Method, device and electronic equipment for detecting physiological signal quality

Abstract

The present invention relates to the field of biomedical signal processing technology, and specifically to a method, device and electronic device for detecting the quality of physiological signals, wherein the method comprises: obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal; aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle; and detecting the signal quality of the waveform of the current cycle according to the correlation coefficient corresponding to the waveform of the current cycle. By calculating the correlation coefficient of two adjacent waveforms, that is, calculating the correlation coefficient by wave-by-wave matching, the problem of low detection efficiency of physiological signal quality caused by the method of establishing a template is avoided; the target physiological signal itself is used to detect the signal quality without relying on the template, and the signal quality detection can be realized while the target physiological signal is obtained, which has a high detection efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of biomedical signal processing, and in particular to a method, a device and an electronic device for detecting the quality of a physiological signal. Background Art

[0002] Biomedical signals are low-frequency weak signals in a strong noise background. They are unstable natural signals emitted by complex living organisms. In each cardiac cycle, the contraction and relaxation of the ventricles cause periodic changes in the pressure in the arteries, and the expansion and retraction signals are called pulsation signals. Based on the principle that the amount of light absorbed by arterial blood changes with arterial pulsation, a blood oxygen pulse signal, namely a PPG (Photo PlethysmoGraphy) signal, can be obtained.

[0003] The most commonly used method for evaluating the signal quality of PPG is the correlation method. The usual steps are to establish a template, then calculate the Pearson correlation coefficient with the signal to be matched, and update or rebuild the template during the continuous matching process. The particularity of the PPG signal is that its amplitude has no strong correlation with physiology. The reasons are: First, it is controlled by the programmable gain adjustment algorithm and is a processed signal. Second, it is interfered by the patient himself, such as body movement, breathing, and coughing. Third, it is interfered by external factors such as changes in the probe holding force, skin pigmentation, nail polish, environmental gases or fluorescent lights (especially those oscillation frequencies close to the harmonic frequency of the pulse oximeter LED pulsation). Therefore, the amplitude of the PPG signal is always random, and the waveform may be affected by arbitrary baseline offsets and sudden amplitude changes. This brings great trouble to the establishment of the PPG signal template: first, it is difficult to establish a reasonable template; second, even if a template is established, the frequency of template reconstruction will be high. Therefore, the efficiency of physiological signal quality detection by establishing a template is low. Summary of the invention

[0004] In view of this, the embodiments of the present invention provide a method, an apparatus and an electronic device for detecting the quality of a physiological signal, so as to solve the problem of low efficiency in detecting the quality of a physiological signal.

[0005] According to a first aspect, an embodiment of the present invention provides a method for detecting quality of a physiological signal, comprising:

[0006] Obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal;

[0007] Aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle;

[0008] The signal quality of the waveform of the current period is detected according to the correlation coefficient corresponding to the waveform of the current period.

[0009] The method for detecting the quality of physiological signals provided in an embodiment of the present invention aligns the waveform of the current cycle with the waveform of the previous cycle and calculates the correlation coefficient of the two adjacent waveforms, that is, adopts a wave-by-wave matching method to calculate the correlation coefficient, thereby avoiding the problem of low detection efficiency of the quality of physiological signals caused by adopting a template-building method; the method adopts the target physiological signal itself without relying on the template to detect the signal quality, and can realize the signal quality detection while acquiring the target physiological signal, and has a high detection efficiency.

[0010] In combination with the first aspect, in a first implementation of the first aspect, aligning the waveform of the current cycle with the waveform of the previous cycle to calculate the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle includes:

[0011] Constructing a matrix grid using the number of sample points of the waveform of the current cycle and the number of sample points of the waveform of the previous cycle; wherein each grid point of the matrix grid represents the distance between two sample points when the waveform of the current cycle is aligned with two sample points in the waveform of the previous cycle;

[0012] The matrix grid is searched using an N*N sliding window to determine the alignment grid points of the waveform of the current cycle and the waveform of the previous cycle in the matrix grid; wherein each grid point of the sliding window represents a cumulative distance, and the middle grid point of the sliding window corresponds to the current search grid point in the matrix grid; and N represents the number of grid points in each row or column of the sliding window;

[0013] Based on the distance represented by the searched alignment grid points, a correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle is calculated.

[0014] The method for detecting the quality of physiological signals provided by the embodiment of the present invention searches the matrix grid by means of a sliding window to perform a local search; the local search method has low spatial complexity and time complexity, occupies little space resources, consumes little time resources, and improves the efficiency of physiological signal quality detection.

[0015] In combination with the first embodiment of the first aspect, in the second embodiment of the first aspect, searching the matrix grid using an N*N sliding window to determine the alignment grid points of the waveform of the current cycle and the waveform of the previous cycle in the matrix grid includes:

[0016] In three preset search directions, respectively determining the next search grid point of the current search grid point in the matrix grid;

[0017] Calculate the cumulative distances between the current search grid point and the corresponding next search grid point in three preset search directions respectively;

[0018] Determine a preset search direction corresponding to the smallest cumulative distance among the three calculated cumulative distances as the target search direction;

[0019] The sliding window is moved along the target search direction, and each grid point of the sliding window is updated to perform the next search on the matrix grid; wherein the grid points of the matrix grid corresponding to all the intermediate grid points are the alignment grid points.

