Apparatus and method for non-contact detecting arrhythmia based on radar
The ultra-wideband radar-based arrhythmia detection method generates a trajectory matrix, decomposes and filters radar signals to accurately classify arrhythmia, addressing the limitations of traditional methods and achieving comparable accuracy to electrocardiograms.
Patent Information
- Application Number
- KR1020230129639
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2022-09-30
- Filing Date
- 2023-09-26
- Publication Date
- 2026-11-06
- Estimated Expiration
- 2043-09-26
AI Technical Summary
Existing arrhythmia detection methods using electrocardiograms are cumbersome and require skilled personnel, while ultrasound is expensive and limited in accessibility, and current UWB-based methods lack the ability to accurately detect arrhythmia by analyzing traditional electrocardiogram waveforms.
A non-contact arrhythmia detection method using ultra-wideband radar that generates a first trajectory matrix from radar signals, decomposes it using Singular Value Decomposition (SVD), filters frequency bands, and reconstructs heart signals to classify arrhythmia based on R peak intervals.
Provides a non-contact, accurate arrhythmia detection method that matches traditional electrocardiogram performance by classifying arrhythmia patients with a high match rate and correlation, overcoming the limitations of existing technologies.
Smart Images

Figure 112023107216893-PAT00059_ABST
Abstract
Description
Technology Field
[0001] The disclosed embodiments relate to a non-contact arrhythmia detection technology based on ultra-wideband radar. Background Technology
[0002] Arrhythmia is a serious condition cited as a cause of sudden cardiac death and a major cause of stroke, requiring prompt examination. Arrhythmia is diagnosed relying on electrocardiograms (ECGs) and ultrasound.
[0003] However, electrocardiograms restrict patient movement due to electrodes attached to the patient and can cause contact infections. Ultrasound is expensive equipment and must be operated by skilled medical personnel.
[0004] To address the aforementioned problems, some studies incorporate Ultra-WideBand (UWB) technology into arrhythmia detection methods. However, some studies estimate arrhythmia by measuring the heart rate or the time difference between heartbeats.
[0005] In other words, current studies merely move away from traditional methods and adopt modified, simplified approaches to integrate with UWB technology. This is because traditional methods detect arrhythmias by analyzing the R peak interval, P-wave characteristics, PR interval, and QRS interval.
[0006] Since the characteristics of some electrocardiogram waveforms require consideration of the heart's electrical properties, electrocardiogram measurement is inevitable. However, although the R peak interval can be detected through UWB technology, there is no relevant research.
[0007] In other words, despite existing research, there is still a need for a new arrhythmia detection method that can overcome the shortcomings of measurement devices while adhering to traditional arrhythmia detection methods. Prior art literature
[0008] Republic of Korea Registered Patent No. 10-1895324 (Published September 5, 2018) The problem to be solved
[0009] The disclosed embodiments are intended to provide a radar-based non-contact arrhythmia detection method and apparatus. means of solving the problem
[0010] An arrhythmia detection device according to one embodiment comprises: one or more processors; and a memory for storing one or more programs executed by the one or more processors, wherein the one or more processors: acquire raw data of a certain period of ultra-wideband radar signal reflected from a subject, generate a first trajectory matrix based on time series data converted from the raw data, decompose the first trajectory matrix to generate a second trajectory matrix, convert the second trajectory matrix to extract a heart signal, and classify the subject as an arrhythmia patient when a specific wavelength interval in the heart signal is greater than or equal to a preset value.
[0011] The above one or more processors can identify the chest wall position of the subject based on the maximum dynamic variation included in the raw data.
[0012] One or more processors above can generate the first trajectory matrix by arranging the time series data into the components of the first trajectory matrix based on the following [Equation 1] such that the components of the remaining diagonals perpendicular to the main diagonal of the first trajectory matrix are all identical.
[0013] [Mathematical Formula 1]
[0014]
[0015] (Here, H is the first locus matrix, x a is the a-th time series data, L is the length of the time series data, n and m are arbitrary natural numbers, provided that n = L - m + 1 and 2 ≤ m ≤ n)
[0016] One or more processors above can calculate the covariance matrix of the first trajectory matrix based on the following [Equation 2].
