Method, program and device for restoring errors in electrocardiogram signals

Abstract

According to one embodiment of the present disclosure, an error restoration method for an electrocardiogram signal performed by a computing device may include a step of acquiring an electrocardiogram signal, a step of estimating an error in the acquired electrocardiogram signal based on a correlation between leads for measuring the acquired electrocardiogram signal, and a step of restoring the error-estimated electrocardiogram signal based on restoration information for correcting the error in the electrocardiogram signal.

Description

[Technical Field]

[0001] The present disclosure relates to an error restoration method for an electrocardiogram signal, and more particularly to a method for detecting errors occurring during the measurement of an electrocardiogram signal and restoring the electrocardiogram signal based on the correlation of the electrocardiogram signal. [Background technology]

[0002] Clinical examinations and imaging tests are used to check for cardiac abnormalities. In particular, for early diagnosis of cardiac diseases, a method is widely used in which an electrocardiogram is measured, the measured electrocardiogram signal is displayed in the form of a graph, and the presence or absence of cardiac abnormalities in a patient is determined based on the graph.

[0003] An electrocardiogram (ECG) records the electrical current generated in the myocardium by the heart's beating as a wave form. The 12-lead ECG method measures signals by attaching 10 electrodes to the patient's body. Of the 10 electrodes, six (V1-V6) are chest lead electrodes that are attached to specific anatomical locations using the unipolar chest lead method. The remaining four electrodes are attached to the patient's limbs, namely the left hand and foot and the right hand and foot. However, the electrodes attached to the left hand and foot and the right hand and foot may also be attached to the left and right upper and lower chest, depending on the situation.

[0004] Generally, to measure an electrocardiogram signal, the attachment positions of 10 electrocardiogram electrodes must be found on the patient's body and attached one by one. However, it is not easy to find the correct positions to attach the 10 electrodes on the patient, and even experienced medical personnel often attach the electrodes incorrectly, resulting in the measurement of signals that are not diagnostic or an incorrect diagnosis.

[0005] Therefore, in order to attach electrocardiogram electrodes and measure correct signals, a method is needed that can quickly provide a normal electrocardiogram signal by determining whether the electrocardiogram electrodes are positioned backward and then restoring the signal where lead reversal has occurred, even if the position of the electrocardiogram electrodes is reversed. Summary of the Invention [Problem to be solved by the invention]

[0006] The present disclosure has been made in response to the above-mentioned background art, and aims to provide a method for determining an error, such as a lead reversal case caused by incorrectly attached electrocardiogram electrodes, based on an electrocardiogram signal, and for restoring the signal determined to be erroneous to a correct signal.

[0007] The present disclosure aims to provide a method for classifying erroneous electrocardiogram signals, such as lead reversal cases, using a trained deep learning model.

[0008] However, the problems to be solved by the present disclosure are not limited to those mentioned above, and other problems not mentioned will be clearly understood from the description below. [Means for solving the problem]

[0009] According to one embodiment of the present disclosure for achieving the above-mentioned object, an error restoration method for an electrocardiogram signal executed by a computing device may include a step of acquiring an electrocardiogram signal, a step of estimating an error in the acquired electrocardiogram signal based on a correlation between leads for measuring the acquired electrocardiogram signal, and a step of restoring the error-estimated electrocardiogram signal based on restoration information for correcting the error in the electrocardiogram signal.

[0010] Alternatively, the step of estimating an error in the acquired electrocardiogram signal based on a correlation between the leads for measuring the acquired electrocardiogram signal may include the steps of: generating a correlation matrix indicating an electrocardiogram signal relationship between the leads based on the acquired electrocardiogram signal; and detecting an electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals based on the correlation matrix.

[0011] Alternatively, the step of detecting an electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals based on the correlation matrix may include inputting the correlation matrix into a pre-trained first neural network model and classifying an electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals.

[0012] Alternatively, the step of estimating an error in the acquired electrocardiogram signal based on the correlation between the leads for measuring the acquired electrocardiogram signal may include inputting the electrocardiogram signal to a pre-trained second neural network model and classifying the electrocardiogram signal in which lead reversal has occurred among the electrocardiogram signals.

[0013] Alternatively, the step of estimating an error in the acquired electrocardiogram signal based on a correlation between the leads for measuring the acquired electrocardiogram signal may include the step of extracting a signal of interest from the acquired electrocardiogram signal for analyzing the correlation using a bandpass filter.

