Electrocardiogram detection method and device, electronic equipment and storage medium

Abstract

The invention provides an electrocardio detection method and device, electronic equipment and a storage medium. The method comprises the steps that an electrocardio signal of a target user and case text information of the target user are obtained; determining an electrocardio feature vector of the electrocardio signal; performing coding processing on the case text information to obtain a semantic vector; and according to the semantic vector and the electrocardio feature vector, obtaining a detection result of the electrocardio signal.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of electrocardiogram (ECG) detection technology, and more specifically, to an ECG detection method, device, electronic device, and storage medium. Background Art

[0002] A variety of diseases can cause electrocardiogram (ECG) abnormalities. The detection of ECG abnormalities is particularly important for the diagnosis of heart disease. ECG diagnosis has become an indispensable part of clinical diagnosis.

[0003] In recent years, artificial intelligence technology has been used more and more widely in medical imaging, especially with the development of deep learning. Deep learning models, as a high-level representation of objects, have achieved certain achievements in the field of medical image recognition. Existing technologies have also used artificial intelligence technology to perform abnormal analysis on electrocardiogram data.

[0004] However, existing automatic analysis methods for electrocardiogram abnormalities are difficult and have low accuracy due to the huge differences in height, weight, medical history, etc. between individuals, and their electrocardiogram waveforms also vary. Summary of the Invention

[0005] One purpose of the embodiments of the present disclosure is to provide a new technical solution for detecting electrocardiogram signals.

[0006] According to a first aspect of an embodiment of the present disclosure, there is provided an electrocardiogram detection method, comprising:

[0007] Obtaining an electrocardiogram (ECG) signal and medical record information of a target user;

[0008] Determining an electrocardiographic characteristic vector of the electrocardiographic signal;

[0009] Encoding the case text information to obtain a semantic vector;

[0010] A detection result of the electrocardiogram signal is obtained according to the semantic vector and the electrocardiogram feature vector.

[0011] Optionally, determining the electrocardiogram (ECG) feature vector of the ECG signal includes:

[0012] Acquiring global waveform features of the electrocardiogram signal;

[0013] The electrocardiogram feature vector is obtained according to the global waveform feature.

[0014] Optionally, obtaining the global waveform features of the electrocardiogram signal includes:

[0015] determining a plurality of heart beats of the electrocardiogram signal;

[0016] Extract local waveform features of each heartbeat;

[0017] The local waveform features of multiple heartbeats are processed to obtain the global waveform features of the electrocardiogram signal.

[0018] Optionally, the method further includes:

[0019] For each heartbeat, when the duration of the to-be-detected band is greater than or equal to a set time threshold, the local waveform feature corresponding to the to-be-detected band is eliminated.

[0020] Optionally, determining the ECG feature vector of the ECG signal further includes:

[0021] Encoding the electrocardiogram signal through a deep learning model based on an attention mechanism to obtain an encoding vector of the electrocardiogram signal;

[0022] The electrocardiogram feature vector is also obtained according to the encoding vector.

[0023] Optionally, obtaining a detection result of the electrocardiogram signal according to the semantic vector and the electrocardiogram feature vector includes:

[0024] Processing the ECG feature vector and the semantic vector through a classification model based on an attention mechanism to obtain a classification result of the ECG signal;

[0025] A detection result of the electrocardiogram signal is obtained according to the classification result.

[0026] Optionally, encoding the case text information to obtain a semantic vector includes:

[0027] The case text information is encoded based on a large language model to obtain the semantic vector.

[0028] According to a second aspect of the present disclosure, there is provided an electrocardiogram detection device, comprising:

[0029] A signal text acquisition module is used to acquire the target user's ECG signal and the target user's case text information;

[0030] a feature vector determination module, configured to determine an electrocardiogram feature vector of the electrocardiogram signal;

[0031] A text information encoding module, used to encode the case text information to obtain a semantic vector;

[0032] The ECG signal detection module is used to obtain a detection result of the ECG signal according to the semantic vector and the ECG feature vector.