[0017] [Mathematical Formula 2]
[0018]
[0019] (Here, C is the covariance matrix, H is the first locus matrix, H T is the transpose of the first trajectory matrix, and n is the number of time series data.
[0020] The above one or more processors can decompose the first trajectory matrix into a plurality of second decomposition matrices based on a plurality of first decomposition matrices corresponding to the covariance matrix, which are produced by performing Singular Value Decomposition (SVD) on the covariance matrix of the first trajectory matrix.
[0021] The above one or more processors are: S matrix U matrix and V included in the plurality of first decomposition matrices. T Using a matrix, the above multiple second decomposition matrices can be calculated based on the following [Equation 3].
[0022] [Mathematical Formula 3]
[0023]
[0024] (Here, is the i-th second decomposition matrix among the plurality of second decomposition matrices above, is the i-th eigenvalue of the above covariance matrix, is the i-th column vector of the above U matrix, is the i-th row vector of matrix V)
[0025] The above one or more processors may: select some of the plurality of second decomposition matrices in a predetermined number according to the magnitude of the heart signal, filter matrices other than the frequency band so that the heart signal corresponding to the some matrices satisfies a predetermined frequency band, and generate the second trajectory matrix based on the remaining matrices after filtering among the some matrices.
[0026] One or more of the above processors can: reconstruct the heart signal extracted from the second trajectory matrix based on the following [Equation 4].
[0027] [Mathematical Formula 4]
[0028] )=
[0029] (Here, is the reconstructed k-th cardiac signal, k is the number of cardiac signals, and m and n are the number of rows or columns of the above trajectory matrix)
[0030] One or more processors above may classify the subject as an arrhythmia patient if the R peak interval in the heart signal is greater than or equal to a preset value.
[0031] A method for detecting arrhythmia according to one embodiment is performed by an arrhythmia detection device comprising one or more processors and a memory for storing one or more programs executed by the one or more processors, and comprises the steps of: acquiring raw data of a constant period ultra-wideband radar signal reflected from a subject; generating a first trajectory matrix based on time series data converted from the raw data; decomposing the first trajectory matrix to generate a second trajectory matrix; converting the second trajectory matrix to extract a heart signal; and classifying the subject as an arrhythmia patient when a specific wavelength interval in the heart signal is greater than or equal to a preset value.
[0032] The above method may further include the step of identifying the chest wall position of the subject based on the maximum dynamic variation included in the raw data.
[0033] The step of generating the first trajectory matrix may include the step of generating the first trajectory matrix by arranging the time series data as components of the first trajectory matrix based on the following [Equation 1] so that the components of the remaining diagonals perpendicular to the main diagonal of the first trajectory matrix are all identical.
[0034] [Mathematical Formula 1]
[0035]
[0036] (Here, H is the first locus matrix, is the a-th time series data, L is the length of the time series data, n and m are arbitrary natural numbers, provided that n = L - m + 1 and 2 ≤ m ≤ n)
[0037] The step of generating the second trajectory matrix may include the step of calculating the covariance matrix of the first trajectory matrix based on the following [Equation 2].
[0038] [Mathematical Formula 2]
[0039]
[0040] (Here, C is the covariance matrix, H is the first locus matrix, H T is the transpose of the first trajectory matrix, and n is the number of time series data.
[0041] The step of generating the second trajectory matrix may include the step of decomposing the first trajectory matrix into a plurality of second decomposition matrices based on a plurality of first decomposition matrices corresponding to the covariance matrix, which are obtained by performing Singular Value Decomposition (SVD) on the covariance matrix of the first trajectory matrix.
[0042] The step of generating the second trajectory matrix comprises the S matrix, U matrix, and V included in the plurality of first decomposition matrices. T It may include a step of calculating the plurality of second decomposition matrices using the matrix based on the following [Equation 3].
[0043] [Mathematical Formula 3]
[0044]
[0045] (Here, is the i-th second decomposition matrix among the plurality of second decomposition matrices above, is the i-th eigenvalue of the above covariance matrix, is the i-th column vector of the above U matrix, is the i-th row vector of the above V matrix)
[0046] The step of generating the second trajectory matrix can include: selecting a predetermined number of some matrices among the plurality of second decomposition matrices in order of the magnitude of the heart signal; filtering matrices other than the frequency band so that the heart signal corresponding to the some matrices satisfies the predetermined frequency band; and generating the second trajectory matrix based on the remaining matrices after filtering among the some matrices.