[0020] In combination with the first aspect, or the first implementation manner of the first aspect, or the second implementation manner of the first aspect, in a third implementation manner of the first aspect, detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period includes:

[0021] Extracting the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle in the target physiological signal waveform;

[0022] Calculating a target correlation coefficient corresponding to the waveform of the current cycle using all the extracted correlation coefficients;

[0023] The target correlation coefficient is compared with a preset correlation coefficient to detect the signal quality of the waveform of the current cycle; wherein the preset correlation coefficient corresponds to the signal quality one by one.

[0024] The method for detecting the quality of physiological signals provided by an embodiment of the present invention, when detecting the correlation coefficient corresponding to the waveform of the current cycle, not only considers the correlation coefficient corresponding to the waveform of the current cycle, but also combines the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle, thereby improving the accuracy of the signal quality of the detected waveform of the current cycle.

[0025] In combination with the third embodiment of the first aspect, in the fourth embodiment of the first aspect, the extracting of the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle in the target physiological signal waveform include:

[0026] Obtaining an array of preset length; wherein the elements in the array are used to represent the correlation coefficients corresponding to the waveforms of each period;

[0027] Circularly moving the array, and inserting the correlation coefficient corresponding to the waveform of the current cycle into the array;

[0028] All elements in the array are extracted to obtain the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficient corresponding to the waveforms of all cycles before the waveform of the current cycle in the target physiological signal waveform.

[0029] In combination with the third implementation of the first aspect, in the fifth implementation of the first aspect, calculating the target correlation coefficient corresponding to the waveform of the current cycle by using all the extracted correlation coefficients includes:

[0030] Calculating the mean of all the correlation coefficients to obtain a first target correlation coefficient;

[0031] Calculating the median of all the correlation coefficients to obtain a second target correlation coefficient;

[0032] A weighted sum is performed on the first target correlation coefficient and the second target correlation coefficient to obtain the target correlation coefficient.

[0033] In combination with the first aspect, in a sixth implementation of the first aspect, the method further includes:

[0034] The waveform of the current period is marked with an identifier corresponding to the signal quality of the waveform of the current period.

[0035] The method for detecting the quality of physiological signals provided in an embodiment of the present invention uses an identifier corresponding to the detected signal quality to mark the waveform of the current cycle, so that the user can intuitively know the signal quality of the waveform of the current cycle, that is, the waveform of the marked target physiological signal can convey signal quality information to achieve intelligent display of the target physiological signal.

[0036] In combination with the first aspect, or the sixth embodiment of the first aspect, in the seventh embodiment of the first aspect, before the step of acquiring the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal, the step further includes:

[0037] Acquiring the target physiological signal waveform;

[0038] The waveforms of the target physiological signal waveforms that have not been subjected to signal quality detection are marked.

[0039] The method for detecting physiological signal quality provided by the embodiment of the present invention marks the waveform that has not been subjected to signal quality detection, so as to reduce the delay between the extraction and real-time sampling point of the waveform of the current cycle and the waveform of the previous cycle.

[0040] According to a second aspect, an embodiment of the present invention further provides a device for detecting quality of a physiological signal, comprising:

[0041] An acquisition module, used to acquire the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal;

[0042] A correlation coefficient calculation module, used for aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle, so as to obtain the correlation coefficient corresponding to the waveform of the current cycle;

[0043] The signal quality detection module is used to detect the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period.

[0044] The device for detecting the quality of physiological signals provided in an embodiment of the present invention aligns the waveform of the current cycle with the waveform of the previous cycle and calculates the correlation coefficient of the two adjacent waveforms, that is, adopts a wave-by-wave matching method to calculate the correlation coefficient, thereby avoiding the problem of low detection efficiency of the quality of physiological signals caused by adopting a template-building method; the device uses the target physiological signal itself instead of relying on the template to detect the signal quality, and can realize the signal quality detection while acquiring the target physiological signal, and has a high detection efficiency.

[0045] According to the third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for detecting physiological signal quality described in the first aspect or any one embodiment of the first aspect by executing the computer instructions.

[0046] According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method for detecting physiological signal quality described in the first aspect or any one embodiment of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 is a flow chart of a method for detecting quality of a physiological signal according to an embodiment of the present invention;

[0049] Figure 2 is a waveform diagram of target physiological signal quality according to an embodiment of the present invention;

[0050] Figure 3is a flow chart of a method for detecting quality of a physiological signal according to an embodiment of the present invention;

[0051] Figure 4a-4d This is a schematic diagram of a 3*3 sliding window and its moving direction according to an embodiment of the present invention;

[0052] Figure 5a-5c A schematic diagram of the movement of a 3*3 sliding window according to an embodiment of the present invention;

[0053] Figure 6 This is a comparison diagram of the effects of DTW and E-DTW in an embodiment of the present invention;

[0054] Figure 7 This is a comparison diagram of the effects of DTW and E-DTW in an embodiment of the present invention;

[0055] Figure 8 is a flow chart of a method for detecting quality of a physiological signal according to an embodiment of the present invention;

