Physiological signal quality evaluation method based on human intelligence, electronic equipment and medium

Abstract

The invention discloses a physiological signal quality evaluation method based on human factor intelligence, electronic equipment and a medium, the physiological signal quality evaluation method based on human factor intelligence comprises the following steps: acquiring a physiological signal, and performing wave crest marking processing on the physiological signal to obtain an interval sequence of the physiological signal; performing segmented signal anomaly identification on the physiological signal based on the interval sequence to obtain an abnormal segment and an anomaly index corresponding to the abnormal segment; and based on the abnormal indexes corresponding to the abnormal segments, carrying out overall quality evaluation on the physiological signal to obtain an overall quality evaluation result of the physiological signal. According to the physiological signal quality assessment method based on human intelligence, the quality of each small fragment in the physiological signal can be assessed, the whole is assessed according to the fragment assessment result, local and whole are both considered, the assessment method is more detailed, and the granularity of the assessment result is better.

Description

Technical Field

[0001] The present application relates to technical fields such as physiological data processing, artificial intelligence, ergonomics and ergonomics, and in particular to a physiological signal quality assessment method, electronic equipment and medium based on ergonomic intelligence. Background Art

[0002] Signal quality assessment usually uses the signal-to-noise ratio as an evaluation indicator. The signal-to-noise ratio parameter is an overall evaluation parameter for the signal. However, signal quality assessment based on the signal-to-noise ratio has shortcomings when applied to physiological signal quality assessment. For example, for PPG (Photo Plethysmo Graphy) signals, PPG signals are subject to various types of interference during acquisition, resulting in the power of the pure signal and the noise signal of the PPG signal being difficult to obtain. Therefore, the signal quality assessment method based on the signal-to-noise ratio is difficult to apply to such signals.

[0003] In addition, the signal quality assessment algorithm usually assesses an entire segment of data and cannot indicate which small segments of the entire segment of data have data quality that meets the requirements, and the granularity of the assessment result is not detailed enough. Summary of the invention

[0004] To this end, the purpose of the implementation mode of the present application is to propose a physiological signal quality assessment method based on human factors intelligence, an electronic device, a storage medium and a computer program product. The physiological signal assessment method of the present invention can evaluate the quality of each small segment in the physiological signal, and evaluate the whole based on the segment assessment results, taking into account both the local and the overall, the assessment method is more detailed, and the granularity of the assessment results is better.

[0005] An embodiment of the present application provides a method for evaluating the quality of physiological signals based on human factors intelligence, the method comprising: acquiring a physiological signal, performing peak marking processing on the physiological signal, and obtaining an interval sequence of the physiological signal; performing segmented signal abnormality identification on the physiological signal based on the interval sequence, and obtaining an abnormal segment and an abnormal index corresponding to the abnormal segment; performing an overall quality evaluation on the physiological signal based on the abnormal index corresponding to the abnormal segment, and obtaining an overall quality evaluation result of the physiological signal.

[0006] Exemplarily, the segmented signal abnormality identification of the physiological signal based on the interval sequence includes: identifying electrode abnormality of the segmented signal of the physiological signal based on the interval sequence; in response to the completion of the electrode abnormality identification result, identifying amplitude abnormality and / or time abnormality of the segmented signal of the physiological signal based on the interval sequence; or, in response to the electrode abnormality identification result satisfying a predetermined condition, removing the abnormal segment and returning to the step of acquiring the physiological signal.

[0007] Exemplarily, the abnormal segment includes an electrode abnormal segment; the abnormal index includes an electrode abnormal index; the segmented signal abnormality identification based on the interval sequence to obtain the abnormal segment in the physiological signal and the abnormal index corresponding to the abnormal segment includes: performing differential processing on the physiological signal based on the interval sequence to obtain a differential physiological signal; when the differential physiological signal is less than a first preset threshold value for a first consecutive preset time, determining that the physiological signal segment corresponding to the first consecutive preset time is an electrode abnormal segment; and determining the electrode abnormality index corresponding to the electrode abnormal segment.

[0008] Exemplarily, the interval sequence includes a peak amplitude sequence; the abnormal segment includes an amplitude abnormal segment; the abnormal index includes an amplitude abnormal index; the segmented signal abnormality identification based on the interval sequence to obtain the abnormal segments in the physiological signal and the abnormal index corresponding to the abnormal segment includes: based on the peak amplitude sequence, determining that the segment corresponding to the signal amplitude greater than the second preset threshold and / or the signal amplitude less than the third preset threshold in the physiological signal is an amplitude abnormal segment; determining the amplitude abnormal index corresponding to the amplitude abnormal segment.

[0009] Exemplarily, the interval sequence includes a pulse wave time difference sequence; the abnormal segment includes a time abnormal segment; the abnormal index includes a time abnormal index; the segmented signal abnormality identification based on the interval sequence to obtain the abnormal segment in the physiological signal and the abnormal index corresponding to the abnormal segment includes: based on the pulse wave time difference sequence, determining that the segment corresponding to the peak interval in the physiological signal is greater than the fourth preset threshold as the time abnormal segment, and / or, based on the pulse wave time difference sequence, determining that the segment corresponding to the heart rate in the physiological signal is greater than the fifth preset threshold as the time abnormal segment, wherein the heart rate is obtained based on the peak interval in the physiological signal; determining the time abnormality index corresponding to the time abnormal segment.

[0010] Exemplarily, the method of performing segmented signal abnormality identification on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments includes: when the time interval between two adjacent electrode abnormal segments is less than a second preset time, updating the two adjacent electrode abnormal segments and the segment between the two adjacent electrode abnormal segments to electrode abnormal segments; and / or calculating a first ratio of the duration of the electrode abnormal segment to the duration of the physiological signal; determining the first ratio as the electrode abnormality index, wherein the electrode abnormality index characterizes the degree of electrode abnormality in the physiological signal.

[0011] Exemplarily, the segmented signal abnormality identification of the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormal indexes corresponding to the abnormal segments includes: when the time interval between two adjacent amplitude abnormal segments is less than a third preset time, updating the two adjacent amplitude abnormal segments and the segment between the two adjacent amplitude abnormal segments to amplitude abnormal segments; and / or calculating a second ratio of the duration of the amplitude abnormal segment to the duration of the physiological signal; determining the second ratio as the amplitude abnormality index, wherein the amplitude abnormality index characterizes the degree of amplitude abnormality in the physiological signal.

[0012] Exemplarily, the segmented signal abnormality identification is performed on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormal indexes corresponding to the abnormal segments, including: when the time interval between two adjacent time abnormality segments is less than a fourth preset time, updating the two adjacent time abnormality segments and the segment between the two adjacent time abnormality segments to time abnormality segments; and / or calculating a third ratio of the duration of the time abnormality segment to the duration of the physiological signal; determining the third ratio as the time abnormality index, wherein the time abnormality index characterizes the degree of time abnormality in the physiological signal.

[0013] Exemplarily, the method further includes: marking the abnormal segments to visualize the overall quality assessment results of the physiological signal, wherein the abnormal segments include electrode abnormal segments, and the marking of the abnormal segments includes: marking the electrode abnormal segments based on the start and end point information of the electrode abnormal segments; and / or, the abnormal segments include amplitude abnormal segments, and the marking of the abnormal segments includes: determining the first peak position information closest to the start and end point information of the amplitude abnormal segment based on the start and end point information of the amplitude abnormal segment, and marking the amplitude abnormal segment based on the first peak position information; and / or, the abnormal segments include time abnormal segments, and the marking of the abnormal segments includes: determining the second peak position information closest to the start and end point information of the time abnormal segment based on the start and end point information of the time abnormal segment, and marking the second peak position information to mark the time abnormal segment.