[0047] The step of extracting the heart signal may further include the step of reconstructing the heart signal extracted from the second trajectory matrix based on the following [Equation 4].
[0048] [Mathematical Formula 4]
[0049] )=
[0050] (Here, is the reconstructed k-th cardiac signal, k is the number of cardiac signals, and m and n are the number of rows or columns of the above trajectory matrix)
[0051] The above-mentioned classification step may include a step of classifying the subject as an arrhythmia patient if the R peak interval in the heart signal is greater than or equal to a preset value. Effects of the invention
[0052] The disclosed embodiments can provide a non-contact arrhythmia detection method by estimating an electrocardiogram signal from an ultra-wideband radar signal using a trajectory matrix.
[0053] The disclosed embodiments can provide a non-contact arrhythmia detection method that follows a traditional arrhythmia diagnosis method by determining the R peak interval in the estimated cardiac rate signal. Brief explanation of the drawing
[0054] FIG. 1 is a block diagram illustrating an arrhythmia detection device according to one embodiment. FIG. 2 is an exemplary diagram illustrating the output of an arrhythmia detection device according to one embodiment. FIG. 3 is a graph illustrating the performance of an arrhythmia detection device according to one embodiment. FIG. 4 is a flowchart illustrating an arrhythmia detection method according to one embodiment. Specific details for implementing the invention
[0055] Hereinafter, a specific embodiment of one embodiment will be described with reference to the drawings. The following detailed description is provided to facilitate a comprehensive understanding of the sensor described herein. However, this is merely an example and the present invention is not limited thereto.
[0056] In describing the embodiments, if it is determined that a detailed description of known technology related to the present invention may unnecessarily obscure the gist of the embodiment, such detailed description will be omitted. Additionally, numbers used in the description of the embodiments (e.g., first, second, etc.) are merely identification symbols to distinguish one component from another.
[0057] FIG. 1 is a block diagram for explaining an arrhythmia detection device (100) according to one embodiment.
[0058] Referring to FIG. 1, an arrhythmia detection device (100) according to one embodiment includes a processor (110) and a memory (120).
[0059] The processor (110) acquires raw data of a periodic ultra-wideband radar signal reflected from the subject.
[0060] Here, ultra-wideband radar is a next-generation sensor that perceives the surrounding environment by transmitting an impulse signal from a transmitting antenna and receiving the signal reflected back from the subject via a receiving antenna.
[0061] That is, the processor (110) can acquire raw data of an impulse-unit ultra-wideband radar signal transmitted by the ultra-wideband radar and reflected to the subject. At this time, the raw data may be composed of a frame signal including signal strength.
[0062] The processor (110) can identify the location of the subject's chest wall based on the maximum dynamic variation detected in the raw data.
[0063] The processor (110) can generate a first trajectory matrix based on time series data converted from raw data.
[0064] At this time, the time series data may be data converted from raw data related to the chest wall position. From this, the processor (110) can extract a heart signal by filtering out noise unrelated to the heart signal (e.g., the subject's breathing, noise caused by the external environment) from the raw data.
[0065] The processor (110) can generate a first trajectory matrix by arranging the time series data converted from raw data as components of the matrix.
[0066] Specifically, the processor (110) can generate a first trajectory matrix by arranging the time series data to be converted into raw data such that the components of the remaining diagonals perpendicular to the main diagonal of the matrix are all identical.
[0067] For example, the processor (110) can arrange time series data based on the following [Equation 1] such that the components of the remaining diagonals perpendicular to the main diagonal of the first trajectory matrix are all identical.
[0068] [Mathematical Formula 1]
[0069]
[0070] Here, H is the first locus matrix, x a represents the a-th time series data, L is the length of the time series data, and n and m can represent any natural numbers. However, n = L - m + 1 and 2 ≤ m ≤ n. Meanwhile, a value of [L / 2] for m is preferred.
