145] Furthermore, precordial leads are primarily used to represent information about the heart's activity in the horizontal plane; therefore, subtracting signals from different precordial leads essentially results in a similar amount of information being conveyed. Referring to Table 3, UV1 can be used as the minuend, and the other five precordial electrode sites (UV2, UV3, UV4, UV5, UV6) as the subtrahends. It is understandable that the minuend can also be other leads; the calculation method for lead differences is shown in the table below. Moreover, subtracting adjacent precordial leads or combining precordial lead differences in other ways is essentially the same.

[0146]

[0147] Furthermore, this application uses both ANN and LSTM methods for lead conversion construction, both of which are nonlinear methods. Other nonlinear lead conversion methods, such as CNN (Convolutional Neural Networks) and DNN (Deep Neural Networks), are also applicable.

[0148] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0149] Based on the same inventive concept, this application also provides an apparatus for implementing the electrocardiogram (ECG) data processing method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more ECG data processing apparatus embodiments provided below can be found in the limitations of the ECG data processing method described above, and will not be repeated here.

[0150] In one embodiment, an apparatus for processing electrocardiogram (ECG) data is provided, comprising: a signal acquisition module, a difference calculation module, and a model conversion module, wherein:

[0151] The signal acquisition module is used to acquire the six precordial lead signals of the monitored object collected by the six patch electrodes.

[0152] The difference calculation module is used to obtain five sets of chest lead difference vectors based on the six precordial lead signals.

[0153] The lead conversion module is used to obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

[0154] In one embodiment, the above-mentioned ECG data processing apparatus further includes a model training module for acquiring the 12-lead ECG signal samples; the 12-lead ECG signal samples include six precordial lead signal samples; the six precordial lead signal samples are collected within a preset range on the chest of the monitored subject; an ECG label dataset is obtained based on the 12-lead ECG signal samples; any two precordial lead signal samples from the six precordial lead signal samples are subtracted to obtain five sets of precordial lead difference vector samples; an ECG sample dataset is obtained based on the five sets of precordial lead difference vector samples; and the lead conversion model to be trained is trained based on the ECG label dataset and the ECG sample dataset to obtain the trained lead conversion model.

[0155] In one embodiment, the model training module is further configured to: determine the six precordial lead signal samples acquired at six target locations, and define the six precordial lead signal samples as six precordial lead signal standard samples; subtract any two precordial lead signal standard samples from the six precordial lead signal standard samples to obtain a precordial lead difference vector standard sample group; the precordial lead difference vector standard sample group includes five precordial lead difference vector standard samples; acquire six precordial lead signal deviation samples according to the six target locations; the six precordial lead signal deviation samples include at least one precordial lead signal sample acquired at a deviation location; the deviation location is determined according to the target location and a preset location deviation amount; subtract any two precordial lead signal deviation samples from the six precordial lead signal deviation samples to obtain a precordial lead difference vector deviation sample group; the precordial lead difference vector deviation sample group includes five precordial lead difference vector deviation samples; and obtain the ECG sample dataset based on the precordial lead difference vector standard sample group and at least one precordial lead difference vector deviation sample group.

[0156] In one embodiment, the model training module is further configured to obtain standard 12-lead ECG signal prediction data based on the ECG sample dataset and the lead conversion model to be trained; determine the maximum preset position deviation for the target position based on the standard 12-lead ECG signal prediction data and the ECG label dataset; and obtain the preset chest range corresponding to the target position based on the maximum preset position deviation.

[0157] In one embodiment, the aforementioned difference calculation module is further used to subtract any two of the six precordial lead signals to obtain five precordial lead difference vectors; the six target locations are respectively located at the right fourth intercostal space, left fourth intercostal space, left fifth intercostal space, the intersection of the left midclavicular line and the fifth intercostal space, the left anterior axillary line, and the left midaxillary line of the monitored object.

[0158] Each module in the aforementioned ECG data processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0159] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for processing electrocardiogram (ECG) data. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0160] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0161] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0162] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0163] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0164] 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 used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0165] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0166] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0167] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An electrocardiogram (ECG) monitoring device, characterized in that, The device includes: a flexible substrate, a signal collector, and a signal processor, wherein six patch electrodes are disposed on the flexible substrate; The six patch electrodes are used to be attached to a preset area on the chest of the monitored object to obtain six precordial lead signals. The signal collector is used to collect the signals from the six precordial leads and send them to the signal processor; The signal processor is configured to obtain five sets of precordial lead difference vectors based on the six precordial lead signals, and to obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of precordial lead difference vectors and a trained lead conversion model; wherein the trained lead conversion model is trained based on 12-lead ECG signal samples.