[0056] Fig. 9 is a signal quality judgment effect diagram based on E-DTW according to an embodiment of the present invention;

[0057] Fig.10 is a flow chart of a method for detecting quality of a physiological signal according to an embodiment of the present invention;

[0058] Fig.11 is a structural block diagram of a device for detecting physiological signal quality according to an embodiment of the present invention;

[0059] Fig.12 It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0061] It should be noted that the periodic signal described in the embodiments of the present invention does not refer to a physiological signal within a fixed time, but refers to a periodic signal caused by the periodic mechanical activity of the target part of the living body. For example, the cardiac cycle refers to the mechanical activity cycle formed by each contraction and relaxation of the heart. The waveform corresponding to each cardiac cycle is the waveform between adjacent end-systoles or adjacent end-diastole periods. The time corresponding to the waveforms of two adjacent cardiac cycles is not necessarily the same. The heartbeat of the living body may accelerate in a certain time period, so the time corresponding to the waveforms of two adjacent cardiac cycles is different.

[0062] The physiological signal waveform is a physiological signal of a living body collected by a monitoring device. Specifically, each point in the physiological signal waveform is a sample point obtained by the monitoring device collecting physiological signals from the living body. As mentioned above, since the time corresponding to the physiological signal waveform of two adjacent cycles is different, the number of sample points of the physiological signal waveform of the two cycles is also different.

[0063] According to an embodiment of the present invention, a method embodiment for detecting the quality of a physiological signal is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0064] In this embodiment, a method for detecting the quality of a physiological signal is provided, which can be used in an electronic device, such as a monitor. Figure 1 is a flow chart of a method for detecting physiological signal quality according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0065] S11, obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal.

[0066] When the electronic device obtains the target physiological signal waveform, it can use the extreme point search algorithm to search for the extreme point in the target physiological signal waveform, and then obtain the waveform of the current cycle. For the convenience of subsequent description, the waveform of the current cycle is called B. The electronic device defines the previous waveform of the current cycle waveform as a reference template that matches the correlation with the waveform of the current cycle, that is, the waveform of the previous cycle, which can be called T. The correlation coefficient is subsequently calculated using the waveform of the current cycle and the waveform of the previous cycle.

[0067] by Figure 2 For example, the waveform of the current cycle obtained by the electronic device is Figure 2 The waveform corresponding to 12 in the previous cycle is Figure 2 The waveform corresponding to 11.

[0068] Optionally, the waveform of the current cycle can also be obtained in other ways, for example, a pre-trained network model can be used, the input of the network model is the target physiological signal waveform, and the output is a waveform of one cycle in the target physiological signal waveform; or other methods can be used. There is no restriction on the method of obtaining the waveform of the current cycle and the waveform of the previous cycle in the target physiological signal waveform, as long as the electronic device can obtain the waveform of the current cycle and the waveform of the previous cycle.

[0069] S12, aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle.

[0070] Before aligning the waveform of the current cycle with the waveform of the previous cycle, the waveform can be converted into a time series. As mentioned above, the waveform is composed of multiple sample points, so the waveform can be expressed using a sample point sequence. That is, the waveform of the current cycle and the waveform of the previous cycle can be expressed using the following time series:

[0071] B={b1,b2,b3,…,b n};

[0072] T={t1,t2,t3,…,t m};

[0073] Among them, b1, b2, b3, …, b n represents the sample points in the waveform of the current cycle, n is the number of sample points in the waveform of the current cycle; t1, t2, t3, ..., t m Represents the sample points in the waveform of the previous cycle, and m is the number of sample points in the waveform of the previous cycle.

[0074] After the electronic device obtains the waveform B of the current cycle and the waveform T of the previous cycle in S11, it can align the waveform of the current cycle with the waveform of the previous cycle by using DTW (Dynamic Time Warping) and calculate the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle.

[0075] Specifically, in order to align the two sequences, we need to construct an n×m matrix grid, where each grid point (i, j) represents b i and t j The distance between two points d(b i , t j ). Each grid point (i, j) represents point b i and t jIt should be noted that the distance between two points can be expressed by Euclidean distance, cosine distance, or other forms of distance, and there is no restriction on the distance representation form.

[0076] The path of each grid point in the matrix grid has only three directions of advancement: if the path has passed through grid point (i, j), then the next grid point passed can only be one of the following three situations: (i+1, j), (i, j+1) or (i+1, j+1). The cumulative Euclidean distance γ(i, j) of each grid point is defined as the distance d(b) found in the current grid point. i , t j ) and the minimum value of the cumulative Euclidean distance between adjacent grid points, that is:

[0077] γ(i,j)=d(b i ,t j )+min{γ(i-1,j),γ(i,j-1),γ(i-1,j-1)}

[0078] The DTW algorithm can be summarized as finding a path that passes through several grid points in this grid. The grid points that the path passes through are the alignment points for calculating the two sequences. After determining the alignment points of the two sequences, the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle can be calculated using the Pearson correlation coefficient calculation method.

[0079] Optionally, in addition to using DTW to align the two sequences, the electronic device may also use other methods. Since DTW uses a global search method to align the two sequences, the electronic device may also use a local search method to align the two sequences. The local search method will be described in detail below.