[0071] The processor (110) generates a second trajectory matrix based on the first trajectory matrix.
[0072] The processor (110) can generate a second trajectory matrix based on the covariance matrix of the first trajectory matrix.
[0073] First of all, the processor (110) can calculate the covariance matrix of the first trajectory matrix based on the following [Equation 2].
[0074] [Mathematical Formula 2]
[0075]
[0076] Here, C is the covariance matrix, H is the first locus matrix, and H T is the transpose of the first locus matrix, and n can represent the number of time series data.
[0077] The processor (110) can generate a second trajectory matrix by performing Singular Value Decomposition (SVD) on the covariance matrix of the first trajectory matrix.
[0078] The processor (110) can generate a second trajectory matrix based on a plurality of first decomposition matrices calculated by performing singular value decomposition on the covariance matrix of the first trajectory matrix.
[0079] Specifically, the processor (110) comprises a plurality of first decomposition matrices, such as S matrix, U matrix, V T At least one of the matrix and the V matrix can be produced.
[0080] Here, the S matrix is the eigenvalues of the covariance matrix ( , , ..., It is a diagonal matrix in which ) is placed as the main diagonal element. In this case, the eigenvalues can be arranged in descending order.
[0081] The U matrix is a left singular matrix, and each column vector may contain orthogonal vectors that capture patterns and correlations of the raw data. In this case, the U matrix may contain normalized orthogonal eigenvectors corresponding to the S matrix.
[0082] V T The matrix is the transpose of the right singular matrix and may include right singular vectors, which are orthogonal vectors describing the relationships of the raw data.
[0083] The processor (110) can produce a plurality of second decomposition matrices that, when summed, become the first decomposition matrix as the decomposition matrix of the first trajectory matrix based on a plurality of first decomposition matrices corresponding to the covariance matrix.
[0084] In other words, the processor (110) can decompose the first trajectory matrix into a plurality of second decomposition matrices using a plurality of first decomposition matrices.
[0085] Specifically, the processor (110) can decompose the first trajectory matrix into a plurality of second decomposition matrices based on the following [Equation 3].
[0086] [Mathematical Formula 3]
[0087]
[0088] Here, is the i-th second decomposition matrix among multiple second decomposition matrices, is the i-th eigenvalue of the covariance matrix, is the i-th column vector of matrix U, can mean the i-th row vector of matrix V.
[0089] The processor (110) generates a second trajectory matrix based on the first trajectory matrix.
[0090] The processor (110) can generate a second trajectory matrix based on a plurality of second decomposition matrices as a decomposition matrix of the first trajectory matrix.
[0091] Specifically, the processor (110) can generate a second trajectory matrix based on some selected matrices among a plurality of second decomposition matrices.
[0092] At this time, the processor (110) can, for example, select some of the multiple second decomposition matrices in order of the magnitude of the heart signal as many times as a preset number.
[0093] Afterward, the processor (110) can filter some matrices so that the heart signal corresponding to some matrices satisfies a preset frequency band.
[0094] Preferably, the frequency band may include a range of 0.7 Hz or more and 3 Hz or less, which is the frequency of a typical heartbeat.
[0095] Through this, the processor (110) can improve the precision of the extracted heart signal by removing noise and outliers other than the heart frequency.
[0096] Afterward, the processor (110) can generate a second trajectory matrix based on the remaining matrix after filtering some of the matrices.
[0097] Specifically, the processor (110) can generate a second trajectory matrix by summing all the remaining matrices.
[0098] The processor (110) extracts the heart signal based on the second trajectory matrix.
[0099] For example, the processor (110) can extract the heart signal in the frequency domain by applying a Fast Fourier Transformation to the second trajectory matrix.
[0100] The processor (110) can reconstruct the extracted heart signal based on the following [Equation 4].
[0101] [Mathematical Formula 4]
[0102] )=
[0103] Here, represents the reconstructed k-th heart signal, k is the number of heart signals, and m and n may represent the number of rows or columns of the first trajectory matrix.
[0104] The processor (110) can classify the subject as an arrhythmia patient if the specific wavelength interval in the heart signal is greater than or equal to a preset value.