[0014] Alternatively, the reconstruction information may include a reconstruction table indicating a conversion relationship between an electrocardiogram signal of a normal lead and an electrocardiogram signal in which lead reversal has occurred.

[0015] Alternatively, the electrodes attached to the body to acquire the electrocardiogram signals may include at least one of a plurality of electrodes attached to the extremities or a plurality of electrodes attached to predetermined locations on the chest.

[0016] Alternatively, the acquired electrocardiogram signal may include one or more of a first electrocardiogram signal generated based on a first electrode corresponding to the right arm and a second electrode corresponding to the left arm, a second electrocardiogram signal generated based on the first electrode and a third electrode corresponding to the left leg, a third electrocardiogram signal generated based on the second electrode and the third electrode, or a chest electrocardiogram signal corresponding to each of a plurality of chest electrodes.

[0017] According to one embodiment of the present disclosure for achieving the above-mentioned object, a computer program for performing an operation for error recovery of an electrocardiogram signal may include an operation of acquiring an electrocardiogram signal, an operation of estimating an error in the acquired electrocardiogram signal based on a correlation between leads for measuring the acquired electrocardiogram signal, and an operation of recovering the electrocardiogram signal with the estimated error based on recovery information for correcting the error in the electrocardiogram signal.

[0018] According to one embodiment of the present disclosure for achieving the above object, a computing device for error reconstruction of an electrocardiogram signal may include a processor including at least one core, a memory including program code executable by the processor, and a network unit for acquiring an electrocardiogram signal. The processor may acquire an electrocardiogram signal, estimate an error in the acquired electrocardiogram signal based on a correlation between leads for measuring the acquired electrocardiogram signal, and reconstruct the error-estimated electrocardiogram signal based on reconstruction information for correcting the error in the electrocardiogram signal. [Effects of the Invention]

[0019] The present disclosure can detect lead reversal cases even when electrodes for measuring electrocardiogram signals are incorrectly attached, and can restore and provide a correct electrocardiogram signal to the user.

[0020] According to the present disclosure, when an electrocardiogram signal is erroneously measured, misdiagnosis can be prevented by specifically identifying and correcting what the lead reversal case is.

[0021] The present disclosure provides a method for restoring an electrocardiogram signal to a user without additional measurements when lead reversal occurs due to incorrect attachment of electrocardiogram electrodes. [Brief explanation of the drawings]

[0022] [Figure 1] FIG. 1 is a block diagram of a computing device according to one embodiment of the present disclosure.

[0023] [Figure 2] FIG. 2 is a block diagram illustrating an error recovery process for an electrocardiogram signal according to an embodiment of the present disclosure.

[0024] [Figure 3] FIG. 1 is a diagram illustrating electrocardiogram electrode attachment locations and corresponding leads according to one embodiment of the present disclosure.

[0025] [Figure 4] 10 is a table illustrating signals generated during lead reversal according to one embodiment of the present disclosure.

[0026] [Figure 5] FIG. 1 is a diagram illustrating a neural network model according to an embodiment of the present disclosure.

[0027] [Figure 6] 1 is a flowchart illustrating a method for electrocardiogram signal error recovery according to one embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0028] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement the present disclosure. The embodiments presented in this disclosure are provided to enable those skilled in the art to use or practice the contents of the present disclosure. Therefore, various modifications to the embodiments of the present disclosure will be apparent to those skilled in the art. That is, the present disclosure may be embodied in various different forms and is not limited to the following embodiments.

[0029] Throughout the specification of the present disclosure, the same or similar reference numerals refer to the same or similar components. In addition, in order to clearly explain the present disclosure, reference numerals of parts that are not relevant to the explanation of the present disclosure may be omitted from the drawings.

[0030] The term "or" as used in this disclosure is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless otherwise specified or otherwise clear from the context in this disclosure, "X uses A or B" should be understood to mean one of the natural inclusive permutations. For example, unless otherwise specified or otherwise clear from the context in this disclosure, "X uses A or B" can be interpreted as either X uses A, X uses B, or X uses both A and B.

[0031] The term "and / or" as used in this disclosure must be understood to indicate and include all possible combinations of one or more of the associated listed concepts.

[0032] The terms "comprises" and / or "comprising" as used in this disclosure should be understood to mean that the specified features and / or components are present. However, the terms "comprises" and / or "comprising" should not be understood to exclude the presence or addition of one or more other features, other components and / or combinations thereof.

[0033] In this disclosure, unless otherwise specified or clear from the context as referring to the singular form, the singular should generally be construed as including "one or more."