[0080] S13, detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period.

[0081] After the electronic device calculates the correlation coefficient corresponding to the waveform of the current cycle, it can use the correlation coefficient to compare with the preset correlation coefficient. When the correlation coefficient is greater than or equal to the preset correlation coefficient, it indicates that the signal quality of the waveform of the current cycle is good; when the correlation coefficient is less than the preset correlation coefficient, it indicates that the signal quality of the waveform of the current cycle is poor. Alternatively, the signal quality can be subdivided, that is, the signal quality is divided into multiple levels, each level corresponds to a preset correlation coefficient, and the correlation coefficient calculated in S12 is compared with the preset correlation coefficient to determine the level corresponding to the signal quality of the waveform of the current cycle. Of course, other methods can also be used to detect the signal quality of the waveform of the current cycle.

[0082] Among them, the preset correlation coefficient can be specifically set according to actual conditions, for example, it can be set using empirical values, or it can be determined by analyzing the correlation coefficients of waveforms of various signal qualities, etc. There is no restriction on the specific method of determining the preset correlation coefficient, it only needs to ensure that the electronic device can obtain the preset correlation coefficient.

[0083] The method for detecting the quality of physiological signals provided in this embodiment aligns the waveform of the current cycle with the waveform of the previous cycle and calculates the correlation coefficient of the two adjacent waveforms, that is, adopts a wave-by-wave matching method to calculate the correlation coefficient, thereby avoiding the problem of low detection efficiency of the quality of physiological signals caused by adopting a template-building method; this method uses the target physiological signal itself to detect the signal quality without relying on the template, and can realize the signal quality detection while acquiring the target physiological signal, and has a high detection efficiency.

[0084] In this embodiment, a method for detecting the quality of a physiological signal is provided, which can be used in an electronic device, such as a monitor. Figure 3 is a flow chart of a method for detecting physiological signal quality according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0085] S21, obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal.

[0086] For details, please see Figure 1 S11 of the illustrated embodiment will not be described in detail here.

[0087] S22, aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle.

[0088] In this embodiment, the electronic device uses a local search method to align the waveform of the current cycle with the waveform of the previous cycle. Specifically, the above S22 may include the following steps:

[0089] S221, constructing a matrix grid using the number of sample points of the waveform of the current cycle and the number of sample points of the waveform of the previous cycle.

[0090] Each grid point of the matrix grid represents the distance between two sample points when the waveform of the current cycle is aligned with two sample points in the waveform of the previous cycle.

[0091] See also Figure 1The description of S12 in the illustrated embodiment about converting the waveform B of the current cycle and the waveform of the previous cycle into the form of a time series. The waveform of the current cycle and the waveform of the previous cycle can be represented by the following time series:

[0092] B={b1,b2,b3,…,b n};

[0093] T={t1,t2,t3,…,t m}.

[0094] The waveform B of the current cycle has n sample points, and the waveform T of the previous cycle has m sample points. Based on the number of sample points of the two waveforms, an n*m matrix grid is constructed. Each grid point (i, j) in the matrix grid represents b i and t j The distance between two points d(b i , t j ).

[0095] S222, searching the matrix grid using an N*N sliding window to determine the alignment grid points of the waveform of the current cycle and the waveform of the previous cycle in the matrix grid.

[0096] Each grid point of the sliding window represents a cumulative distance, and the middle grid point of the sliding window corresponds to the current search grid point in the matrix grid.

[0097] The sliding window is N*N, where N is the number of grid points in each row or column of the sliding window. The size of the sliding window can be set according to the actual situation, for example, it can be 3*3, 4*4, or 5*5, etc. In the following description, a 3*3 sliding window is used as an example, and the distance between sample points is the Euclidean distance. The use of a 3*3 sliding window can occupy less computing resources and has a better computing effect.

[0098] A 3*3 sliding window is used to search the matrix grid starting from the grid point (1, 1) of the matrix grid, that is, the middle grid point of the sliding window is aligned with the grid point (1, 1) of the matrix grid. The search direction of the sliding window is determined by calculating the cumulative distance to determine the alignment grid points of the waveform of the current cycle and the waveform of the previous cycle in the matrix grid.

[0099] Specifically, the above S222 includes the following steps:

[0100] (1) In the three preset search directions, the next search grid point of the current search grid point is determined in the matrix grid respectively.

[0101] Figure 4a-4dA 3*3 sliding window and its moving direction are shown. The sliding window has three preset search directions. If the current search grid point is (i, j), the next search grid point is one of (i+1, j), (i, j+1) or (i+1, j+1).

[0102] like Figure 4a As shown, there are 9 grid points (1) to (9), each of which stores the cumulative Euclidean distance. The middle grid point (5) of the sliding window corresponds to the current search grid point in the matrix grid, and the shortest cumulative Euclidean distance is selected in the three directions (6), (8), and (9).

[0103] (2) Calculate the cumulative distance between the current search grid point and the corresponding next search grid point in the three preset search directions respectively.