[0105] The processor (110) can classify a subject as an arrhythmia patient if the standard deviation of a specific wavelength interval in the heart signal is greater than or equal to a preset value.
[0106] In particular, the processor (110) can classify the subject as an arrhythmia patient if the R peak interval or the standard deviation of the R peak interval in the heart signal is greater than or equal to a preset value.
[0107] Memory (120) stores one or more instructions executed by the processor (110).
[0108] The memory (120) can store various data used by the processor (110). For example, the memory (120) may include input data or output data for software (e.g., a program executed by the processor (110) and / or instructions associated with the program).
[0109] FIG. 2 is an exemplary diagram illustrating the output of an arrhythmia detection device (100) according to one embodiment.
[0110] Referring to Figure 2, the top shows the log value of the eigenvalue of the covariance matrix, and the bottom shows the heart signal corresponding to the top.
[0111] As shown above, the processor (110) can calculate the eigenvalues of the corresponding raw data by performing singular value decomposition on the covariance matrix. At this time, for scaling, the processor (110) can plot the distribution of the eigenvalues by taking the logarithm of the eigenvalues.
[0112] As shown below, the processor (110) can extract a heart signal based on a second trajectory matrix. Here, the processor (110) can identify heart signal characteristics by decomposing frequency components using wavelets.
[0113] FIG. 3 is a graph illustrating the performance of an arrhythmia detection device (100) according to one embodiment.
[0114] Referring to FIG. 3, the error between an arrhythmia detection device (100) according to one embodiment and a method of detecting arrhythmia using a conventional electrocardiogram signal is compared.
[0115] With respect to heart rate, the arrhythmia detection device (100) according to one embodiment shows a match rate of about 90% when compared to a method of detecting arrhythmia using a conventional electrocardiogram signal.
[0116] In relation to ICC, an arrhythmia detection device (100) according to one embodiment indicates that there is a correlation of less than 0.5 based on IBI (inter-beat interval) when compared to a method of detecting arrhythmia using a conventional electrocardiogram signal.
[0117] In particular, an arrhythmia detection device (100) according to one embodiment can classify normal people and arrhythmia patients based on characteristic values for IBI, similar to a method of detecting arrhythmia using existing electrocardiogram signals.
[0118] Normal subjects classified according to the method of detecting arrhythmia using conventional electrocardiograms have an IBI standard value of 105 ms or less, while arrhythmia patients have a value exceeding 105 ms.
[0119] Similarly, a normal subject classified according to an arrhythmia detection device (100) according to one embodiment has an IBI standard value of 200ms or less, and an arrhythmia patient has an IBI value exceeding 200ms.
[0120] Thus, an arrhythmia detection device (100) according to one embodiment can classify whether a subject is an arrhythmia patient based on whether the standard deviation of a specific wavelength interval exceeds a specific threshold value (e.g., 200 ms).
[0121] FIG. 4 is a flowchart illustrating a method for detecting arrhythmia according to one embodiment.
[0122] Referring to FIG. 4, the method of FIG. 4 can be performed by an arrhythmia detection device (100) according to one embodiment of FIG. 1.
[0123] First, an arrhythmia detection device (100) according to one embodiment acquires (410) raw data of a constant period ultra-wideband radar signal reflected from a subject.
[0124] Subsequently, an arrhythmia detection device (100) according to one embodiment generates a first trajectory matrix (420) based on time series data converted from raw data.
[0125] Subsequently, an arrhythmia detection device (100) according to one embodiment generates a second trajectory matrix (430) based on a first trajectory matrix.
[0126] Subsequently, an arrhythmia detection device (100) according to one embodiment extracts a heart signal (440) based on a second trajectory matrix.
[0127] Subsequently, an arrhythmia detection device (100) according to one embodiment determines (450) whether a specific wavelength interval in a heart signal is greater than or equal to a preset value.
[0128] At this time, the arrhythmia detection device (100) according to one embodiment classifies the subject as an arrhythmia patient (451) when a specific wavelength interval in the heart signal is greater than or equal to a preset value.
[0129] Meanwhile, an arrhythmia detection device (100) according to one embodiment classifies a subject as an arrhythmia patient (452) when a specific wavelength interval in the heart signal is less than a preset value.