[0034] The term "Nth (N is a natural number)" used in this disclosure can be understood as an expression used to distinguish components of the present disclosure from one another based on a predetermined criterion, such as functional, structural, or convenience of description. For example, in this disclosure, components that perform different functional roles can be classified as a first component or a second component. However, components that are substantially identical within the technical concept of the present disclosure but must be distinguished for convenience of description can also be classified as a first component or a second component.

[0035] The term "acquire" as used in this disclosure may be understood to mean not only receiving data from an external device or system via a wired or wireless communication network, but also generating data in an on-device form.

[0036] Meanwhile, the terms "module" or "unit" used in this disclosure may be understood to refer to an independent functional unit that processes computing resources, such as a computer-related entity, firmware, software or a portion thereof, hardware or a portion thereof, or a combination of software and hardware. Here, a "module" or "unit" may refer to a unit composed of a single element or a unit expressed as a combination or collection of multiple elements. For example, as a concept of connotation, a "module" or "unit" may refer to a hardware element or a collection of hardware elements of a computing device, an application program that performs a specific software function, a processing procedure implemented by executing software, or a collection of instructions for executing a program. Furthermore, as a broad concept, a "module" or "unit" may refer to a computing device itself that constitutes a system, or an application executed on a computing device. However, the above concepts are merely examples, and the concepts of a "module" or "unit" may be defined in various ways within the scope of what one skilled in the art can understand based on the contents of this disclosure.

[0037] The term "model" as used in this disclosure may be understood as a system implemented using mathematical concepts and language to solve a specific problem, a set of software units for solving a specific problem, or an abstract model of a processing process for solving a specific problem. For example, a neural network "model" may refer to a system implemented as a neural network that has problem-solving capabilities through learning. Here, a neural network may have problem-solving capabilities by optimizing parameters connecting nodes or neurons through learning. A neural network "model" may include a single neural network or a neural network ensemble that combines multiple neural networks.

[0038] As used in this disclosure, "data" may include "image," signals, etc. As used in this disclosure, the term "image" may refer to multidimensional data made up of discrete image elements. In other words, "image" may be understood as a term referring to a digital representation of an object that can be seen by the human eye. For example, "image" may refer to multidimensional data made up of elements that correspond to pixels in a two-dimensional image. "Image" may refer to multidimensional data made up of elements that correspond to voxels in a three-dimensional image.

[0039] As used herein, "lead reversal" refers to incorrect lead placement, which can occur when electrodes used for electrocardiogram measurement are misplaced. For example, in the case of lead reversal, where limb electrodes other than the neutral electrode are misplaced, certain lead signals may be altered, certain other lead signals may be inverted or rotated, and certain other lead signals may remain unchanged. For example, misplacement of the neutral electrode may distort not only limb lead signals but also precordial lead signals. In this case, lead signals may appear to resemble other lead signals or may appear compressed, with signal magnitudes approaching zero.

[0040] The explanations of the above terms are intended to aid in understanding the present disclosure. Therefore, unless the above terms are explicitly stated as matters limiting the contents of the present disclosure, care should be taken not to use them to limit the technical ideas of the contents of the present disclosure.

[0041] FIG. 1 is a block diagram of a computing device according to one embodiment of the present disclosure.

[0042] The computing device 100 according to an embodiment of the present disclosure may be a hardware device or part of a hardware device that performs comprehensive data processing and calculations, or may be a software-based computing environment connected via a communication network. For example, the computing device 100 may be a server that performs intensive data processing functions and shares resources, or a client that shares resources by interacting with the server. The computing device 100 may also be a cloud system in which multiple servers and clients interact with each other to comprehensively process data. The above description is merely an example of a type of computing device 100, and various types of computing devices 100 may be configured within the scope of what one skilled in the art would understand based on the contents of this disclosure.

[0043] 1, a computing device 100 according to an embodiment of the present disclosure may include a processor 110, a memory 120, and a network unit 130. However, since FIG. 1 is merely an example, the computing device 100 may include other components for implementing a computer environment. Also, the computing device 100 may include only some of the disclosed components.

[0044] The processor 110 according to an embodiment of the present disclosure may be understood as a component including hardware and / or software for performing computing operations. For example, the processor 110 may read a computer program to perform data processing for machine learning. The processor 110 may process operations such as input data processing for machine learning, feature extraction for machine learning, and error calculation based on backpropagation. The processor 110 for performing such data processing may include a central processing unit (CPU), a general-purpose graphics processing unit (GPGPU), a tensor processing unit (TPU), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc. The types of processor 110 described above are merely examples, and various types of processor 110 may be configured within the scope of what one skilled in the art would understand based on the present disclosure.