[0014] Exemplarily, the overall quality assessment result of the physiological signal includes an overall quality assessment index of the physiological signal, and the overall quality assessment result of the physiological signal is obtained by performing an overall quality assessment on the physiological signal based on the abnormality index corresponding to the abnormal segment, including: weighting at least two of the electrode abnormality index, the amplitude abnormality index, and the time abnormality index based on the weight corresponding to each index to obtain a first sum; and obtaining the overall quality assessment index of the physiological signal based on the first sum; wherein the electrode abnormality index corresponds to a first weight, the amplitude abnormality index corresponds to a second weight, and the time abnormality index corresponds to a third weight, and the first weight is greater than the second weight, and the second weight is greater than the third weight.

[0015] Exemplarily, the method further includes: when the overall quality assessment index of the physiological signal is less than a preset threshold, removing the physiological signal corresponding to the overall quality assessment index of the physiological signal.

[0016] Exemplarily, the method further includes: removing the abnormal segment based on the overall quality assessment result of the physiological signal to obtain a first physiological signal; and integrating the first physiological signal so as to perform algorithm processing based on the integrated first physiological signal.

[0017] Exemplarily, the physiological signal includes multiple pulse waves, and the peak marking processing of the physiological signal includes: identifying the peak position sequence of the physiological signal and the peak amplitude sequence corresponding to the peak position sequence based on a preset algorithm; obtaining a pulse wave time difference sequence based on the peak position sequence; wherein, before identifying the physiological signal based on the interval sequence, the method also includes: determining the peak amplitude median based on the peak amplitude sequence; and normalizing the physiological signal based on the peak amplitude median.

[0018] Exemplarily, the first preset time is the product of the mean of the pulse wave time difference and a first preset coefficient, the second preset threshold is the product of the median of the peak amplitude and the second preset coefficient, the third preset threshold is the product of the median of the peak amplitude and the third preset coefficient, and the fourth preset threshold is the superposition or difference of the mean of the pulse wave time difference and a preset multiple of standard deviations; wherein the mean of the pulse wave time difference is obtained based on the pulse wave time difference sequence, and the second preset coefficient is greater than the third preset coefficient.

[0019] Another embodiment of the present application provides a physiological signal evaluation device, which includes: a marking module, which is used to acquire a physiological signal, perform peak marking processing on the physiological signal, and obtain an interval sequence of the physiological signal; an identification module, which is used to perform segmented signal abnormality identification on the physiological signal based on the interval sequence, and obtain abnormal segments and abnormal indexes corresponding to the abnormal segments; an evaluation module, which is used to perform overall quality evaluation on the physiological signal based on the abnormal index corresponding to the abnormal segment, and obtain an overall quality evaluation result of the physiological signal.

[0020] Another embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of any of the above embodiments when executing the computer program.

[0021] Another embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method of any of the above embodiments are implemented.

[0022] Another embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed by a processor of a computer device, the computer device is enabled to perform the steps of the method of any of the above embodiments.

[0023] In the above implementation, the physiological signal quality assessment method based on human intelligence: obtains the physiological signal, performs peak marking processing on the physiological signal, and obtains the interval sequence of the physiological signal; based on the interval sequence, segmented signal abnormality identification is performed on the physiological signal to obtain abnormal segments and abnormal indexes corresponding to the abnormal segments; based on the abnormal indexes corresponding to the abnormal segments, the overall quality assessment of the physiological signal is performed to obtain the overall quality assessment result of the physiological signal. The physiological signal quality assessment method based on human intelligence of the present invention can evaluate the quality of each small segment in the physiological signal, and evaluate the whole according to the segment evaluation results, taking into account both the local and the whole, the evaluation method is more detailed, and the granularity of the evaluation result is better. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of a method for evaluating physiological signal quality based on human intelligence provided in an embodiment of the present application;

[0025] Figure 2 A flowchart of a peak marking process provided for an embodiment of the present application;

[0026] Figure 3 A flowchart of the normalization process provided for the implementation mode of the present application;

[0027] Figure 4 A flow chart of a cascade anomaly assessment scheme provided for an embodiment of the present application;

[0028] Figure 5 A flow chart of electrode abnormality assessment provided for an embodiment of the present application;

[0029] Figure 6 A flowchart of an amplitude anomaly assessment provided for an embodiment of the present application;

[0030] Figure 7 A flow chart of temporal anomaly assessment provided for an embodiment of the present application;

[0031] Figure 8 A flow chart of obtaining an overall quality assessment result of a physiological signal provided in an embodiment of the present application;

[0032] Fig. 9 The overall framework diagram of the PPG signal evaluation method provided in the embodiment of the present application;

[0033] Fig.10 (a) is a schematic diagram of the physiological signal evaluation results provided by an embodiment of the present application;

[0034] Fig.10 (b) is a schematic diagram of a physiological signal evaluation result provided by another embodiment of the present application;

[0035] Fig.10 (c) is a schematic diagram of a physiological signal evaluation result provided by another embodiment of the present application;

[0036] Fig.11 A schematic diagram of a human factor intelligent physiological signal quality assessment device provided in an embodiment of the present application;

[0037] Fig.12 A block diagram of an electronic device provided for an embodiment of the present application. DETAILED DESCRIPTION

[0038] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0039] In signal quality assessment, the parameter of signal-to-noise ratio is usually used as an evaluation index. However, the inventors found in the process of practicing this application that the signal quality assessment based on signal-to-noise ratio is insufficient when applied to the quality assessment of physiological signals. For example, physiological signals that record heartbeats or pulse waves, including but not limited to ECG (Electro Cardio Gram) signals, PPG (Photo Plethysmo Graphy) signals, and other types of physiological signals including ECG and / or PPG are all random signals with obvious individual differences. Signal quality assessment based on signal-to-noise ratio is difficult to show the individual differences of signals. In addition, these signals are subject to various types of interference during collection, and the influence of interference signals on physiological signals is uncertain, which makes it difficult to obtain the power of pure signals and noise signals, so the signal quality assessment method based on signal-to-noise ratio is difficult to apply to such signals.

[0040] In addition, the signal quality assessment algorithm usually assesses an entire segment of data and cannot indicate which small segments of the entire segment of data have data quality that meets the requirements, and the granularity of the assessment result is not detailed enough.

[0041] Based on this, this application proposes a physiological signal quality assessment method based on human intelligence, designs a multi-angle signal quality assessment algorithm, realizes the quality assessment of each small fragment in the physiological signal, and evaluates the whole according to the fragment assessment results, taking into account both the local and the whole, and avoiding the calculation of the power of pure signals and noise.

[0042] In some examples, the physiological signals recording heartbeats or pulse waves are random signals that vary with the state of the person; at the same time, the signal acquisition process will be subject to various types of interference, such as baseline drift, power frequency interference, motion interference, respiratory rhythm interference, etc. These interference signals are all random signals, and the sources of these signals are such as AC interference, limb movement, breathing, etc. This application performs abnormal evaluation on physiological signals and determines abnormal segments in physiological signals to eliminate the influence of interference signals on physiological signals, and can obtain physiological signals with better signal quality for subsequent algorithm training or state evaluation and other processing.