[0130] In the illustrated Figure 4, the above method is described by dividing it into multiple steps, but at least some of the steps may be performed in a different order, combined with other steps, omitted, divided into detailed steps, or performed with one or more steps not illustrated added.
[0131] Embodiments of the present invention may include a program for performing the methods described herein on a computer, and a computer-readable recording medium comprising said program. The computer-readable recording medium may include program instructions, local data files, local data structures, etc., either alone or in combination. The medium may be one specifically designed and configured for the present invention, or one that is commonly available in the field of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of said programs may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.
[0132] Although representative embodiments of the present invention have been described in detail above, those skilled in the art will understand that various modifications can be made to the above-described embodiments without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims set forth below as well as equivalents thereof.
[0133] The terms described below are defined with consideration of their functions in the present invention, and these may vary depending on the operator's intent or practice. Therefore, their definitions should be based on the content throughout this specification. Terms used in the detailed description are intended merely to describe specific embodiments and should not be limiting. Unless explicitly stated otherwise, expressions in the singular form include the meaning of the plural form. In this description, expressions such as "include" or "compose" are intended to refer to certain components, numbers, steps, actions, elements, parts thereof, or combinations thereof, and should not be interpreted to exclude the existence or possibility of one or more other components, numbers, steps, actions, elements, parts thereof, or combinations thereof other than those described.
[0134] Additionally, the embodiments described herein may have aspects that are entirely hardware, partially hardware and partially software, or entirely software. In this specification, "unit," etc., refers to computer-related entities such as hardware, a combination of hardware and software, or software. Explanation of the symbols
[0135] 100: Arrhythmia detection device 110: Processor 120: Memory
Claims
Claim 1 An arrhythmia detection device comprising: one or more processors; and a memory storing one or more programs executed by the one or more processors, wherein the one or more processors: acquire raw data of a constant period ultra-wideband radar signal reflected from a subject, generate a first trajectory matrix based on time series data converted from the raw data, decompose the first trajectory matrix to generate a second trajectory matrix, convert the second trajectory matrix to extract a heart signal, and classify the subject as an arrhythmia patient when a specific wavelength interval in the heart signal is greater than or equal to a preset value. Claim 2 In claim 1, the one or more processors: an arrhythmia detection device that identifies the chest wall position of the subject based on the maximum dynamic variation included in the raw data. Claim 3 An arrhythmia detection device according to claim 1, wherein one or more processors: generate the first trajectory matrix by arranging the time series data as components of the first trajectory matrix based on the following [Equation 1] such that the components of the remaining diagonals perpendicular to the main diagonal of the first trajectory matrix are all identical. [Equation 1] (Here, H is the first locus matrix, x a is the a-th time series data, L is the length of the time series data, n and m are arbitrary natural numbers, provided that n = L - m + 1 and 2 ≤ m ≤ n) Claim 4 In claim 1, the one or more processors: an arrhythmia detection device that calculates the covariance matrix of the first trajectory matrix based on the following [Equation 2]. [Equation 2] (Here, C is the covariance matrix, H is the first locus matrix, H T is the transpose of the first trajectory matrix, and n is the number of time series data. Claim 5 An arrhythmia detection device according to claim 1, wherein one or more processors: decompose the first trajectory matrix into a plurality of second decomposition matrices based on a plurality of first decomposition matrices corresponding to the covariance matrix, which are calculated by performing Singular Value Decomposition (SVD) on the covariance matrix of the first trajectory matrix. Claim 6 In paragraph 5, the above one or more processors: S matrix U matrix and V included in the plurality of first decomposition matrices T An arrhythmia detection device that calculates the plurality of second decomposition matrices using a matrix based on the following [Equation 3]. [Equation 3] (Here, is the i-th second decomposition matrix among the plurality of second decomposition matrices above, is the i-th eigenvalue of the above covariance matrix, is the i-th column vector of the above U matrix, is the i-th row vector of matrix V) Claim 7 In claim 5, the one or more processors: select some of the plurality