[0045] The processor 110 can detect errors in the electrocardiogram signal caused by incorrect placement of electrodes for electrocardiogram measurement. "Incorrect placement of electrodes" can be understood as the placement of electrodes for measuring the electrocardiogram signal of a particular lead out of their normal positions. When electrodes are incorrectly placed, the likelihood of measuring an electrocardiogram signal that is not diagnostic or of the measured electrocardiogram signal leading to an erroneous diagnosis increases. To prevent such problems, the processor 110 can detect errors in the electrocardiogram signal caused by incorrect placement of electrodes based on the measured electrocardiogram signal.

[0046] For example, the processor 110 may analyze an electrocardiogram signal to detect a lead reversal caused by misplaced electrodes. Specifically, the processor 110 may analyze correlations between leads used to measure the electrocardiogram signal based on the electrocardiogram signal. Here, the correlation between leads may be understood as an analysis result indicating the correlation between the electrocardiogram signals between the leads based on the measured lead-specific signals. The processor 110 may detect a lead reversal using the correlation between leads analyzed based on the measured electrocardiogram signal. The processor 110 may detect which lead's signal was measured as a reversal due to misplaced electrodes based on the correlation between leads. Here, the calculation process for detecting a lead reversal may be performed using logic based on rules defined by program code or may be performed based on a deep learning algorithm based on program code. Through this calculation process, the processor 110 may easily detect which reversal occurred in the measured signal of which lead for all leads used to measure the electrocardiogram signal.

[0047] The processor 110 can not only detect errors in the electrocardiogram signal but also restore the detected errors in the electrocardiogram signal. The processor 110 can correct the detected errors in the electrocardiogram signal based on restoration information determined in advance to correct the errors in the electrocardiogram signal. Here, the restoration information may be predefined information for converting lead-specific signals measured when the electrodes are placed in incorrect positions into lead-specific signals measured when the electrodes are placed in normal positions. That is, the processor 110 can use the restoration information to convert the electrocardiogram signal in which an error has been detected into a signal measured when the electrodes are placed in normal positions.

[0048] The process of detecting and restoring an error in an electrocardiogram signal executed by the processor 110 can minimize resources in a hospital environment required to ensure that an electrocardiogram signal can be normally used for diagnosing a disease, etc. For example, even if an electrocardiogram signal is measured with an incorrect electrode setting for measuring the electrocardiogram signal, the above-described error detection and restoration process can minimize the hassle of having to reset the electrode setting and measure the signal again.

[0049] The memory 120 according to an embodiment of the present disclosure may be understood as a component including hardware and / or software for storing and managing data processed by the computing device 100. That is, the memory 120 may store any type of data generated or determined by the processor 110 and any type of data received by the network unit 130. For example, the memory 120 may include at least one type of storage medium selected from the group consisting of flash memory, hard disk, multimedia card micro, card-type memory, random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, and optical disk. The memory 120 may also include a database system that manages data in a predetermined manner. The types of memory 120 described above are merely examples, and various configurations of the memory 120 are possible within the scope of what would be understood by one skilled in the art based on the present disclosure.

[0050] The memory 120 may structure and organize and manage data, data combinations, and program code executable by the processor 110 required for the processor 110 to perform calculations. For example, the memory 120 may store medical data received via the network unit 130 (described below). The memory 120 may store program code for operating a neural network model to receive medical data and perform learning, program code for operating the neural network model to receive medical data and perform inference according to the intended use of the computing device 100, and processed data generated by executing the program code.

[0051] The network unit 130 according to an embodiment of the present disclosure may be understood as a component that transmits and receives data via any type of known wired or wireless communication system. For example, the network unit 130 may transmit and receive data using a wired or wireless communication system such as a local area network (LAN), wideband code division multiple access (WCDMA), long term evolution (LTE), wireless broadband internet (WiBro), 5G, ultra wide-band, ZigBee, radio frequency (RF) communication, wireless LAN, wireless fidelity (WiFi), near field communication (NFC), or Bluetooth. The above-described communication systems are merely examples, and various wired and wireless communication systems for transmitting and receiving data by the network unit 130 may be applied in addition to the above examples.