[0043] Figure 1 It is a flowchart of a method for evaluating physiological signal quality based on human factor intelligence according to an embodiment of the present application.

[0044] As an example, Figure 1 As shown, the physiological signal evaluation method includes:

[0045] S101, acquiring a physiological signal, performing peak marking processing on the physiological signal, and obtaining an interval sequence of the physiological signal.

[0046] S102, performing segmented signal abnormality identification on the physiological signal based on the interval sequence to obtain abnormal segments and abnormal indexes corresponding to the abnormal segments.

[0047] S103, performing an overall quality assessment on the physiological signal based on the abnormality index corresponding to the abnormal segment, and obtaining an overall quality assessment result of the physiological signal.

[0048] Exemplarily, physiological signals can be acquired through an acquisition device, and the acquisition device includes, for example, a portable bio-information acquisition device. The subject wears the portable bio-information acquisition device and acquires physiological signals through the portable bio-information acquisition device. Physiological signals include signals that record heartbeats or pulse waves, including but not limited to ECG signals and PPG signals. In the process of acquiring physiological signals, the subject can perform low-intensity activities to introduce various types of interference signals. The collected physiological signals are subjected to peak marking processing. The physiological signals include multiple pulse waves. The pulse wave peaks in the physiological signals are marked to obtain the interval sequence of the physiological signals. Based on the interval sequence, the physiological signals are segmented and signal abnormalities are identified to obtain abnormal segments, which are small segments in the physiological signals. The present application adopts a multi-angle evaluation algorithm to identify physiological signals. It can be understood that abnormal segments include multiple types of abnormal segments. And the abnormal segments correspond to abnormal indexes, which can be used to measure the severity of a certain abnormal situation. Based on the abnormal index corresponding to the abnormal segment, the physiological signal is evaluated for overall quality, and finally the overall quality evaluation result of the physiological signal is obtained.

[0049] The physiological signal evaluation method of the present application evaluates the quality of each small segment in the physiological signal, and evaluates the whole according to the segment evaluation results, taking into account both the local and the whole, and avoiding the calculation of the power of the pure signal and the noise.

[0050] In some examples, when collecting physiological signals such as heartbeat or pulse wave, due to inconsistent wearing conditions of the collected sensors or different baseline data of different people, the physiological signals have individual differences and will change with the status of the person. At the same time, the signal collection process will be subject to various types of interference, such as baseline drift, power frequency interference, motion interference, respiratory rhythm interference, etc. These interference signals are all random signals, and the sources of these signals include AC interference, limb movement, breathing, etc.

[0051] The present application performs segmented abnormality assessment on physiological signals, determines abnormal segments in physiological signals, and eliminates the impact of individual differences on physiological signals, thereby obtaining physiological signals with better signal quality for subsequent algorithm training or state assessment and other processing.

[0052] As an example, the physiological signal is explained by taking the PPG signal as an example. The PPG signal of the subject is collected using a portable bio-information collection device.s After collecting physiological signals, they are preprocessed, including filtering. For example, the effective frequency band of PPG signals is [0.5-8] Hz, and other frequency band signals can be considered as noise signals. It is necessary to filter out such noise signals that can be distinguished by frequency band. 0.5-8 Hz bandpass filtering can be used to obtain the filtered PPG signal, which can be recorded as PPG clean .

[0053] As an example, Figure 2 As shown, the physiological signal includes multiple pulse waves, and peak marking processing is performed on the physiological signal, including:

[0054] S201, identifying and obtaining a peak position sequence of a physiological signal and a peak amplitude sequence corresponding to the peak position sequence based on a preset algorithm.

[0055] S202, obtaining a pulse wave time difference sequence based on the wave peak position sequence.

[0056] For example, a physiological signal such as a PPG signal includes multiple pulse waves, and a peak marking process is performed on the physiological signal. The physiological signal can be identified based on a preset algorithm to obtain a peak position sequence of the physiological signal. The preset algorithm is, for example, the Pan-Tompkins algorithm, and of course, other peak recognition algorithms can also be used. The position of each peak of the pulse wave of the physiological signal is identified based on the preset algorithm, and the peak position can be recorded as R. The positions of the peaks of the multiple pulse waves constitute a peak position sequence, which is composed of R1, R2, R3, R4, ..., R n Among them, n is the physiological signal PPG after preprocessing clean The total number of pulse waves in the. The amplitude of each pulse wave peak can be recorded as A, and the peak amplitude sequence corresponding to the peak position sequence is A1, A2, A3, A4, ..., A n Based on the peak position sequence, the pulse wave time difference sequence is obtained. The time difference between the two pulse waves is obtained based on the peak position sequence. For example, if the peak positions of R2 and R1 are known, the difference between the two is the time interval between the pulse waves, which can be recorded as NN. The pulse wave time difference sequence consists of N1, N2, N3, ..., N n-1 ,composition.

[0057] As an example, Figure 3 As shown, before the physiological signal is identified based on the interval sequence, the physiological signal evaluation method further includes:

[0058] S301, determining a peak amplitude median value based on a peak amplitude sequence.

[0059] S302: Normalize the physiological signal based on the median value of the peak amplitude.

[0060] For example, the physiological signal after preprocessing is normalized. Based on the peak amplitude sequence A1, A2, A3, A4, ..., A n The median value of the peak amplitude is obtained, which can be recorded as A median , the physiological signal is normalized based on the median value of the peak amplitude. For example, the value of each sampling point in the physiological signal can be divided by the median value of the peak amplitude A median , the normalized physiological signal can be recorded as PPG normalized It should be noted that the pulse wave of the physiological signal is composed of a large number of sampling points.

[0061] Normalizing the physiological signals unifies the amplitude range of the physiological signals, which facilitates the subsequent evaluation of the physiological signals. It should be noted that the subsequent evaluation is the physiological signals after normalization and peak marking. This application uses a multi-angle signal evaluation algorithm to evaluate the quality of physiological signals. The following is a specific description of the quality evaluation of physiological signals.

[0062] The present application evaluates the quality of physiological signals based on multiple abnormal identification methods. For example, during the data collection process, there may be a situation where the electrode falls off. At this time, the signal collected by the sensor is not the actual required signal. Therefore, when performing data quality evaluation, the present application determines whether the human-related intelligent physiological signal contains a lead-off situation based on the electrode abnormality identification method. For example, during the signal collection process, a higher amplitude may appear. The larger amplitude will have a negative impact on the subsequent PPG analysis and use. Therefore, the present application also considers the impact of high amplitudes in the signal quality evaluation process of human-related intelligent physiological signals based on the amplitude abnormality identification method. For example, the human-related intelligent physiological signal acquisition process will introduce interference, affecting the calculation of the NN interval. The present application also evaluates the quality of physiological signals based on the time abnormality identification method.

[0063] It should be noted that the abnormality identification method of the present application is not limited to the above three types, and can also be other types of abnormality identification methods, such as power frequency interference abnormality identification, etc.

[0064] As an example, Figure 4 As shown, segmented signal abnormality recognition of physiological signals is performed based on the interval sequence, including:

[0065] S401, identifying electrode abnormalities of segmented signals of physiological signals based on the interval sequence.