[0104] According to the principle of DTW algorithm, Figure 4b As shown, the cumulative Euclidean distance of (6) is the sum of the Euclidean distance of (6) and the minimum of the cumulative Euclidean distances in (2), (3), and (5); Figure 4c As shown, the cumulative Euclidean distance of (8) is the sum of the Euclidean distance of (8) and the minimum of the cumulative Euclidean distances in (4), (5), and (7); Figure 4d As shown, the cumulative Euclidean distance of (9) is the sum of the Euclidean distance of (9) and the minimum of the cumulative Euclidean distances in (5), (6), and (8).

[0105] (3) Determine the preset search direction corresponding to the smallest cumulative distance among the three calculated cumulative distances as the target search direction.

[0106] After calculating the cumulative Euclidean distances of (6), (8), and (9), the cumulative Euclidean distances of the three are compared, and the preset search direction with the smallest cumulative Euclidean distance among the three is taken as the target search direction.

[0107] (4) Move the sliding window along the target search direction and update each grid point of the sliding window to perform the next search on the matrix grid.

[0108] Among them, the grid points of the matrix grid corresponding to all the intermediate grid points are the alignment grid points.

[0109] If the cumulative Euclidean distance of (9) is the smallest, the sliding window moves diagonally downward, see Figure 5a At the same time, the sliding window is updated, that is, (5) becomes (1), (6) becomes (2), (8) becomes (4), (9) becomes (5), and (3), (6), (7), (8), (9) recalculate the cumulative Euclidean distance. If the cumulative Euclidean distance of (6) is the smallest, the sliding window moves to the right, see Figure 5bAt the same time, the sliding window is updated, that is, (2) becomes (1), (3) becomes (2), (5) becomes (4), (6) becomes (5), (8) becomes (7), (9) becomes (8), and (3), (6), and (9) recalculate the cumulative Euclidean distance. If the cumulative Euclidean distance of (8) is the smallest, the sliding window moves downward, see Figure 5c At the same time, the sliding window is updated, that is, (4) becomes (1), (5) becomes (2), (6) becomes (3), (7) becomes (4), (8) becomes (5), (9) becomes (6), and (7), (8), and (9) recalculate the cumulative Euclidean distance.

[0110] When one of the sliding window points (6), (8), (9) reaches (n, m), the sliding stops. The (i, j) corresponding to all the sliding grid points (5) of the sliding window is the point where the two sequences should be aligned.

[0111] S223, calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle based on the distance represented by the searched alignment grid points.

[0112] After aligning the two signals, the electronic device can calculate the correlation coefficient ρ. In essence, the local search method of this embodiment can be called the E-DTW method, which replaces the global search of DTW with a local search.

[0113] The waveform B of the current cycle and the waveform T of the next cycle are aligned using DTW and E-DTW respectively, and the correlation coefficient is calculated.

[0114] From the perspective of space complexity, in order to align B and T, we first need to construct an n×m matrix grid. If the sampling rate is 250Hz and the pulse rate is 60bpm, then the width of B and T is 250 sampling points. When calculating DTW, we need to build a 250×250 32-bit matrix, that is, 4×250×250=250000Bytes≈244KBytes, which has too high space complexity. If used on embedded platforms, the cost will be high due to the large amount of RAM occupied. In addition, according to the DTW method, to find the shortest cumulative Euclidean distance path in the constructed matrix, at least 250×250=62500 cumulative Euclidean distance calculations are required. If used on embedded platforms, the time complexity is too high and it will consume too many time resources.

[0115] If the E-DTW method in this embodiment is used and a 3*3 sliding window is used to store the accumulated Euclidean distance, the space complexity of this method is 4×3×3=36Bytes. For a signal with a sampling rate of 250Hz and a pulse rate of 60bpm, the space complexity is reduced by about 6900 times; only about 1244 distance and calculations are required, and the time complexity is reduced by about 50 times. It can be seen that the E-DTW method described in this embodiment has the advantages of occupying less space resources and consuming less time resources. If used in an embedded platform, it has the advantage of low cost.

[0116] From the calculated correlation coefficient, Figure 6 and Figure 7 This is a comparison of the effects of DTW and E-DTW. Figure 6 as well as Figure 7 The original signal in is the target physiological signal that has not been aligned, the waveform under DTW is the waveform after the target physiological signal is processed by DTW, and the waveform under E-DTW is the waveform after the target physiological signal is processed by E-DTW. The E-DTW described in this embodiment can achieve the same calculation results as DTW while reducing the spatial complexity and time complexity.

[0117] S23, detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period.

[0118] For details, please see Figure 1 S13 of the illustrated embodiment will not be described in detail here.

[0119] The method for detecting the quality of physiological signals provided in this embodiment searches the matrix grid by means of a sliding window to perform a local search; the local search method has low spatial complexity and time complexity, occupies little space resources, consumes little time resources, and improves the efficiency of physiological signal quality detection.

[0120] In this embodiment, a method for detecting the quality of a physiological signal is provided, which can be used in an electronic device, such as a monitor. Figure 8 is a flow chart of a method for detecting physiological signal quality according to an embodiment of the present invention. Figure 8 As shown, the process includes the following steps:

[0121] S31, obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal.

[0122] For details, please see Figure 3 S21 of the illustrated embodiment will not be described in detail here.