[0052] The network unit 130 may receive data necessary for the processor 110 to perform calculations via wired or wireless communication with any system or any client. The network unit 130 may also transmit data generated by calculations by the processor 110 via wired or wireless communication with any system or any client. For example, the network unit 130 may receive medical data via communication with a database in a hospital environment, a cloud server that performs tasks such as standardizing medical data, or a computing device. The network unit 130 may transmit output data of the neural network model, and intermediate data and processed data derived during the calculation process of the processor 110 via communication with the database, server, computing device, or the like.

[0053] Figure 2 is a block diagram illustrating an error recovery process for an electrocardiogram signal according to one embodiment of the present disclosure, and Figure 4 is a table illustrating signals generated during lead reversal according to one embodiment of the present disclosure.

[0054] 2, a computing device 200 for error recovery of an electrocardiogram signal may include a reversal detection module 220 for detecting errors in input data 210 and a recovery module 230 for correcting errors in the input data 210. In the present disclosure, the input data 210 may include an electrocardiogram signal corresponding to at least one lead. The input data 210 may be generated by an apparatus for electrocardiogram measurement and transmitted to the computing device 200. If the computing device 200 includes electrodes for measuring the electrocardiogram signal, the input data 210 may also be generated directly by the computing device 200.

[0055] The reversal sensing module 220 may include a filter unit 222 , a correlation analysis unit 224 , and a case classification unit 226 .

[0056] The filter unit 222 can filter noise from the input data 210 to facilitate analysis of the input data 210. For example, the filter unit 222 can include a band pass filter that passes only components of a specific frequency band to be analyzed from the input data 210. The band pass filter can pass only signals from 0.5 Hz to 50 Hz from the input data 210. However, the range of 0.5 Hz to 50 Hz is merely an example, and the present disclosure is not limited to such values. The filter unit 222 can further include a notch filter to reduce hum, which is power supply noise generated from a power cable, for example.

[0057] The correlation analysis unit 224 may analyze correlations for each lead. The correlation analysis by the correlation analysis unit 224 may be performed based on the input data 210 or the output data of the filter unit 222. The correlation analysis unit 224 may analyze the correlation of electrocardiogram signals between the leads for each lead, thereby generating an N×N correlation matrix (N is a natural number) indicating the relationship between the electrocardiogram signals between the leads. For example, if the input data 210 includes 12-lead electrocardiogram signals, the correlation analysis unit 224 may analyze the morphological correlations between the 12 leads and generate a 12×12 correlation matrix. The calculation process for generating a 12×12 correlation matrix may be described in Equation 1 below.

[0058]

number

[0059] In Equation 1, X may be an electrocardiogram signal, C may be a channel or lead, T may be time, A may be an adjacency matrix, and E[X] may be an expectation value of the random variable X.

[0060] In this way, the correlation analysis unit 224 can organize the correlations between all leads into a single matrix based on the ECG signal. The correlation analysis unit 224 can express the signal relationships of all leads as a correlation matrix so that the case classification unit 226 can easily interpret complex signal relationships. In other words, the case classification unit 226 can easily interpret the characteristics of the signals for each lead using the matrix generated by the correlation analysis unit 224, and can effectively detect lead reversal.

[0061] The case classification unit 226 may determine whether or not a signal included in the input data 210 has a lead reversal and classify the lead reversal case based on the correlation matrix or correlation information generated by the correlation analysis unit 224. For example, the case classification unit 226 may determine whether or not a lead reversal exists by comparing the degree of agreement between the information included in the correlation matrix and a predetermined normal case. The case classification unit 226 may then classify the lead reversal case using the correlation matrix based on the determination result. The case classification unit 226 may also classify the lead reversal case included in the input data 210 based on the correlation matrix using a first neural network model. The first neural network model may include a convolutional neural network for receiving a visualized matrix and detecting a lead reversal case. The first neural network model may be trained based on supervised learning using a matrix of a normal case without a lead reversal as a label. However, the neural network model may also be trained based on unsupervised learning, reinforcement learning, or other learning methods in addition to supervised learning.

[0062] Meanwhile, according to an alternative embodiment of the present disclosure, the operations of the correlation analysis unit 224 and the case classification unit 226 may be replaced by a second neural network model. For example, the second neural network model may classify lead reversal cases based on the input data 210 or the output data of the filter unit 222. The second neural network model may extract features for interpreting correlations between lead-specific signals based on the input data 210 or the output data of the filter unit 222, and may be trained to classify lead reversal cases based on the features. Here, the second neural network model may include a neural network that performs operations based on self-attention.