[0033] According to a third aspect of the present disclosure, an electronic device is provided, comprising a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the method described in the first aspect of the present disclosure under the control of the computer program.

[0034] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0035] Through the embodiments of the present disclosure, by integrating the ECG feature vector of the target user's ECG signal and the semantic vector of the target user's case text information through a cross-modal attention mechanism, individual differences can be fully analyzed to obtain the detection results of the ECG signal, so that the ECG detection results can more accurately and comprehensively reflect the heart status of the target user.

[0036] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0038] Figure 1 is a block diagram illustrating a hardware configuration of an electronic device that can implement an embodiment of the present disclosure;

[0039] Figure 2 is a flow chart of an electrocardiogram detection method according to an embodiment of the present disclosure;

[0040] Figure 3 is a structural diagram of a heartbeat according to one embodiment of the present disclosure;

[0041] Figure 4 is a block diagram of an electrocardiogram detection device according to an embodiment of the present disclosure;

[0042] Figure 5 is a block diagram of an electronic device according to one embodiment of the present disclosure. DETAILED DESCRIPTION

[0043] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention.

[0044] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0045] Technologies, methods and equipment known to persons of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, such technologies, methods and equipment should be considered part of the specification.

[0046] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0047] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0048] <Hardware Configuration>

[0049] Figure 1 is a block diagram illustrating a hardware configuration of an electronic device 1000 that can implement an embodiment of the present disclosure.

[0050] The electronic device 1000 may be a wristband, a watch, a computer, an electrocardiogram detection device or other electronic products. Figure 1 As shown, electronic device 1000 may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, and the like. Processor 1100 may be a CPU, a microprocessor MCU, or the like. Memory 1200 may include, for example, ROM (read-only memory), RAM (random access memory), or a non-volatile memory such as a hard disk. Interface device 1300 may include, for example, a USB interface or a headphone jack. Communication device 1400 may be capable of wired or wireless communication, specifically, Wi-Fi, Bluetooth, 2G / 3G / 4G / 5G communication, or the like. Display device 1500 may be, for example, an LCD display or a touchscreen display. Input device 1600 may include, for example, a touchscreen, a keyboard, or somatosensory input. Users may input and output voice information through speaker 1700 and microphone 1800.

[0051] Figure 1 The electronic device shown is merely illustrative and does not in any way limit the present disclosure, its application or use. In the embodiments of the present disclosure, the memory 1200 of the electronic device 1000 is used to store instructions, which are used to control the processor 1100 to operate to perform any of the methods provided in the embodiments of the present disclosure. It should be understood by those skilled in the art that although Figure 1While multiple devices are shown for electronic device 1000, the present disclosure may only relate to some of these devices. For example, electronic device 1000 may only relate to processor 1100 and memory 1200. A skilled person can design instructions based on the solutions disclosed in this disclosure. How instructions control processor operations is well known in the art and will not be described in detail here.

[0052] <Method Example>

[0053] The present disclosure provides an electrocardiogram detection method, which can be implemented by an electronic device, specifically, by Figure 1 The electronic device 1000 is shown as an implementation.

[0054] Figure 2 Flowchart of an electrocardiogram detection method according to an embodiment of the present disclosure.

[0055] like Figure 2 As shown, the method includes steps S2100 to S2400 as shown below:

[0056] Step S2100: Acquire the target user's electrocardiogram signal and case text information.

[0057] The ECG signal in this embodiment may be a single-lead ECG signal or a multi-lead ECG signal, which is not limited here.

[0058] In this embodiment, the target user's electrocardiogram signal may be collected by an electrocardiogram detection device.

[0059] The electrocardiogram signal in this embodiment may be time series data obtained by digitizing an electrocardiogram.

[0060] An electrocardiogram is a graph of various potential changes that are drawn from the body surface through an electrocardiograph during each cardiac cycle, as the pacemaker, atria, and ventricles are excited successively, accompanied by changes in bioelectricity.