[0123] S32, aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle.

[0124] For details, please see Figure 3 S22 of the illustrated embodiment will not be described in detail here.

[0125] S33, detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period.

[0126] In this embodiment, when detecting the signal quality of the waveform of the current cycle, not only the correlation coefficient calculated in S32 is considered, but also the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle are combined. Specifically, the above S33 includes the following steps:

[0127] S331, extracting the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle in the waveform of the target physiological signal.

[0128] Since the target physiological signal waveform is a continuous waveform, the electronic device has calculated the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle and stored them in the electronic device before calculating the correlation coefficient corresponding to the waveform of the current cycle. That is, the electronic device can directly extract the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficients corresponding to the waveforms of all previous cycles from the corresponding storage space.

[0129] As an optional implementation of this embodiment, the correlation coefficients may be stored in an array. Specifically, the above S331 includes the following steps:

[0130] (1) Obtaining an array of preset length; wherein the elements in the array are used to represent the correlation coefficients corresponding to the waveforms of each period.

[0131] The preset length is the number of waveforms that need to be calculated for the correlation coefficient, which is set in advance. The preset length can be specifically set according to actual conditions and is not limited here.

[0132] For example, the preset length is set to L. After L correlation coefficients are calculated, the correlation coefficient corresponding to the waveform of the current period can be used to replace the earliest stored correlation coefficient to achieve the circular use of the array and avoid using a longer array to occupy more computing resources.

[0133] (2) Circularly move the array and insert the correlation coefficient corresponding to the waveform of the current cycle into the array.

[0134] After calculating the correlation coefficient corresponding to the waveform of the current cycle, the array can be moved cyclically to the first position and the correlation coefficient can be inserted into the last position of the array; or the array can be moved cyclically to the last position and the correlation coefficient can be inserted into the first position of the array.

[0135] (3) Extracting all elements in the array to obtain the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficient corresponding to the waveform of all cycles before the waveform of the current cycle in the target physiological signal waveform.

[0136] S332, using all the extracted correlation coefficients, calculate the target correlation coefficient corresponding to the waveform of the current cycle.

[0137] The electronic device may calculate the mean of all correlation coefficients as the target correlation coefficient corresponding to the waveform of the current period; may also calculate the median of all correlation coefficients as the target correlation coefficient; or may adopt other methods to achieve it.

[0138] As an optional implementation of this embodiment, the above S332 includes the following steps:

[0139] (1) Calculate the mean of all correlation coefficients to obtain a first target correlation coefficient;

[0140] (2) calculating the median of all correlation coefficients to obtain a second target correlation coefficient;

[0141] (3) Perform a weighted summation on the first target correlation coefficient and the second target correlation coefficient to obtain the target correlation coefficient.

[0142] For example, the target correlation coefficient can be expressed as follows:

[0143] ρ B =coef1×ρ mean +coef2×ρ median ;

[0144] Among them, ρ B is the target correlation coefficient; ρ mean is the first target correlation coefficient; coef1 is the weight corresponding to the first target correlation coefficient; ρ median is the second target correlation coefficient; coef2 is the weight corresponding to the second target correlation coefficient. coef1 and coef2 can be set according to actual conditions and are not limited here.

[0145] The target correlation coefficient ρ of waveform B is obtained according to the following formula B , the higher the value, the better the current signal quality, and vice versa, the worse the signal quality. Optionally, the electronic device calculates the target correlation coefficient ρ B Afterwards, if Fig. 9As shown, the target correlation coefficient can be superimposed and displayed on the waveform of the target physiological signal (ie, the original signal in the figure).

[0146] S333, comparing the target correlation coefficient with the preset correlation coefficient to detect the signal quality of the waveform of the current cycle.

[0147] The preset correlation coefficient corresponds to the signal quality one by one.

[0148] In this embodiment, the signal quality is divided into 4 levels, and the level division rules are as follows:

[0149]

[0150] Thd1, Thd2, and Thd3 are preset correlation coefficients corresponding to each level. For example, Thd1, Thd2, and Thd3 are 0.9, 0.79, and 0.5, respectively, and are not limited thereto in other examples. The smaller the level, the better the signal quality.

[0151] The method for detecting the quality of physiological signals provided by an embodiment of the present invention, when detecting the correlation coefficient corresponding to the waveform of the current cycle, not only considers the correlation coefficient corresponding to the waveform of the current cycle, but also combines the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle, thereby improving the accuracy of the signal quality of the detected waveform of the current cycle.

[0152] As an optional implementation of this embodiment, after the above S33, the following may be included: marking the waveform of the current period with an identifier corresponding to the signal quality of the waveform of the current period.

[0153] The mark corresponding to the signal quality can be a color, a line type, or other marks, as long as the mark can distinguish the waveforms of various signal qualities. In this embodiment, colors can be used to distinguish waveforms of different signal qualities, for example:

[0154] If B At level 1, the waveform of the current cycle is colored green.

[0155] If B At level 2, the waveform of the current period is colored blue.

[0156] If B At level 3, the waveform of the current cycle is colored yellow.

[0157] If B At level 4, the waveform of the current cycle is colored red.