[0066] S402, in response to the electrode abnormality identification result being completed, amplitude abnormality identification and / or time abnormality identification of segmented signals of the physiological signal are performed based on the interval sequence; or, in response to the electrode abnormality identification result satisfying a predetermined condition, the abnormal segment is removed and the step of acquiring the physiological signal is returned.

[0067] Exemplarily, the present application may also perform a cascade physiological signal evaluation method according to the degree of influence of the above-mentioned abnormal conditions on the physiological signals. In actual physiological signal applications, the influence of the electrode detachment abnormality on the physiological signals is greater than the influence of other types of abnormal conditions on the physiological signals. Therefore, the electrode abnormality of the segmented signal of the physiological signal can be first identified based on the interval sequence, and after the electrode abnormality identification is completed, the amplitude abnormality and / or time abnormality of the segmented signal of the physiological signal can be identified based on the interval sequence.

[0068] Exemplarily, the electrode abnormality recognition result includes an electrode abnormality segment. There is also a situation where the electrode falls off seriously. At this time, the wearable device is not worn correctly or the wearable device has no contact with the subject, resulting in the acquisition device failing to collect physiological signals. In this case, in response to the electrode abnormality recognition result meeting a predetermined condition, the preset condition can be set to a duration of the electrode abnormality segment that is greater than a set continuous long time. At this time, the electrode abnormality segment can be directly discarded without subsequent quality assessments such as amplitude abnormality assessment and time abnormality assessment, and the step of obtaining physiological signals is returned to reacquire the physiological signals. For example, the staff can readjust the wearable device to obtain physiological signals. This can save computing resources and improve the overall signal evaluation efficiency.

[0069] The cascade quality assessment method proposed in this application is conducive to saving computing resources, improving quality analysis efficiency and reducing processing delays when performing real-time analysis of data acquisition quality.

[0070] The following is a detailed description of electrode anomaly recognition, amplitude anomaly recognition, and time anomaly recognition.

[0071] As an example, Figure 5 As shown, the abnormal segment includes the electrode abnormal segment; the abnormal index includes the electrode abnormal index; based on the interval sequence, the physiological signal is segmented and the signal abnormality is identified to obtain the abnormal segment in the physiological signal and the abnormal index corresponding to the abnormal segment, including:

[0072] S501, performing differential processing on the physiological signal based on the interval sequence to obtain a differential physiological signal.

[0073] S502: When the differential physiological signal is less than the first preset threshold value for a first preset time, determine that the physiological signal segment corresponding to the first preset time is an electrode abnormality segment.

[0074] S503, determining the electrode abnormality index corresponding to the electrode abnormality segment.

[0075] For example, since during the data collection process, there may be an electrode detachment, the signal collected by the sensor is not the actual required signal. Therefore, when performing data quality assessment, it can be determined whether the data contains an electrode detachment. Usually, the values ​​collected after the electrode detachment are close to 0 and relatively fixed values, which can be used as a basis for evaluating the lead detachment.

[0076] Exemplarily, the physiological signal is identified based on the interval sequence, including identifying whether there is an electrode abnormality, and the abnormal segment includes the electrode abnormal segment. First, the physiological signal is differentially processed based on the interval sequence to obtain a differential physiological signal, for example, to calculate the PPG normalized The first-order difference of is used to obtain the differential physiological signal PPG diff1 , determine whether there is a continuous first preset time when the differential physiological signal is less than the first preset threshold. If the differential physiological signal is less than the first preset threshold for a continuous first preset time, it is considered that the physiological signal segment corresponding to the continuous first preset time has an electrode detachment, and the physiological signal segment corresponding to the continuous first preset time is determined to be an electrode abnormal segment. The first preset threshold can be set to 0.01. Determine the electrode abnormality index corresponding to the electrode abnormal segment according to the electrode abnormal segment.

[0077] For example, the R peak positions closest to the start and end of the electrode abnormality segment can be recorded. After traversing the entire data, a series of electrode abnormality segments and the start and end points R of the corresponding segments may be generated. sd ,R ed (R sd As the starting point, R ed is the end point), where R sd Can include R sd1 ,R sd2 ,R sd3 ,...,R sdk1 , corresponding to, R ed Contains R ed1 ,R ed2 ,R ed3 ,...,R edk1 , k1 is the number of abnormal electrode fragments.

[0078] As an example, segmented signal abnormality recognition is performed on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments, including:

[0079] When the time interval between two adjacent electrode abnormality segments is less than a second preset time, the two adjacent electrode abnormality segments and the segment between the two adjacent electrode abnormality segments are updated as electrode abnormality segments;

[0080] and / or,

[0081] Calculating a first ratio of the duration of the electrode abnormal segment to the duration of the physiological signal;

[0082] The first ratio is determined as an electrode abnormality index, wherein the electrode abnormality index represents a degree of electrode abnormality in the physiological signal.

[0083] Exemplarily, the electrode abnormality segment may include multiple segments. The present application also integrates the obtained electrode abnormality segments. When the time interval between two adjacent electrode abnormality segments is less than the second preset time, the two adjacent electrode abnormality segments and the segment between the two adjacent electrode abnormality segments are updated as electrode abnormality segments. The second preset time can be set to 0.2s. For example, if the time interval between two adjacent electrode abnormality segments is less than 0.2s, the two electrode abnormality segments are merged into one electrode abnormality segment. For example, the middle of two electrode abnormality segments can also be marked as an electrode abnormality segment.

[0084] For example, the present application can also calculate the electrode abnormality index according to the electrode abnormality segment, and the electrode abnormality index can be recorded as m1. The electrode abnormality index m1 can be obtained according to the duration of the electrode abnormality segment and the duration of the physiological signal. According to the start and end points R of the electrode abnormality segment recorded above, sd ,R ed , the duration of each electrode abnormal segment can be obtained, and the total electrode abnormal segment duration is added up. The electrode abnormal segment duration is divided by the physiological signal duration to obtain the electrode abnormality index m1. It can be understood that the electrode abnormality index m1 is used to measure the severity of electrode detachment in the physiological signal. The larger the electrode abnormality index m1, the more serious the electrode detachment.

[0085] As an example, Figure 6 As shown, the interval sequence includes a peak amplitude sequence, the abnormal segment includes an amplitude abnormal segment, and the abnormal index includes an amplitude abnormal index. Based on the interval sequence, segmented signal abnormality recognition is performed on the physiological signal to obtain abnormal segments in the physiological signal and abnormal indexes corresponding to the abnormal segments, including:

[0086] S601, based on a peak amplitude sequence, determining that a segment in a physiological signal in which a signal amplitude is greater than a second preset threshold and / or a signal amplitude is less than a third preset threshold is an abnormal amplitude segment.

[0087] S602, determining an amplitude anomaly index corresponding to the amplitude anomaly segment.

[0088] For example, during the signal acquisition process, a higher amplitude may appear, which will have a negative impact on the subsequent PPG analysis and use. Therefore, the signal quality evaluation process of the present application also considers the impact of the high amplitude. Based on the peak amplitude sequence, the segments corresponding to the signal amplitude greater than the second preset threshold and / or the signal amplitude less than the third preset threshold in the physiological signal are determined as amplitude abnormal segments. For example, when the PPG is found normalized The signal amplitude is greater than the second preset threshold and PPG normalized The segment whose signal amplitude is greater than the third preset threshold is determined as an amplitude abnormal segment. The second preset threshold can be the median of the peak amplitude A. median Set, for example, the second preset threshold to 1.8*A median , the third preset threshold is 0.4*A mrdian And determine the amplitude anomaly index corresponding to the amplitude anomaly segment according to the amplitude anomaly segment.