of second decomposition matrices in a predetermined number according to the magnitude of the heart signal, filter matrices other than the frequency band so that the heart signal corresponding to the some matrices satisfies a predetermined frequency band, and generate the second trajectory matrix based on the remaining matrix after filtering among the some matrices, an arrhythmia detection device. Claim 8 In claim 1, the one or more processors: an arrhythmia detection device that reconstructs a heart signal extracted from the second trajectory matrix based on the following [Equation 4]. [Equation 4] )= (Here, is the reconstructed k-th cardiac signal, k is the number of cardiac signals, and m and n are the number of rows or columns of the above trajectory matrix) Claim 9 In claim 1, the one or more processors are: an arrhythmia detection device that classifies the subject as an arrhythmia patient if the R peak interval in the heart signal is greater than or equal to a preset value. Claim 10 A method performed by an arrhythmia detection device comprising one or more processors; and a memory storing one or more programs executed by said one or more processors, the method comprising: acquiring raw data of a constant period ultra-wideband radar signal reflected from a subject; generating a first trajectory matrix based on time series data converted from said raw data; decomposing the first trajectory matrix to generate a second trajectory matrix; converting the second trajectory matrix to extract a heart signal; and classifying said subject as an arrhythmia patient if a specific wavelength interval in said heart signal is greater than or equal to a preset value. Claim 11 In claim 10, the above method further comprises the step of identifying the chest wall position of the subject based on the maximum dynamic variation included in the raw data, a method for detecting arrhythmia. Claim 12 A method for detecting arrhythmia according to claim 10, wherein the step of generating the first trajectory matrix comprises the step of generating the first trajectory matrix by arranging the time series data as components of the first trajectory matrix based on the following [Equation 1] such that the components of the remaining diagonals perpendicular to the main diagonal of the first trajectory matrix are all identical. [Equation 1] (Here, H is the first locus matrix, is the a-th time series data, L is the length of the time series data, n and m are arbitrary natural numbers, provided that n = L - m + 1 and 2 ≤ m ≤ n) Claim 13 A method for detecting arrhythmia according to claim 10, wherein the step of generating the second trajectory matrix comprises the step of calculating the covariance matrix of the first trajectory matrix based on the following [Equation 2]. [Equation 2] (Here, C is the covariance matrix, H is the first locus matrix, H T is the transpose of the first trajectory matrix, and n is the number of time series data. Claim 14 A method for detecting arrhythmia according to claim 10, wherein the step of generating the second trajectory matrix comprises the step of decomposing the first trajectory matrix into a plurality of second decomposition matrices based on a plurality of first decomposition matrices corresponding to the covariance matrix, which are obtained by performing Singular Value Decomposition (SVD) on the covariance matrix of the first trajectory matrix. Claim 15 In claim 14, the step of generating the second trajectory matrix comprises the S matrix U matrix and V included in the plurality of first decomposition matrices. T A method for detecting arrhythmia comprising the step of calculating the plurality of second decomposition matrices using a matrix based on the following [Equation 3]. [Equation 3] (Here, is the i-th second decomposition matrix among the plurality of second decomposition matrices above, is the i-th eigenvalue of the above covariance matrix, is the i-th column vector of the above U matrix, is the i-th row vector of the above V matrix) Claim 16 A method for detecting arrhythmia according to claim 14, wherein the step of generating the second trajectory matrix comprises: selecting a predetermined number of some matrices among the plurality of second decomposition matrices in order of the magnitude of the heart signal; filtering matrices other than the frequency band so that the heart signal corresponding to the some matrices satisfies a predetermined frequency band; and generating the second trajectory matrix based on the remaining matrices after filtering among the some matrices. Claim 17 A method for detecting arrhythmia according to claim 10, wherein the step of extracting the heart signal further comprises the step of reconstructing the heart signal extracted from the second trajectory matrix based on the following [Equation 4]. [Equation 4] )= (Here, is the reconstructed k-th cardiac signal, k is the number of cardiac signals, and m and n are the number of rows or columns of the above trajectory matrix) Claim 18 A method for detecting arrhythmia according to claim 10, wherein the classifying step includes classifying the subject as an arrhythmia patient if the R peak interval in the heart signal is greater than or equal to a preset value.
Citation Information
Patent Citations
Vital check method using Ultra-wideband radar
KR101895324B1