[0158] The waveform of the current cycle is marked with an identifier corresponding to the detected signal quality, so that the user can intuitively know the signal quality of the waveform of the current cycle, that is, the waveform of the marked target physiological signal can convey signal quality information to achieve intelligent display of the target physiological signal.

[0159] In this embodiment, a method for detecting the quality of a physiological signal is provided, which can be used in an electronic device, such as a monitor. Fig.10 is a flow chart of a method for detecting physiological signal quality according to an embodiment of the present invention. Fig.10 As shown, the process includes the following steps:

[0160] S41, obtaining a target physiological signal waveform.

[0161] When the electronic device is a monitor, the monitor monitors the target physiological signal waveform of the living body in real time and performs the following processing.

[0162] S42, marking waveforms in the target physiological signal waveform that have not been subjected to signal quality detection.

[0163] When the electronic device obtains the target physiological signal waveform, it can mark the waveform that has not been subjected to signal quality detection. It can be distinguished by color or other methods, and there is no limitation here. For example, the electronic device colors this part of the waveform with white, and the processed waveform can be colored according to the color corresponding to the signal quality of each waveform.

[0164] S43, obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal.

[0165] For details, please see Figure 8 S31 of the illustrated embodiment will not be described in detail here.

[0166] S44, aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle.

[0167] For details, please see Figure 8 S32 of the illustrated embodiment will not be described in detail here.

[0168] S45, detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period.

[0169] For details, please see Figure 8 S33 of the illustrated embodiment will not be described in detail here.

[0170] S46, marking the waveform of the current cycle with an identifier corresponding to the detected signal quality.

[0171] For details, please see Figure 8 The optional implementations after S33 of the illustrated embodiment will not be described in detail here.

[0172] The method for detecting the quality of physiological signals provided in this embodiment marks the waveforms that have not been subjected to signal quality detection, so as to reduce the delay between the extraction and real-time sampling points of the waveforms of the current cycle and the waveforms of the previous cycle.

[0173] In this embodiment, a device for detecting the quality of physiological signals is also provided, and the device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made are not repeated here. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0174] This embodiment provides a device for detecting the quality of physiological signals, such as Fig.11 As shown, including:

[0175] The acquisition module 51 is used to acquire the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal.

[0176] The correlation coefficient calculation module 52 is used to align the waveform of the current cycle with the waveform of the previous cycle, and calculate the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle.

[0177] The signal quality detection module 53 is used to detect the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period.

[0178] The device for detecting the quality of physiological signals provided in this embodiment aligns the waveform of the current cycle with the waveform of the previous cycle and calculates the correlation coefficient of the two adjacent waveforms, that is, adopts a wave-by-wave matching method to calculate the correlation coefficient, thereby avoiding the problem of low detection efficiency of the quality of physiological signals caused by adopting a template-building method; the device uses the target physiological signal itself instead of relying on the template to detect the signal quality, and can realize the signal quality detection while acquiring the target physiological signal, and has a high detection efficiency.

[0179] The device for detecting the quality of physiological signals in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and a memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0180] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0181] The embodiment of the present invention also provides an electronic device having the above Fig.11 The device for detecting the quality of physiological signals is shown.

[0182] See also Fig.12 , Fig.12 is a schematic diagram of the structure of an electronic device provided by an optional embodiment of the present invention, such as Fig.12 As shown, the electronic device may include: at least one processor 61, such as a CPU (Central Processing Unit), at least one communication interface 63, a memory 64, and at least one communication bus 62. The communication bus 62 is used to realize the connection and communication between these components. The communication interface 63 may include a display screen (Display), a keyboard (Keyboard), and the optional communication interface 63 may also include a standard wired interface and a wireless interface. The memory 64 may be a high-speed RAM memory (Random Access Memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 64 may optionally be at least one storage device located away from the aforementioned processor 61. The processor 61 may be combined with Fig.11 In the described device, the memory 64 stores an application program, and the processor 61 calls the program code stored in the memory 64 to execute any of the above method steps.

[0183] The communication bus 62 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 62 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.12 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0184] Among them, the memory 64 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as a flash memory (English: flash memory), a hard disk drive (English: hard disk drive, abbreviated: HDD) or a solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 64 may also include a combination of the above types of memory.

[0185] The processor 61 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.

[0186] The processor 61 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0187] Optionally, the memory 64 is also used to store program instructions. The processor 61 can call the program instructions to implement the present application Figure 1-10 The method for detecting the quality of physiological signals is shown in the embodiment.

[0188] The embodiment of the present invention further provides a non-transitory computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions can execute the method for detecting the quality of physiological signals in any of the above method embodiments. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above types of memories.