[0089] For example, the R peak positions closest to the start and end of the amplitude abnormal segment can be recorded. After the analysis of the entire data is completed, a series of amplitude abnormal segments and the start and end points R of the corresponding segments are obtained. sa ,R ea , where R sa Contains R sa1 ,R sa2 ,R sa3 ,...,R sak2 , k2 is the number of abnormal amplitude segments, R ea Contains R ea1 ,R ea2 ,R ea3 ,...,R eak2 .

[0090] As an example, segmented signal abnormality recognition is performed on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments, including:

[0091] When the time interval between two adjacent amplitude abnormal segments is less than a third preset time, the two adjacent amplitude abnormal segments and the segment between the two adjacent amplitude abnormal segments are updated as the amplitude abnormal segment;

[0092] and / or,

[0093] Calculating a second ratio of the duration of the abnormal amplitude segment to the duration of the physiological signal;

[0094] The second ratio is determined as an amplitude anomaly index, wherein the amplitude anomaly index represents a degree of amplitude anomaly in the physiological signal.

[0095] Exemplarily, the amplitude abnormality segment may include multiple segments. The present application also integrates the obtained amplitude abnormality segments. When the time interval between two adjacent amplitude abnormality segments is less than the third preset time, the two adjacent amplitude abnormality segments and the segment between the two adjacent amplitude abnormality segments are updated to the amplitude abnormality segment. The third preset time can be set to 0.2s. For example, if the time interval between two adjacent amplitude abnormality segments is less than 0.2s, the two amplitude abnormality segments are merged into one amplitude abnormality segment. For example, the middle of two amplitude abnormality segments can also be marked as an amplitude abnormality segment.

[0096] For example, the present application can also calculate the amplitude abnormality index according to the amplitude abnormality segment, which can be recorded as m2. The amplitude abnormality index m2 can be obtained according to the duration of the amplitude abnormality segment and the duration of the physiological signal. According to the start and end points R of the amplitude abnormality segment recorded above, sa ,R ea , the duration of each amplitude abnormal segment can be obtained, and the total amplitude abnormal segment duration is added up. The amplitude abnormal segment duration is divided by the physiological signal duration to obtain the amplitude abnormality index m2. It can be understood that the amplitude abnormality index m2 is used to measure the severity of the amplitude abnormality in the physiological signal. The larger the amplitude abnormality index m2, the more serious the amplitude abnormality.

[0097] As an example, Figure 7 As shown, the interval sequence includes the pulse wave time difference sequence, the abnormal segment includes the time abnormal segment, and the abnormal index includes the time abnormal index. Based on the interval sequence, the physiological signal is segmented and the signal abnormality is identified to obtain the abnormal segment in the physiological signal and the abnormal index corresponding to the abnormal segment, including:

[0098] S701, based on the pulse wave time difference sequence, determine that the segment corresponding to the peak interval in the physiological signal is greater than the fourth preset threshold as a time abnormal segment, and / or, based on the pulse wave time difference sequence, determine that the segment corresponding to the heart rate in the physiological signal is greater than the fifth preset threshold as a time abnormal segment; wherein the heart rate is obtained based on the peak time interval in the physiological signal.

[0099] S702: Determine a temporal anomaly index corresponding to the temporal anomaly segment.

[0100] For example, the time difference sequence of the PPG signal is a very important basis for PPG analysis. However, since the acquisition process may introduce interference, such as baseline drift, power frequency interference, etc., which may affect the calculation of the NN interval, when evaluating the quality of the PPG signal, this application also considers the problem of abnormal time intervals.

[0101] Exemplarily, according to the pulse wave time difference sequence marked in the above steps, the segment corresponding to the peak interval in the physiological signal greater than the fourth preset threshold is determined as a time abnormal segment, and the fourth preset threshold can be set to exceed the time difference sequence NN by three times the standard deviation. For example, in the pulse wave time difference sequence NN, the segment corresponding to the time abnormal segment that exceeds the time difference sequence NN by three times the standard deviation is found. The time difference sequence NN three times the standard deviation can be obtained by first calculating the time difference sequence N1, N2, N3, ..., N n-1 The average value of NN mean , and standard deviation σ, over [NN mean -3σ,NN mean The fragment with a value of +3σ] is a time anomaly fragment. The time anomaly index corresponding to the time anomaly fragment is determined according to the time anomaly fragment.

[0102] For example, the R peak positions closest to the start and end of the time anomaly segment are recorded to obtain a series of time anomaly segments and the start and end points R of the corresponding segments. sn ,R en , where R sn Contains R sn1 ,R sn2 ,R sn3 ,...,R snk3 , k3 is the number of time anomaly segments, R en Contains R en1 ,R en2 ,R en3 ,...,R enk3 .

[0103] Exemplarily, the present application also determines that the segment corresponding to the heart rate in the physiological signal greater than the fifth preset threshold is a time abnormal segment based on the pulse wave time difference sequence, and the heart rate is obtained based on the peak time interval in the physiological signal. For example, the heart rate is obtained by multiplying the inverse of the time interval by 60. The fifth preset threshold can be set to [30, 200], and the segment with a heart rate exceeding the range of [30, 200] is determined as a time abnormal segment. The entire physiological signal is traversed to obtain the starting point closest to the time abnormal segment, and the above R is added. sn ,R en .

[0104] As an example, segmented signal abnormality recognition is performed on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments, including:

[0105] When the time interval between two adjacent time anomaly segments is less than a fourth preset time, the two adjacent time anomaly segments and the segment between the two adjacent time anomaly segments are updated as time anomaly segments;

[0106] and / or,

[0107] calculating a third ratio of the duration of the temporal abnormality segment to the duration of the physiological signal;

[0108] The third ratio is determined as a temporal anomaly index, wherein the temporal anomaly index represents a degree of temporal anomaly in the physiological signal.

[0109] Exemplarily, the time anomaly segment may include multiple segments. The present application also integrates the obtained time anomaly segments. When the time interval between two adjacent time anomaly segments is less than the fourth preset time, the two adjacent time anomaly segments and the segment between the two adjacent time anomaly segments are updated as time anomaly segments. The fourth preset time can be set to 1s. For example, if the time interval between two adjacent time anomaly segments is less than 1s, the two time anomaly segments are merged into a time anomaly segment. For example, the segment between two time anomaly segments can also be marked as an abnormal segment.

[0110] For example, the present application can also calculate the time anomaly index according to the time anomaly segment, which can be recorded as m3. The time anomaly index m3 can be obtained according to the time anomaly segment duration and the physiological signal duration. According to the start and end points R of the time anomaly segment recorded above, sn ,R en , the duration of the abnormal segment in each time period can be obtained, and the total duration of the abnormal time segment can be obtained by adding up. The duration of the abnormal time segment is divided by the duration of the physiological signal to obtain the time anomaly index m3. It can be understood that the time anomaly index m3 is used to measure the severity of the time anomaly in the physiological signal. The larger the time anomaly index m3, the more serious the time anomaly.