[0063] The restoration module 230 can restore an ECG signal in which an error has been detected by the inversion detection module 220 based on predetermined restoration information. The restoration module 230 can correct the lead reversal case detected by the case classification unit 226 using the restoration information. The restoration module 230 can use a signal table corresponding to each lead reversal case (hereinafter referred to as the restoration table) as restoration information to restore the ECG signal. The restoration table can be information in the form of a table showing the conversion relationship between an ECG signal of a normal lead and an ECG signal in which a lead reversal has occurred. Here, "normal" can be understood as a state in which the electrodes are attached at positions determined for measuring a lead signal in which the electrodes are normally placed. The restoration table can be seen in FIG. 4. For example, according to the restoration table of FIG. 4, in the case of an LA / RA lead in which the left arm electrode and the right arm electrode are attached in a back-to-back order, Lead I is an inverted signal, and Leads II and III can be changed (Lead I → -(Lead I), Lead II → Lead III, Lead III → Lead II).

[0064] In the table of FIG. 4, "-" may mean a signal with reversed polarity, and "≒" may mean an approximately identical signal. CW may mean a rotation in the direction RA → LA → LL → RA, and CCW may mean a rotation in the direction RA → LL → LA → RA. The reconstruction module 230 can reconstruct an error-containing ECG signal by converting an ECG signal with lead reversal into an ECG signal with a normal lead based on a reconstruction table including the table of FIG. 4. Meanwhile, the ECG signal error reconstruction method of the present disclosure has been described focusing on the limb leads out of the 12 leads, but the application is not limited thereto and may also include ECG measurement methods with 5 leads, 3 leads, or other lead formats.

[0065] 2, the output data 240 may be data obtained by restoring an electrocardiogram signal in which a lead reversal has occurred to an electrocardiogram signal with a normal lead. The output data 240 may be transmitted to various types of clients and displayed on a display device.

[0066] The computing device 200 for recovering errors in an electrocardiogram signal may be used in a form combined with or included in an apparatus for measuring an electrocardiogram signal, or may be executed in a separate apparatus. The recovery method executed by the computing device 200 may be performed in a server or computing device located physically separate from the electrocardiogram signal measuring apparatus. The recovery method executed by the computing device 200 may be performed in a manner of receiving the measured signal via a network, analyzing it, and transmitting the analysis results again via the network.

[0067] FIG. 3 is a diagram illustrating electrocardiogram electrode attachment locations and corresponding leads according to one embodiment of the present disclosure.

[0068] 3, a standard 12-lead (or 12-lead) electrocardiogram can include limb leads and chest leads. The limb leads can include four electrodes attached to the limbs (hereinafter referred to as limb electrodes). The chest leads can include six electrodes attached to the chest (hereinafter referred to as chest electrodes).

[0069] The limb electrodes may include a right arm electrode RA, a left arm electrode LA, a right leg electrode RL, and a left leg electrode LL. The right leg electrode RL may be a common electrode or a ground electrode. The limb electrodes may be attached to positions corresponding to the right arm, left arm, right leg, and left leg, respectively.

[0070] The chest electrodes (or precordial electrodes) may include a first chest electrode V1, a second chest electrode V2, a third chest electrode V3, a fourth chest electrode V4, a fifth chest electrode V5, and a sixth chest electrode V6.

[0071] The limb leads can include standard limb leads I, II, and II and amplified limb leads aVR, aVL, and aVF. Among the limb leads, the standard limb leads are bipolar leads, so the voltage difference between two electrodes can represent the trace recorded on an electrocardiogram recorder. For example, the first lead L1 can represent the voltage difference between the right arm electrode RA and the left arm electrode LA, the second lead L2 can represent the voltage difference between the right arm electrode RA and the left leg electrode LL, and the third lead L3 can represent the voltage difference between the left arm electrode LA and the left leg electrode LL. The first electrocardiogram signal or first channel signal can be a signal generated in the first lead L1, the second electrocardiogram signal or second channel signal can be a signal generated in the second lead L2, and the third electrocardiogram signal or third channel signal can be a signal generated in the third lead L3. Among the limb leads, the amplified limb lead is a unipolar lead, so that the sixth electrocardiogram signal can be generated from the fourth electrocardiogram signals corresponding to the right arm electrode RA, the left arm electrode LA, and the left leg electrode LL, respectively.

[0072] Since the chest leads are unipolar leads, the twelfth electrocardiogram signal can be generated from the seventh electrocardiogram signal corresponding to each of the chest electrodes.