2. The electrocardiogram monitoring device according to claim 1, characterized in that, The device also includes an alignment aid for covering the flexible substrate; the alignment aid has the same number of holes as the patch electrodes, the holes corresponding to the preset range of the chest, and the holes are used to guide the user to move the patch electrodes to the corresponding preset range of the chest.

3. The electrocardiogram monitoring device according to claim 2, characterized in that, The flexible substrate has at least one opening, and the patch electrode is slidably mounted in the opening to move to a preset range on the chest according to the hole position.

4. The electrocardiogram monitoring device according to claim 1, characterized in that, The flexible substrate is also provided with a conductive layer, which is electrically connected to the patch electrode. By designing conductive layers of different lengths, the patch electrode can be attached to a preset area on the chest.

5. The electrocardiogram monitoring device according to claim 4, characterized in that, The flexible substrate is further provided with a connection layer, and the conductive layer is electrically connected to the patch electrode through the connection layer.

6. The electrocardiogram monitoring device according to any one of claims 1 to 5, characterized in that, The signal processor includes a chest lead vector calculation module and a lead conversion module; the chest lead vector calculation module is used to subtract any two chest lead vectors from the six precordial lead signals to obtain the five sets of chest lead difference vectors; the lead conversion module is used to obtain the current standard 12-lead ECG signal of the monitored object based on the five sets of chest lead difference vectors and the trained lead conversion model.

7. A method for processing electrocardiogram (ECG) data, characterized in that, The method, applied to the electrocardiogram monitoring device as described in any one of claims 1 to 6, comprises: Acquire the six precordial leads of the monitored object by six patch electrodes; Based on the six precordial lead signals, five sets of precordial lead difference vectors are obtained; The current standard 12-lead ECG signal of the monitored object is obtained based on the five sets of chest lead difference vectors and the trained lead conversion model; wherein, the trained lead conversion model is trained based on 12-lead ECG signal samples.

8. The method according to claim 7, characterized in that, The method further includes: The 12-lead ECG signal samples are acquired; the 12-lead ECG signal samples include six precordial lead signal samples; the six precordial lead signal samples are collected within a preset area of ​​the chest of the monitored subject; Based on the 12-lead ECG signal samples, an ECG tag dataset was obtained; Subtract any two precordial lead signal samples from the six precordial lead signal samples to obtain five sets of precordial lead difference vector samples. Based on the five sets of chest lead difference vector samples, an electrocardiogram sample dataset is obtained; Based on the ECG label dataset and the ECG sample dataset, the lead conversion model to be trained is trained to obtain the trained lead conversion model.

9. The method according to claim 8, characterized in that, The subtraction of any two precordial lead signal samples from the six precordial lead signal samples yields five sets of precordial lead difference vector samples, including: The six precordial lead signal samples collected at the six target locations are identified and designated as the six precordial lead signal standard samples. Subtract any two precordial lead signal standard samples from the six precordial lead signal standard samples to obtain a precordial lead difference vector standard sample group; the precordial lead difference vector standard sample group includes five precordial lead difference vector standard samples. Based on the six target locations, six precordial lead signal deviation samples are obtained; the six precordial lead signal deviation samples include at least one precordial lead signal sample acquired at the deviation location; the deviation location is determined based on the target location and the deviation amount of the preset location. Subtract any two precordial lead signal deviation samples from the six precordial lead signal deviation samples to obtain a precordial lead difference vector deviation sample group; the precordial lead difference vector deviation sample group includes five precordial lead difference vector deviation samples. The ECG sample dataset is obtained based on the five sets of chest lead difference vector samples, including: The ECG sample dataset is obtained based on the standard sample group of the precordial lead difference vectors and at least one sample group of the precordial lead difference vectors deviation.

10. The method according to claim 9, characterized in that, The step of training the lead conversion model to be trained based on the ECG label dataset and the ECG sample dataset includes: Based on the ECG sample dataset and the lead conversion model to be trained, standard 12-lead ECG signal prediction data are obtained. Based on the standard 12-lead ECG signal prediction data and the ECG tag dataset, determine the maximum preset position deviation for the target position; Based on the maximum preset position deviation, the preset range of the chest corresponding to the target position is obtained.

Images