[0061] In some embodiments, when an electrocardiogram signal is obtained, the electrocardiogram signal may be further standardized.

[0062] In this embodiment, the normalization process may include at least one of a high-frequency noise removal process and a baseline correction process. By performing the baseline correction process on the ECG signal, the baseline drift of the ECG signal can be eliminated.

[0063] In this embodiment, the target user's case text information may include case information recorded in text form, and may also include the target user's medication history information. The case information may include at least one of the target user's medical history information, physical examination information, laboratory test information, imaging examination information, diagnosis information, and treatment information.

[0064] Medical history information may include previous medical history, the onset time of the current illness, main symptoms and their characteristics, the development and changes of the disease, accompanying symptoms, diagnosis and treatment process, and other conditions related to the current illness, such as diet, sleep, urination, defecation, weight changes, etc.

[0065] Physical examination information includes the measured values of basic vital signs such as body temperature, pulse, respiration, and blood pressure. It may also include descriptive information on general conditions such as consciousness, mental state, nutritional status, body shape, facial expression, etc. It may also include system examination information such as the head and neck, chest, abdomen, spine, limbs, and nervous system.

[0066] Laboratory test information may include the results of routine tests such as blood, urine, and stool, as well as the results of special tests such as biochemical indicators, coagulation function, immune indicators, and microbiological tests.

[0067] Imaging examination information may include the results of X-ray, CT, MRI, ultrasound, radionuclide imaging, and other examinations.

[0068] Diagnostic information may include the doctor's judgment on the target user's disease based on comprehensive information such as the target user's medical history, physical examination, auxiliary examinations, etc., including the name, type, and stage of the disease.

[0069] Treatment information may include records of treatment measures received by the target user in the hospital, including drug treatment (drug name, dosage, usage, medication time, etc.), surgical treatment (surgery name, surgery time, surgery effect, etc.), physical therapy, rehabilitation treatment, etc.

[0070] Step S2200: determining an ECG feature vector of an ECG signal.

[0071] In some embodiments, determining an ECG feature vector of an ECG signal includes: acquiring a global waveform feature of the ECG signal; and obtaining an ECG feature vector based on the global waveform feature.

[0072] In this embodiment, global waveform features may include features reflecting the waveform morphology of the ECG signal. Specifically, global features may include heart rate, heart rhythm, and the overall morphology of the ECG signal. The overall morphology of the ECG signal includes the morphology, amplitude, and time interval of each band. For example, global waveform features may include one or more of the following: P wave morphology, QRS wave morphology, T wave morphology, RR interval, ST segment deviation, and the like.

[0073] In some embodiments, obtaining the global waveform features of the ECG signal includes: determining multiple heartbeats of the ECG signal; extracting the local waveform features of each heartbeat; and processing the local waveform features of the multiple heartbeats to obtain the global waveform features of the ECG signal.

[0074] The local waveform features of this embodiment may include features reflecting the waveform shape of a heartbeat. The local waveform features may include the amplitude, width, shape, direction, depth, slope, etc. of the P wave, QRS wave, T wave, ST wave, etc.

[0075] The local waveform features selected in this embodiment have clear physiological meanings, are easy to detect and identify abnormalities, and help improve the accuracy and stability of abnormal ECG identification.

[0076] Since each heartbeat is regular, the waveform in the electrocardiogram is also regular. The waveform in the electrocardiogram that can fully represent one cardiac cycle is called a heartbeat.

[0077] In this embodiment, different intervals and key points within a heartbeat can be defined for the ECG signal according to the order of myocardial depolarization and repolarization: PR interval, PR segment, QRS complex interval, ST segment, JT interval, QT interval, P wave peak, R wave peak, T wave peak, U wave peak, Q wave trough, S wave trough, and J point. The structural diagram of a heartbeat can be as follows: Figure 3 shown.

[0078] In this embodiment, the heartbeat may be determined in the electrocardiogram signal by one or more of a plurality of intervals and key points.