[0111] As an example, the human factor intelligent physiological signal evaluation method further includes: marking abnormal segments to visualize the overall quality evaluation result of the physiological signal, wherein:

[0112] The abnormal segment includes an electrode abnormal segment, and marking the abnormal segment includes: marking the electrode abnormal segment based on the start and end point information of the electrode abnormal segment; and / or,

[0113] The abnormal segment includes an amplitude abnormal segment, and marking the abnormal segment includes: determining first wave peak position information closest to the start and end point information of the amplitude abnormal segment based on the start and end point information of the amplitude abnormal segment, and marking the amplitude abnormal segment based on the first wave peak position information; and / or,

[0114] The abnormal segment includes a time abnormal segment, and marking the abnormal segment includes: determining the second peak position information closest to the start and end point information of the time abnormal segment based on the start and end point information of the time abnormal segment, and marking the second peak position information to mark the time abnormal segment.

[0115] Exemplarily, in order to intuitively display the abnormal fragments, the present application also marks the abnormal fragments to visualize the abnormal fragments. For example, for the electrode abnormal fragment, the electrode abnormal fragment can be marked according to the start and end point information of the electrode abnormal fragment, for example, the entire segment within the start and end point of the electrode abnormal fragment is marked with a preset color.

[0116] Exemplarily, in order to intuitively display the amplitude abnormality segment, the present application also performs marking processing on the amplitude abnormality segment to visualize the amplitude abnormality segment. For example, for the amplitude abnormality segment, the closest first wave peak position information can be found according to the start and end point information of the amplitude abnormality segment, and the entire segment within the start and end point of the first wave peak position information is marked with another color.

[0117] Exemplarily, in order to intuitively display the time anomaly segment, the present application also performs marking processing on the time anomaly segment to visualize the time anomaly segment. For example, for the time anomaly segment, the second wave peak position information closest to the start and end point information of the time anomaly segment can be marked with another color. For example, a wave peak dot is marked in dark blue, indicating that the instantaneous heart rate corresponding to the current pulse wave exceeds the range of [30,200], and a wave peak dot is marked in red, indicating that the interval between the current pulse wave peak and the next pulse wave peak is within NN. mean Outside 3 standard deviations.

[0118] After the abnormal segment is identified, the physiological signal is evaluated based on the abnormal index corresponding to the abnormal segment to obtain the overall quality evaluation result of the physiological signal.

[0119] As an example, the overall quality assessment result of the temporal physiological signal includes an overall quality assessment index of the physiological signal, such as Figure 8 As shown, based on the abnormal index corresponding to the abnormal time segment, the overall quality assessment of the temporal physiological signal is performed to obtain the overall quality assessment result of the physiological signal, including:

[0120] S801, performing weighted processing on at least two of the electrode abnormality index, the amplitude abnormality index, and the time abnormality index based on the weight corresponding to each index to obtain a first sum value.

[0121] S802: Obtain an overall quality assessment index of the physiological signal based on the first sum value.

[0122] Among them, the electrode anomaly index corresponds to the first weight, the amplitude anomaly index corresponds to the second weight, and the time anomaly index corresponds to the third weight. The first weight is greater than the second time weight, and the second weight is greater than the third time weight.

[0123] Exemplarily, the abnormality index includes at least two of the electrode abnormality index, the amplitude abnormality index, and the time abnormality index, and at least two of the electrode abnormality index, the amplitude abnormality index, and the time abnormality index are weighted according to the weight corresponding to each index to obtain a first sum value. For example, the abnormality index includes three indexes, namely, the electrode abnormality index, the amplitude abnormality index, and the time abnormality index, and the weighted processing is performed according to the weights corresponding to the three indexes to obtain a first sum value. For example, the electrode abnormality index m1 corresponds to the first weight, the amplitude abnormality index m2 corresponds to the second weight, and the time abnormality index m3 corresponds to the third weight.

[0124] Exemplarily, the size of the weight can be determined according to the degree of influence of abnormal conditions on physiological signals. For example, since electrode detachment has a decisive influence on signal quality, the first weight can be set to 0.5. Under normal circumstances, the PPG amplitude will be stable within a certain range. Excessive or small amplitude is also caused by noise interference. Abnormal amplitude will affect the recognition of pulse waves, so the second weight can be set to 0.3. Since the frequency of the pulse wave changes slowly over time, when the continuous NN sequence changes greatly, the corresponding PPG signal also has certain interference, so the third weight can be set to 0.2. Of course, the setting of the first weight, the second weight and the third weight can also be determined according to needs.

[0125] Exemplarily, weighted processing is performed according to the weights corresponding to the three indexes to obtain a first sum value, for example, the first sum value is equal to (0.5*m1+0.3*m2+0.2*m3). Based on the first sum value, the overall quality assessment index of the physiological signal is obtained, and the overall quality assessment index of the physiological signal can be recorded as m signal , m signal =1-(0.5*m1+0.3*m2+0.2*m3) It can be understood that the larger the overall quality evaluation index of the physiological signal is, the better the quality of the physiological signal is.

[0126] As an example, the human factors intelligent physiological signal evaluation method further includes: when the overall quality evaluation index of the physiological signal is less than a preset threshold, removing the physiological signal corresponding to the overall quality evaluation index of the physiological signal.

[0127] Exemplarily, the present application further processes the physiological signal according to the above-mentioned physiological signal quality evaluation. For example, when the overall quality evaluation index of the physiological signal is less than a preset threshold, it indicates that the overall quality of the physiological signal is poor. The physiological signal can be removed and the physiological signal with good quality can be retained.

[0128] As an example, the human factors intelligent physiological signal evaluation method also includes: removing abnormal segments based on the overall quality evaluation result of the physiological signal to obtain a first physiological signal; integrating the first physiological signal so as to perform algorithm processing based on the integrated first physiological signal.

[0129] Exemplarily, the present application can also optimize the physiological signals based on the overall quality assessment results of the physiological signals, for example, removing abnormal segments in the physiological signals and integrating the remaining physiological signals. The integration method is, for example, based on linear interpolation calculation integration. The physiological signals without abnormal segments are obtained, so as to facilitate subsequent algorithm training, state recognition and other algorithm processing based on the physiological signals.

[0130] For example, during integration, the start and end points R of the abnormal electrode segment recorded above can be used. sd ,R ed , the starting and ending points of the amplitude abnormal segment R sn ,R en ,, the start and end points of the time anomaly segment R sa ,R ea Integrate and obtain the abnormal fragments in the entire data. Remove all abnormal fragments and the rest are usable data, which are provided to users for subsequent applications.

[0131] Fig. 9 It is an overall framework diagram of a PPG signal evaluation method according to an embodiment of the present application.

[0132] like Fig. 9 As shown, the PPG signal is first collected, and then the PPG signal is filtered and processed by using a 0.5-8 Hz bandpass filter to obtain a filtered PPG signal, which is recorded as PPG clean The algorithm is used to identify the pulse wave peak of the PPG signal, and the pulse wave peak position sequence R, the pulse wave peak amplitude sequence A and the time difference sequence NN between the pulse waves are obtained. The median of the pulse wave peak amplitude A is used to identify the PPG signal pulse wave peak. clean Normalization processing is performed. Electrode abnormality assessment, amplitude abnormality assessment and time abnormality assessment are performed on the physiological signals after normalization processing. Finally, the PPG signal quality index and abnormal fragment are calculated based on the evaluation results of electrode abnormality assessment, amplitude abnormality assessment and time abnormality assessment.