[0073] The above description is merely one example for a standard 12-lead electrocardiogram measurement, and the present disclosure is not limited to such example.

[0074] FIG. 5 is a diagram illustrating a neural network model according to an embodiment of the present disclosure.

[0075] 5, a neural network model 320 for detecting lead reversal can classify lead reversal cases 330 among electrocardiogram signals using correlation data 310 indicating the overall lead relevance for measurement of the electrocardiogram signals as an input, where the correlation data 310 can be the correlation matrix generated by the correlation analysis unit 224 of FIG.

[0076] The neural network model 320 can interpret the characteristics of the lead-by-lead ECG signal relationship by receiving a correlation matrix indicating the ECG signal relationship between the leads. The neural network model 320 can determine whether lead reversal exists among the signals present in the input during the feature interpretation process. The neural network model 320 can then classify the input signals as a lead reversal case based on the determination result.

[0077] For example, the neural network model 320 can classify lead reversal cases for limb leads among 12 leads based on a correlation matrix. The algorithm for classifying lead reversal cases for limb leads among 12 leads can be found in Equation 2 below.

[0078]

number

[0079] Z can be a correlation matrix for three limb leads in an ECG signal to determine lead reversal. θ can be a parameter of a neural network model to classify lead reversal cases. For example, flatten() can be a function that flattens a multidimensional array space into one dimension.

[0080] FIG. 6 is a flowchart illustrating an electrocardiogram signal error recovery method according to one embodiment of the present disclosure.

[0081] 1 and 6, a computing device 100 according to an embodiment of the present disclosure may acquire an electrocardiogram signal (S110). The computing device 100 may be connected to an electrocardiogram measurement device and may use the measured electrocardiogram signal or an electrocardiogram signal transmitted via a network. Here, the electrocardiogram signal acquired by the computing device 100 may include an electrocardiogram signal measured in a standard 12-lead format, and an electrocardiogram signal measured in a 5-lead, 3-lead, or various other lead formats.

[0082] For example, electrodes attached to the body to acquire electrocardiogram signals may include at least one of a plurality of limb electrodes attached to the extremities of the limbs in a standard 12-lead format and a plurality of chest electrodes attached to predetermined locations on the chest.

[0083] For example, the electrocardiogram signal may include one or more of a first electrocardiogram signal generated based on a first electrode corresponding to the right arm and a second electrode corresponding to the left arm, a second electrocardiogram signal generated based on the first electrode and a third electrode corresponding to the left leg, or a third electrocardiogram signal generated based on the second electrode and the third electrode in a standard 12-lead format.

[0084] The computing device 100 may estimate an error in the electrocardiogram signal based on the correlation between the leads (S120). The computing device 100 may generate a correlation matrix indicating the relationship between the electrocardiogram signals between the leads based on the acquired electrocardiogram signals. The computing device 100 may then detect an electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals based on the correlation matrix. The computing device 100 may generate the correlation matrix by preprocessing the input signal or by taking into account weights, etc., in the process of calculating the correlation between the electrocardiogram signals for each lead.

[0085] In the process of detecting an electrocardiogram signal in which lead reversal has occurred among acquired electrocardiogram signals based on the correlation matrix, the computing device 100 may input the correlation matrix to a first pre-trained neural network model to classify the electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals. The computing device 100 may input the electrocardiogram signal to a second pre-trained neural network model based on self-attention to classify the electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals.

[0086] The computing device 100 can use a bandpass filter to remove noise from the electrocardiogram signal and extract a signal of interest from the acquired electrocardiogram signal for analyzing the correlation.

[0087] The computing device 100 may restore the ECG signal in which an error is estimated and provide the restored signal to a user (S130). The computing device 100 may restore the ECG signal in which lead reversal is estimated by using a restoration table that indicates a conversion relationship between an ECG signal of a normal lead and an ECG signal in which lead reversal has occurred. For example, the restoration table may include the table illustrated in FIG. 4.

[0088] According to an embodiment of the present disclosure, even if lead reversal occurs due to incorrectly attached ECG electrodes, information about the incorrectly attached electrodes and the case of lead reversal can be accurately determined. Furthermore, according to the embodiment of the present disclosure, an ECG signal can be quickly provided without reattaching electrodes or re-measuring the signal.