[0079] In some embodiments, invalid heartbeats in the ECG signal may be removed. Invalid heartbeats may include heartbeats with a period that is too long or too short, or heartbeats caused by a peak of at least one band that is too large or too small.

[0080] In this embodiment, the statistical values of the local waveform features of multiple heartbeats may be determined as the global waveform features of the ECG signal, wherein the statistical values may include any one or more of the average value, maximum value, minimum value, and median value.

[0081] In some embodiments, the ECG feature vector is obtained based on the global waveform features by arranging the global waveform features in a certain order to form a vector, namely the ECG feature vector.

[0082] In some embodiments, the method may further include: for each heartbeat, when the duration of the to-be-detected band is greater than or equal to a set time threshold, eliminating the local waveform feature corresponding to the to-be-detected band.

[0083] In this embodiment, the time threshold may be pre-set based on the cardiac cycle, and the time threshold may be less than the duration of two cardiac cycles of the target user. For example, the time threshold may be the duration of any one cardiac cycle of the target user, or the average duration of multiple cardiac cycles of the target user.

[0084] In this embodiment, a heartbeat includes the following wavebands: P wave, Q wave, R wave, S wave, T wave, and U wave. The waveband to be detected can be any one of the P wave, QRS wave, T wave, and U wave.

[0085] The local waveform feature corresponding to the band to be detected may be the local waveform feature obtained based on the band to be detected.

[0086] Through this embodiment, by eliminating local waveform features corresponding to bands with durations greater than or equal to a set time threshold, the obtained global waveform features can be made more accurate, thereby improving the accuracy of the detection results of the electrocardiogram signal.

[0087] In some embodiments, determining the ECG feature vector of the ECG signal further includes: encoding the ECG signal through a deep learning model based on an attention mechanism to obtain the encoding vector of the ECG signal; and obtaining the ECG feature vector based on the encoding vector.

[0088] In this embodiment, the global features and / or local features of the electrocardiogram signal may be coded to obtain a coding vector.

[0089] Step S2300: Encode the case text information to obtain a semantic vector.

[0090] In this embodiment, the case text information can be encoded based on any one of a bag-of-words model, a word embedding model, a deep learning-based model, and a pre-trained language model to obtain a semantic vector.

[0091] In some embodiments, encoding the case text information to obtain a semantic vector may include: encoding the case text information based on a large language model to obtain a semantic vector.

[0092] The large language model has a more accurate understanding of the semantics of case text information. By encoding the case text information with the large language model, the semantic vector can more accurately represent the semantic dependencies in the case text information.

[0093] Step S2400 , obtaining the detection result of the ECG signal according to the semantic vector and the ECG feature vector.

[0094] The ECG feature vector of this embodiment may include at least one of obtaining the ECG feature vector according to the coding vector and obtaining the ECG feature vector according to the global waveform feature.

[0095] In some embodiments, the detection result of the ECG signal is obtained based on the semantic vector and the ECG feature vector, including: processing the ECG feature vector and the semantic vector through a classification model based on the attention mechanism to obtain the classification result of the ECG signal; and obtaining the detection result of the ECG signal based on the classification result.

[0096] The classification model of this embodiment can be trained based on a set of common abnormal waveforms in different intervals and labels for the abnormal categories corresponding to each waveform in the abnormal waveform set. The abnormal waveform set can include waveforms with absent P waves, elevated P waves, inverted P waves, absent QRS complexes, chaotic QRS complexes, absent T waves, elevated T waves, inverted T waves, prolonged PR intervals, shortened PR intervals, delayed ST segments, and elevated ST segments.

[0097] Furthermore, medical terminology can be used to enable the classification model to fully analyze individual differences, and the classification model can also learn medical prior knowledge to improve the model's understanding of the pathological causes behind abnormal signals, thereby improving the medical interpretability of abnormality detection.

[0098] In one embodiment, the classification result may be the probability that the ECG signal belongs to each category (including a normal category and multiple abnormal categories).