[0133] As an example, the first preset time is the product of the mean of the pulse wave time difference and the first preset coefficient, the second preset threshold is the product of the median of the peak amplitude and the second preset coefficient, the third preset threshold is the product of the median of the peak amplitude and the third preset coefficient, and the fourth preset threshold is the superposition or difference of the mean of the pulse wave time difference and a preset multiple of standard deviations;

[0134] The mean value of the pulse wave time difference is obtained based on the pulse wave time difference sequence, and the second preset coefficient is greater than the third preset coefficient.

[0135] For example, the first preset time may be the product of the mean value of the pulse wave time difference and the first preset coefficient. When evaluating electrode abnormality, the PPG value may be first calculated based on the pulse wave peak position R. normalized The average value of the interval NN between two pulse wave fluctuations, that is, the NN mean, is recorded as NN mean , NN mean That is, N1, N2, N3, ..., N n-1 The first preset coefficient can be set to 0.4, and the first preset time can be 0.4*NN meam When evaluating the amplitude abnormality, the second preset threshold can be set as the product of the median value of the peak amplitude and the second preset coefficient, and the third preset threshold can be set as the product of the median value of the peak amplitude and the third preset coefficient. The second preset coefficient can be 1.8, the third preset coefficient can be 0.4, and the second preset threshold can be set to 1.8*A median , the third preset threshold can be set to 0.4*A median When performing time anomaly assessment, the fourth preset threshold can be set to the superposition or difference of the mean of the pulse wave time difference and the preset multiple standard deviation, the preset multiple can be set to 3 times, and the fourth preset threshold can be NN mean ±3σ.

[0136] Fig.10 It is a schematic diagram of the physiological signal evaluation result of an embodiment of the present application.

[0137] like Fig.10 As shown, Fig.10 (a) contains amplitude anomalies and time anomalies, where the time anomaly is due to the heart rate exceeding the range of [30,200]. According to the above signal quality evaluation algorithm, the signal quality index of this data is calculated to be 0.916. Fig.10 (b) contains multiple amplitude anomalies and time anomalies, where the time anomaly is also due to the heart rate exceeding the range of [30,200]. According to the above quality assessment algorithm, the signal quality index of the current data is 0.864. Fig.10 (c) contains two sections of electrode detachment, one section of amplitude anomaly, and one section of time anomaly caused by electrode detachment. Based on the above signal quality evaluation algorithm, the signal quality index of the current data is 0.845.

[0138] For example, it can be seen from the above three figures and their signal quality results that as the signal quality decreases, the signal quality index also decreases. The closer the signal quality index value is to 1, the better the data quality. In actual application, the signal quality can be evaluated according to the signal quality index. If the signal quality index is above 0.95, it can be considered that the signal quality is high. If the signal quality index is in the range of 0.90-0.95, it can be considered that the signal quality is medium. If the data quality is lower than 0.9, the signal quality can be considered to be low.

[0139] For example, after obtaining the abnormal segment, the data can be segmented according to the start and end points of the abnormal segment to obtain a segment with better quality. The data can also be spliced ​​by shifting the pulse wave peak and then interpolating to obtain a complete signal with better quality. The specific method is as follows: for an abnormal segment, such as R sd ,R ed , find R ed Previous and R ed The closest minimum point P min , connect R by interpolation sd and P min , the data length corresponding to the interpolation can be set to NN mean / 2, interpolation method such as linear interpolation method.

[0140] The physiological signal evaluation method of the present application can evaluate physiological signals in a more detailed manner, evaluate physiological signals from multiple angles, and can intuitively display abnormal segments.

[0141] The present application also proposes a physiological signal quality assessment device based on human intelligence.

[0142] like Fig.11 As shown, the human factors intelligent physiological signal quality assessment device includes a marking module 1101, which is used to acquire physiological signals, perform peak marking processing on the time physiological signals, and obtain an interval sequence of the physiological signals; an identification module 1102, which is used to perform segmented signal abnormality identification on the time physiological signals based on the time interval sequence, and obtain abnormal segments and abnormal indexes corresponding to the time abnormal segments; an assessment module 1103, which is used to perform overall quality assessment on the time physiological signals based on the abnormal indexes corresponding to the time abnormal segments, and obtain an overall quality assessment result of the physiological signals.

[0143] The application also proposes a computer-readable storage medium.

[0144] In this embodiment, a computer program is stored on a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned physiological signal quality assessment method based on human factor intelligence are implemented.

[0145] Fig.12 A block diagram of an electronic device provided for an embodiment of the present application.

[0146] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned physiological signal quality assessment method based on human factors intelligence when executing the computer program.

[0147] like Fig.12As shown, for ease of understanding, the embodiment of the present application shows a specific electronic device.

[0148] Electronic devices are intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0149] like Fig.12 As shown, the device includes a computing unit 1201, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1202 or a computer program loaded from a storage unit 1208 into a random access memory (RAM) 1203. In RAM 1203, various programs and data required for the operation of the electronic device can also be stored. The computing unit 1201, ROM 1202, and RAM 1203 are connected to each other via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0150] Multiple components in the electronic device are connected to the I / O interface 1205, including: an input unit 1206, such as a keyboard, a mouse, etc.; an output unit 1207, such as various types of displays, speakers, etc.; a storage unit 1208, such as a disk, an optical disk, etc.; and a communication unit 1209, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0151] The computing unit 1201 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1201 executes the various methods described above, such as the physiological signal quality assessment method based on human intelligence. For example, in some embodiments, the physiological signal quality assessment method based on human intelligence may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 1208. In some embodiments, part or all of the computer program may be loaded and / or installed on an electronic device via ROM 1202 and / or communication unit 1209. When the computer program is loaded into RAM 1203 and executed by the computing unit 1201, the physiological signal quality assessment method based on human intelligence described above may be executed. Alternatively, in other embodiments, the computing unit 1201 may be configured to execute the physiological signal quality assessment method based on human intelligence in any other appropriate manner (for example, by means of firmware).

[0152] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this application, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0153] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0154] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present application, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0155] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0156] In addition, the terms "first", "second", etc. used in the embodiments of the present application are only used for descriptive purposes and should not be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated in the present embodiment. Therefore, the features defined by the terms "first", "second", etc. in the embodiments of the present application can explicitly or implicitly indicate that at least one of the features is included in the embodiment. In the description of the present application, the word "multiple" means at least two or two or more, such as two, three, four, etc., unless otherwise clearly and specifically defined in the embodiments.

[0157] In this application, unless otherwise clearly specified or limited in the embodiments, the terms "installed", "connected", "connected" and "fixed" etc. appearing in the embodiments should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integrated connection. It can be understood that it can also be a mechanical connection, an electrical connection, etc.; of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal connection of two elements, or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific implementation situation.

[0158] In the present application, unless otherwise clearly specified and limited, a first feature being “above” or “below” a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being “above”, “above”, and “above” a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being “below”, “below”, and “below” a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0159] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A physiological signal quality assessment method based on human intelligence, characterized in that: The method comprises: Acquiring a physiological signal, performing peak marking processing on the physiological signal, and obtaining an interval sequence of the physiological signal; Based on the interval sequence, segmented signal abnormality recognition is performed on the physiological signal to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments; Based on the abnormal index corresponding to the abnormal segment, an overall quality assessment of the physiological signal is performed to obtain an overall quality assessment result of the physiological signal.