[0189] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for detecting the quality of a physiological signal, characterized in that: include: Acquire the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal, wherein the periodic signal in the waveform of the target physiological signal refers to the periodic signal caused by the periodic mechanical activity of the target part in the living body, and the waveform of the previous cycle is the waveform immediately before the waveform of the current cycle; Aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle to obtain the correlation coefficient corresponding to the waveform of the current cycle; Detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period; Wherein, aligning the waveform of the current cycle with the waveform of the previous cycle includes: Constructing a matrix grid using the number of sample points of the waveform of the current cycle and the number of sample points of the waveform of the previous cycle; wherein each grid point of the matrix grid represents the distance between two sample points when the waveform of the current cycle is aligned with two sample points in the waveform of the previous cycle; The matrix grid is searched using an N*N sliding window to determine the alignment grid points of the waveform of the current cycle and the waveform of the previous cycle in the matrix grid; wherein each grid point of the sliding window represents a cumulative distance, and the middle grid point of the sliding window corresponds to the current search grid point in the matrix grid; N represents the number of grid points in each row or column in the sliding window, and the search direction of the sliding window is determined based on the cumulative distance from the current search grid point to the next search grid point.

2. The method according to claim 1, characterized in that The calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle includes: Based on the distance represented by the searched alignment grid points, a correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle is calculated.

3. The method according to claim 2, characterized in that The step of searching the matrix grid using an N*N sliding window to determine, in the matrix grid, alignment grid points of the waveform of the current cycle and the waveform of the previous cycle, includes: In three preset search directions, respectively determining the next search grid point of the current search grid point in the matrix grid; Calculate the cumulative distances between the current search grid point and the corresponding next search grid point in three preset search directions respectively; Determine a preset search direction corresponding to the smallest cumulative distance among the three calculated cumulative distances as the target search direction; The sliding window is moved along the target search direction, and each grid point of the sliding window is updated to perform the next search on the matrix grid; wherein the grid points of the matrix grid corresponding to all the intermediate grid points are the aligned grid points.

4. The method according to any one of claims 1 to 3, characterized in that The detecting the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period includes: Extracting the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficients corresponding to the waveforms of all cycles before the waveform of the current cycle in the target physiological signal waveform; Calculating a target correlation coefficient corresponding to the waveform of the current cycle using all the extracted correlation coefficients; The target correlation coefficient is compared with a preset correlation coefficient to detect the signal quality of the waveform of the current cycle; wherein the preset correlation coefficient corresponds to the signal quality one by one.

5. The method according to claim 4, characterized in that The extracting of the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficient corresponding to the waveforms of all cycles before the waveform of the current cycle in the target physiological signal waveform includes: Obtaining an array of preset length; wherein the elements in the array are used to represent the correlation coefficients corresponding to the waveforms of each period; Circularly moving the array, and inserting the correlation coefficient corresponding to the waveform of the current cycle into the array; All elements in the array are extracted to obtain the correlation coefficient corresponding to the waveform of the current cycle and the correlation coefficient corresponding to the waveforms of all cycles before the waveform of the current cycle in the target physiological signal waveform.

6. The method according to claim 4, characterized in that The step of calculating the target correlation coefficient corresponding to the waveform of the current cycle by using all the extracted correlation coefficients includes: Calculating the mean of all the correlation coefficients to obtain a first target correlation coefficient; Calculating the median of all the correlation coefficients to obtain a second target correlation coefficient; A weighted sum is performed on the first target correlation coefficient and the second target correlation coefficient to obtain the target correlation coefficient.

7. The method according to claim 1, characterized in that The method further comprises: The waveform of the current period is marked with an identifier corresponding to the signal quality of the waveform of the current period.

8. The method according to claim 1 or 7, characterized in that: Before the step of obtaining the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal, the method further includes: Acquiring the target physiological signal waveform; The waveforms of the target physiological signal waveforms that have not been subjected to signal quality detection are marked.

9. A device for detecting the quality of a physiological signal, characterized in that: include: An acquisition module, used to acquire the waveform of the current cycle and the waveform of the previous cycle in the waveform of the target physiological signal, wherein the periodic signal in the waveform of the target physiological signal refers to the periodic signal caused by the periodic mechanical activity of the target part in the living body, and the waveform of the previous cycle is the waveform immediately before the waveform of the current cycle; A correlation coefficient calculation module, used for aligning the waveform of the current cycle with the waveform of the previous cycle, and calculating the correlation coefficient between the waveform of the current cycle and the waveform of the previous cycle, so as to obtain the correlation coefficient corresponding to the waveform of the current cycle; A signal quality detection module, used to detect the signal quality of the waveform of the current period according to the correlation coefficient corresponding to the waveform of the current period; Wherein, aligning the waveform of the current cycle with the waveform of the previous cycle includes: Constructing a matrix grid using the number of sample points of the waveform of the current cycle and the number of sample points of the waveform of the previous cycle; wherein each grid point of the matrix grid represents the distance between two sample points when the waveform of the current cycle is aligned with two sample points in the waveform of the previous cycle; The matrix grid is searched using an N*N sliding window to determine the alignment grid points of the waveform of the current cycle and the waveform of the previous cycle in the matrix grid; wherein each grid point of the sliding window represents a cumulative distance, and the middle grid point of the sliding window corresponds to the current search grid point in the matrix grid; N represents the number of grid points in each row or column in the sliding window, and the search direction of the sliding window is determined based on the cumulative distance from the current search grid point to the next search grid point.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for detecting the quality of a physiological signal according to any one of claims 1 to 8 by executing the computer instructions.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method for detecting physiological signal quality according to any one of claims 1 to 8.

Images