[0099] Furthermore, the category with the highest probability can be used as the category to which the ECG signal belongs. Alternatively, the category with a probability greater than or equal to a corresponding threshold can be used as the category to which the ECG signal belongs. The thresholds corresponding to each category can be the same or different, and are not limited here.

[0100] In another embodiment, the classification result may be a category label of the category to which the ECG signal belongs.

[0101] In this embodiment, the detection result of the ECG signal may include whether the ECG signal is abnormal and the reason why the ECG signal is abnormal.

[0102] Then, the category to which the ECG signal belongs can be determined according to the classification result, and the detection result of the ECG signal can be determined according to the category to which the ECG signal belongs.

[0103] In this embodiment, a mapping table of waveform categories and detection results may be provided, and each mapping relationship in the mapping table may correspond to a waveform category combination and a detection result. A waveform category combination may include at least one waveform category.

[0104] Furthermore, the detection result of the ECG signal can be obtained according to the category to which the ECG signal belongs and the mapping table.

[0105] Based on the test results obtained, the test results can be displayed, uploaded to a server for storage, or sent to other electronic devices for display.

[0106] Through the embodiments of the present disclosure, by integrating the ECG feature vector of the target user's ECG signal and the semantic vector of the target user's case text information through a cross-modal attention mechanism, individual differences can be fully analyzed to obtain the detection results of the ECG signal, so that the ECG detection results can more accurately and comprehensively reflect the heart status of the target user.

[0107] In addition, when the electronic device executing the method of this embodiment is an electrocardiogram detection device, the method of this embodiment can be used to detect the heart state of the target user non-invasively and non-destructively, realize real-time and continuous electrocardiogram abnormality monitoring, and provide the target user with a portable electrocardiogram abnormality monitoring solution, which is convenient for the target user's daily use.

[0108] On this basis, the electronic device can be a product in the form of a watch, a bracelet, a ring, etc.

[0109] <Device Example>

[0110] The present disclosure also provides an electrocardiogram detection device, such as Figure 4 As shown, the electrocardiogram detection device 4000 may include a signal text acquisition module 4100 , a feature vector determination module 4200 , a text information encoding module 4300 and an electrocardiogram signal detection module 4400 .

[0111] The signal text acquisition module 4100 is used to acquire the target user's electrocardiogram signal and the target user's case text information.

[0112] The feature vector determination module 4200 is used to determine the ECG feature vector of the ECG signal.

[0113] The text information encoding module 4300 is used to encode the case text information to obtain a semantic vector.

[0114] The ECG signal detection module 4400 is configured to obtain a detection result of the ECG signal according to the semantic vector and the ECG feature vector.

[0115] In some embodiments, the feature vector determination module 4200 is configured to:

[0116] Acquiring global waveform features of the electrocardiogram signal;

[0117] The electrocardiogram feature vector is obtained according to the global waveform feature.

[0118] In some embodiments, obtaining the global waveform characteristics of the electrocardiogram signal includes:

[0119] determining a plurality of heart beats of the electrocardiogram signal;

[0120] Extract local waveform features of each heartbeat;

[0121] The local waveform features of multiple heartbeats are processed to obtain the global waveform features of the electrocardiogram signal.

[0122] In some embodiments, the electrocardiogram detection device 4000 further includes:

[0123] A module for removing the local waveform features corresponding to the to-be-detected waveband when the duration of the to-be-detected waveband is greater than or equal to a set time threshold for each heartbeat.

[0124] In some embodiments, the feature vector determination module 4200 is configured to:

[0125] Encoding the electrocardiogram signal through a deep learning model based on an attention mechanism to obtain an encoding vector of the electrocardiogram signal;

[0126] The electrocardiogram feature vector is also obtained according to the encoding vector.

[0127] In some embodiments, the ECG signal detection module 4400 is used to:

[0128] Processing the ECG feature vector and the semantic vector through a classification model based on an attention mechanism to obtain a classification result of the ECG signal;

[0129] A detection result of the electrocardiogram signal is obtained according to the classification result.