2. The method according to claim 1, characterized in that: The segmented signal abnormality identification of the physiological signal based on the interval sequence includes: Performing electrode abnormality identification of segmented signals of the physiological signal based on the interval sequence; In response to the completion of the electrode abnormality identification result, the amplitude abnormality and / or time abnormality identification of the segmented signal of the physiological signal is performed based on the interval sequence; or, in response to the electrode abnormality identification result meeting a predetermined condition, the abnormal segment is removed and the step of obtaining the physiological signal is returned.

3. The method according to claim 1, characterized in that The abnormal segment includes an electrode abnormal segment; the abnormal index includes an electrode abnormal index; The step of performing segmented signal abnormality identification on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments includes: performing differential processing on the physiological signal based on the interval sequence to obtain a differential physiological signal; When the differential physiological signal is less than the first preset threshold value for a continuous first preset time, determining that the physiological signal segment corresponding to the continuous first preset time is an electrode abnormal segment; An electrode abnormality index corresponding to the electrode abnormality segment is determined.

4. The method according to any one of claims 1 to 3, characterized in that: The interval sequence includes a peak amplitude sequence; the abnormal segment includes an amplitude abnormal segment; the abnormal index includes an amplitude abnormal index; the segmented signal abnormality identification based on the interval sequence is performed on the physiological signal to obtain the abnormal segment in the physiological signal and the abnormal index corresponding to the abnormal segment, including: based on the peak amplitude sequence, determining that the segment corresponding to the signal amplitude greater than the second preset threshold and / or the signal amplitude less than the third preset threshold in the physiological signal is an amplitude abnormal segment; determining the amplitude abnormal index corresponding to the amplitude abnormal segment; and / or, The interval sequence includes a pulse wave time difference sequence; the abnormal segment includes a time abnormal segment; the abnormal index includes a time abnormal index; the segmented signal abnormality identification based on the interval sequence is performed on the physiological signal to obtain the abnormal segment in the physiological signal and the abnormal index corresponding to the abnormal segment, including: based on the pulse wave time difference sequence, determining that the segment corresponding to the peak interval in the physiological signal is greater than the fourth preset threshold as the time abnormal segment, and / or, based on the pulse wave time difference sequence, determining that the segment corresponding to the heart rate in the physiological signal is greater than the fifth preset threshold as the time abnormal segment, wherein the heart rate is obtained based on the peak interval in the physiological signal; determining the time abnormality index corresponding to the time abnormal segment.

5. The method according to claim 3, characterized in that: The step of performing segmented signal abnormality identification on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments includes: When the time interval between two adjacent electrode abnormality segments is less than a second preset time, the two adjacent electrode abnormality segments and the segment between the two adjacent electrode abnormality segments are updated as the electrode abnormality segment; and / or, Calculating a first ratio of the duration of the electrode abnormal segment to the duration of the physiological signal; The first ratio is determined as the electrode abnormality index, wherein the electrode abnormality index represents the degree of electrode abnormality in the physiological signal.

6. The method according to claim 4, characterized in that The step of performing segmented signal abnormality identification on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments includes: When the time interval between two adjacent amplitude abnormal segments is less than a third preset time, the two adjacent amplitude abnormal segments and the segment between the two adjacent amplitude abnormal segments are updated as the amplitude abnormal segment; and / or, Calculating a second ratio of the duration of the abnormal amplitude segment to the duration of the physiological signal; The second ratio is determined as the amplitude abnormality index, wherein the amplitude abnormality index represents the degree of amplitude abnormality in the physiological signal.

7. The method according to claim 4, characterized in that The step of performing segmented signal abnormality identification on the physiological signal based on the interval sequence to obtain abnormal segments in the physiological signal and abnormality indexes corresponding to the abnormal segments includes: When the time interval between two adjacent time anomaly segments is less than a fourth preset time, the two adjacent time anomaly segments and the segment between the two adjacent time anomaly segments are updated as time anomaly segments; and / or, Calculating a third ratio of the duration of the temporal abnormality segment to the duration of the physiological signal; The third ratio is determined as the temporal anomaly index, wherein the temporal anomaly index represents a degree of temporal anomaly in the physiological signal.

8. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: The abnormal segments are marked to visualize the overall quality assessment result of the physiological signal, wherein: The abnormal segment includes an electrode abnormal segment, and the marking process of the abnormal segment includes: marking the electrode abnormal segment based on the start and end point information of the electrode abnormal segment; and / or, The abnormal segment includes an amplitude abnormal segment, and the marking process for the abnormal segment includes: determining first wave peak position information closest to the start and end point information of the amplitude abnormal segment based on the start and end point information of the amplitude abnormal segment, and marking the amplitude abnormal segment based on the first wave peak position information; and / or, The abnormal segment includes a time abnormal segment, and the marking process for the abnormal segment includes: based on the start and end point information of the time abnormal segment, determining the second peak position information closest to the start and end point information of the time abnormal segment, and marking the second peak position information to mark the time abnormal segment.

9. The method according to claim 1, characterized in that: The overall quality assessment result of the physiological signal includes an overall quality assessment index of the physiological signal. The overall quality assessment of the physiological signal is performed based on the abnormality index corresponding to the abnormal segment to obtain the overall quality assessment result of the physiological signal, including: Performing weighted processing on at least two of the electrode abnormality index, the amplitude abnormality index, and the time abnormality index based on the weight corresponding to each index to obtain a first sum value; Obtaining an overall quality assessment index of the physiological signal based on the first sum; Among them, the electrode anomaly index corresponds to a first weight, the amplitude anomaly index corresponds to a second weight, and the time anomaly index corresponds to a third weight. The first weight is greater than the second weight, and the second weight is greater than the third weight.

10. The method according to claim 9, characterized in that The method further comprises: When the overall quality assessment index of the physiological signal is less than a preset threshold, the physiological signal corresponding to the overall quality assessment index of the physiological signal is removed.

11. The method according to claim 1, characterized in that The method further comprises: Based on the overall quality evaluation result of the physiological signal, the abnormal segment is removed to obtain a first physiological signal; The first physiological signal is integrated so as to perform algorithm processing based on the integrated first physiological signal.

12. The method according to claim 1, characterized in that The physiological signal includes a plurality of pulse waves, and the peak marking process of the physiological signal includes: Identifying a peak position sequence of the physiological signal and a peak amplitude sequence corresponding to the peak position sequence; Obtaining a pulse wave time difference sequence based on the wave peak position sequence; Before identifying the physiological signal based on the interval sequence, the method further includes: Determining a peak amplitude median value based on the peak amplitude sequence; The physiological signal is normalized based on the peak amplitude median value.

13. The method according to claim 3 or 4, characterized in that: The first preset time is the product of the mean of the pulse wave time difference and the first preset coefficient, the second preset threshold is the product of the median of the peak amplitude and the second preset coefficient, the third preset threshold is the product of the median of the peak amplitude and the third preset coefficient, and the fourth preset threshold is the superposition or difference of the mean of the pulse wave time difference and a preset multiple of standard deviations; The mean of the pulse wave time difference is obtained based on the pulse wave time difference sequence, and the second preset coefficient is greater than the third preset coefficient.

14. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the method described in any one of claims 1 to 13 when executing the computer program.

15. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 13 are implemented.

Images