[0089] The various embodiments of the present disclosure described above can be combined with additional embodiments and can be modified within the scope that can be understood by those skilled in the art from the above detailed description. The embodiments of the present disclosure are illustrative in all respects and should not be construed as limiting. For example, each component described as a single type can also be implemented in a distributed form, and similarly, components described as distributed can also be implemented in a combined form. Therefore, all modifications and variations derived from the meaning, scope, and equivalent concepts of the claims of the present disclosure should be construed as being within the scope of the present disclosure. [Explanation of symbols]

[0090] 22: Filter unit 100: Computing equipment 110: Processor 120: Memory 130: Network Department 200: Computing equipment 210: Input data 220: Reverse sensing module 222: Filter unit 224: Correlation Analysis Unit 226: Case Classification Unit 230: Recovery module 240: Output data 310: Correlation Data 320: Neural Network Model 330: Lead reversal case

Claims

1. 1. A method for error recovery of electrocardiogram signals, performed by a computing device including at least one processor, comprising: acquiring electrocardiogram signals including a plurality of leads from electrocardiogram electrodes; estimating an error in the acquired electrocardiogram signal caused by mispositioning of electrocardiogram electrodes based on the correlation between the leads; reconstructing the error-estimated electrocardiogram signal based on reconstruction information for correcting the error in the electrocardiogram signal; Including, wherein the reconstruction information includes a transformation relationship between a normal electrocardiogram signal and an error-inducing electrocardiogram signal.

2. estimating an error in the acquired electrocardiogram signal based on correlations between the leads, generating a correlation matrix indicating electrocardiogram signal relationships between the leads; determining lead reversals among the acquired electrocardiogram signals based on the correlation matrix; detecting an electrocardiogram signal in which a reversal has occurred; The method of claim 1 , comprising:

3. 3. The method of claim 2, wherein detecting an electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals based on the correlation matrix includes inputting the correlation matrix into a pre-trained first neural network model to classify an electrocardiogram signal in which lead reversal has occurred among the acquired electrocardiogram signals.

4. 2. The method of claim 1, wherein estimating an error in the acquired electrocardiogram signals based on the correlation between the leads includes inputting the electrocardiogram signals to a pre-trained second neural network model and classifying the electrocardiogram signals as having lead reversal.

5. 10. The method of claim 1, wherein estimating an error in the acquired electrocardiogram signal based on the correlation between the leads comprises extracting a signal of interest from the acquired electrocardiogram signal using a band pass filter for analyzing the correlation.

6. The method of claim 1 , wherein the reconstruction information includes a reconstruction table indicating a conversion relationship between electrocardiogram signals of normal leads and electrocardiogram signals in which lead reversal has occurred.

7. 2. The method of claim 1, wherein the electrodes attached to the body to acquire the electrocardiogram signals include at least one of a plurality of electrodes attached to the extremities or a plurality of electrodes attached to predetermined locations on the chest.

8. 2. The method of claim 1, wherein the acquired electrocardiogram signals include one or more of a first electrocardiogram signal generated based on a first electrode corresponding to a right arm and a second electrode corresponding to a left arm, a second electrocardiogram signal generated based on the first electrode and a third electrode corresponding to a left leg, a third electrocardiogram signal generated based on the second electrode and the third electrode, or a chest electrocardiogram signal corresponding to each of a plurality of chest electrodes.

9. A computer program stored on a computer-readable storage medium, the computer program being adapted to run on one or more processors. ) performs an operation for error recovery of the electrocardiogram signal, The operation is acquiring electrocardiogram signals including a plurality of leads from electrocardiogram electrodes; estimating an error in the acquired electrocardiogram signal caused by mispositioning of electrocardiogram electrodes based on correlation between the leads; an operation of reconstructing the error-estimated electrocardiogram signal based on reconstruction information for correcting the error in the electrocardiogram signal; Including, wherein the reconstruction information includes a transformation relationship between a normal electrocardiogram signal and an error-occurring electrocardiogram signal; Computer program.

10. 1. A computing device for error reconstruction of electrocardiogram signals, comprising: a processor including at least one core; a memory containing program code executable by the processor; a network unit for acquiring an electrocardiogram signal; Including, The device is characterized in that the processor acquires an electrocardiogram signal including multiple leads from electrocardiogram electrodes, estimates an error in the acquired electrocardiogram signal caused by incorrect placement of the electrocardiogram electrodes based on correlation between the leads, and restores the electrocardiogram signal with the estimated error based on restoration information for correcting the error in the electrocardiogram signal, wherein the restoration information includes a conversion relationship between a normal electrocardiogram signal and an electrocardiogram signal with an error.

Images