[0130] In some embodiments, the text information encoding module 4300 is used to:

[0131] The case text information is encoded based on a large language model to obtain the semantic vector.

[0132] <Electronic Equipment Example>

[0133] This embodiment provides an electronic device. In one aspect, the electronic device may include the aforementioned electrocardiogram detection device 4000 .

[0134] On the other hand, Figure 5 As shown, the electronic device 5000 may include a processor 5100 and a memory 5200, the memory 5200 is used to store computer programs, and the processor 5100 is used to control the electronic device to execute the method of any embodiment of the present disclosure under the control of the computer program.

[0135] <Readable Storage Medium Embodiment>

[0136] This embodiment provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the method described in any method embodiment of the present disclosure is executed.

[0137] The present invention may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0138] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0139] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0140] The computer program instructions for performing the operation of the present invention can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, and conventional procedural programming languages such as "C" language or similar programming languages. The computer readable program instructions can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), is personalized by utilizing the state information of the computer readable program instructions, and the electronic circuit can execute the computer readable program instructions, thereby realizing various aspects of the present invention.

[0141] Various aspects of the present invention are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0142] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0143] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0144] The flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of an instruction, and the module, program segment or part of the instruction contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

[0145] While various embodiments of the present invention have been described above, the foregoing description is intended to be illustrative, non-exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or technological improvements in the marketplace, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of the present invention is defined by the appended claims.

Claims

1. A method for detecting an electrocardiogram, characterized in that: include: Obtaining an electrocardiogram (ECG) signal and medical record information of a target user; Determining an electrocardiographic characteristic vector of the electrocardiographic signal; Encoding the case text information to obtain a semantic vector; A detection result of the electrocardiogram signal is obtained according to the semantic vector and the electrocardiogram feature vector.

2. The method according to claim 1, characterized in that The determining of the electrocardiographic characteristic vector of the electrocardiographic signal includes: Acquiring global waveform features of the electrocardiogram signal; The electrocardiogram feature vector is obtained according to the global waveform feature.

3. The method according to claim 2, characterized in that The obtaining of the global waveform characteristics of the electrocardiogram signal includes: determining a plurality of heart beats of the electrocardiogram signal; Extract local waveform features of each heartbeat; The local waveform features of multiple heartbeats are processed to obtain the global waveform features of the electrocardiogram signal.

4. The method according to claim 3, characterized in that The method further comprises: For each heartbeat, when the duration of the to-be-detected band is greater than or equal to a set time threshold, the local waveform feature corresponding to the to-be-detected band is eliminated.

5. The method according to claim 1, wherein The determining of the electrocardiographic characteristic vector of the electrocardiographic signal further includes: Encoding the electrocardiogram signal through a deep learning model based on an attention mechanism to obtain an encoding vector of the electrocardiogram signal; The electrocardiogram feature vector is also obtained according to the encoding vector.

6. The method according to claim 1, characterized in that Obtaining a detection result of the electrocardiogram signal according to the semantic vector and the electrocardiogram feature vector includes: Processing the ECG feature vector and the semantic vector through a classification model based on an attention mechanism to obtain a classification result of the ECG signal; A detection result of the electrocardiogram signal is obtained according to the classification result.

7. The method according to claim 1, characterized in that The encoding process of the case text information to obtain a semantic vector includes: The case text information is encoded based on a large language model to obtain the semantic vector.

8. An electrocardiogram detection device, characterized in that: include: A signal text acquisition module is used to acquire the target user's ECG signal and the target user's case text information; a feature vector determination module, configured to determine an electrocardiogram feature vector of the electrocardiogram signal; A text information encoding module, used to encode the case text information to obtain a semantic vector; The ECG signal detection module is used to obtain a detection result of the ECG signal according to the semantic vector and the ECG feature vector.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the method according to any one of claims 1 to 7 under the control of the computer program.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, which implements the method according to any one of claims 1 to 7 when executed by a processor.

Images