A method, device and system for monitoring a patient's respiratory signal

Abstract

The present invention relates to the field of respiratory monitoring technology, and specifically to a method, device and system for monitoring the respiratory signal of a patient. The present invention analyzes the distribution characteristics of extreme points on a respiratory signal curve to obtain an initial respiratory abnormality factor between adjacent extreme points; according to the time change characteristics between different adjacent extreme points and the change trend of data points between adjacent moments, the noise influence degree between adjacent extreme points is obtained, and the corrected respiratory abnormality factor between adjacent extreme points is obtained; according to the change trend between adjacent extreme points of the same type and the corresponding respiratory abnormality fluctuation factor, the trend characteristic factor of each extreme point of the same type is obtained; and then the data credibility between adjacent extreme points is obtained; the respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve. The present invention takes into account noise interference and improves the accuracy of respiratory monitoring by obtaining relatively accurate respiratory signal data of the patient.

Description

Technical Field

[0001] The present invention relates to the technical field of respiratory monitoring, and in particular to a method, device and system for monitoring a patient's respiratory signal. Background Art

[0002] Monitoring of patients' respiratory signals is an important part of the medical field. It can not only provide real-time assessment of patients' health status, but also improve patients' treatment outcomes through early warning, guided treatment, and other means. However, in the process of collecting patients' respiratory signals, due to interference from sounds in the surrounding environment and noise from medical devices, the collected patients' respiratory signals are not very accurate, resulting in errors in abnormal monitoring of patients' respiratory signals, which reduces the accuracy of monitoring patients' abnormal breathing conditions.

[0003] In the prior art, a filtering algorithm is used to denoise the respiratory signal. However, since the changes in the patient's respiratory volume are different in different situations, the traditional filtering algorithm based on fixed parameter settings often cannot adapt to the signal processing requirements in different situations, cannot effectively remove noise, and the monitoring effect of the respiratory signal is poor. Summary of the invention

[0004] In order to solve the technical problem that the fixed parameter settings under the traditional filtering algorithm often cannot adapt to the signal processing requirements in different situations and the monitoring effect of the respiratory signal is poor, the purpose of the present invention is to provide a patient's respiratory signal monitoring method, device and system, and the technical solutions adopted are as follows:

[0005] The present invention provides a method for monitoring a patient's respiratory signal, the method comprising:

[0006] Obtain the patient's respiratory volume at each moment to form data points of the respiratory signal curve;

[0007] According to the distribution characteristics of the respiratory volume in the neighborhood of each extreme point on the respiratory signal curve, the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points are obtained; according to the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments, the noise influence degree between adjacent extreme points is obtained; according to the initial respiratory abnormality factor and the noise influence degree between adjacent extreme points, the modified respiratory abnormality factor between adjacent extreme points is obtained;

[0008] For extreme value points of the same type, the possibility of emotional mutation of each extreme value point of the same type is obtained according to the change trend between adjacent extreme value points of the same type and the corresponding respiratory abnormality fluctuation factor; the trend characteristic factor of each extreme value point of the same type is obtained according to the distribution of emotional mutation possibility of the extreme value points of the same type and the fluctuation characteristics between the extreme value points of the same type; the data credibility between adjacent extreme value points is obtained according to the difference in trend characteristic factors between adjacent extreme value points and the corrected respiratory abnormality factor;

[0009] According to the data credibility between adjacent extreme value points where each data point on the respiratory signal curve is located, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve.

[0010] Furthermore, the step of obtaining the abnormal breathing fluctuation factor of each extreme point and the initial abnormal breathing factor between adjacent extreme points includes:

[0011] The mean of the respiratory volume in the neighborhood of each extreme point is obtained as the average respiratory volume; the standard deviation of the respiratory volume in the neighborhood of each extreme point is obtained;

[0012] Calculate the difference between the respiratory volume at the corresponding extreme point and the average respiratory volume. If the difference is greater than three times the standard deviation, normalize the difference and use it as the abnormal respiratory fluctuation factor for each extreme point. If the difference is less than or equal to three times the standard deviation, the abnormal respiratory fluctuation factor for each extreme point is 0.

[0013] The mean of the abnormal breathing fluctuation factors between adjacent extreme value points is obtained and normalized as the initial abnormal breathing factor between adjacent extreme value points.

[0014] Furthermore, the method for obtaining the noise impact degree includes:

[0015] Obtain the monotonic curve segments between adjacent extreme value points, obtain the time length of each monotonic curve segment, and obtain the average of the time lengths between adjacent monotonic curve segments before and after each monotonic curve segment as the comparison time length; obtain the average of the slopes of the data points between all adjacent moments on each monotonic curve segment as the first slope; obtain the average of the first slopes between adjacent monotonic curve segments before and after each monotonic curve segment as the second slope;

[0016] If the monotonic curve segment between adjacent extreme value points is monotonically increasing, the length difference between the preset multiple of the time length and the comparison time length is obtained, and the slope difference between the first slope and the preset multiple of the second slope is obtained; if the monotonic curve segment between adjacent extreme value points is monotonically decreasing, the length difference between the time length and the preset multiple of the comparison time length is obtained, and the slope difference between the preset multiple of the first slope and the second slope is obtained;

[0017] The product of the length difference and the slope difference is calculated and normalized to obtain the degree of noise influence between adjacent extreme points.

[0018] Furthermore, the method for obtaining the corrected abnormal breathing factor includes:

[0019] The product of the initial abnormal breathing factor and the noise influence degree between adjacent extreme value points is obtained as the corrected abnormal breathing factor between adjacent extreme value points.

[0020] Furthermore, the method for obtaining the possibility of sudden emotional change includes:

[0021] For extreme value points of the same type, the difference in slope between each extreme value point of the same type and the adjacent extreme value points of the same type before and after is obtained as the first slope difference;

[0022] The product of the first slope difference and the corresponding respiratory abnormality fluctuation factor is calculated as the possibility of emotional mutation at each similar extreme point.

[0023] Furthermore, the method for obtaining the trend characteristic factor includes:

[0024] According to the possibility of emotional mutation of similar extreme value points, all similar extreme value points are clustered to obtain multiple similar clusters;

[0025] Obtain the average of the likelihood of emotional mutation of all similar extreme value points in each similar cluster as the overall emotional mutation likelihood; select the extreme value point in the corresponding similar cluster with the largest overall emotional mutation likelihood as the emotional mutation point;

[0026] Obtain the absolute value of the slope between adjacent emotion mutation points. If the absolute value of the slope is greater than or equal to the preset slope threshold, obtain the absolute value of the accumulated slopes between all adjacent extreme value points of the same type between adjacent emotion mutation points, and perform normalization to obtain the trend characteristic factor of each extreme value point of the same type between adjacent emotion mutation points.

[0027] If the absolute value of the slope is less than the preset slope threshold, the mean of the abnormal breathing fluctuation factors of all similar extreme points between adjacent emotional mutation points is obtained, and negative correlation mapping is performed to obtain the trend characteristic factor of each similar extreme point between adjacent emotional mutation points.

[0028] Furthermore, the method for obtaining the data credibility includes:

[0029] The difference in trend characteristic factors between adjacent extreme value points is obtained as the factor difference; the product of the factor difference and the corrected respiratory abnormality factor is calculated, and negative correlation normalization mapping is performed to obtain the data credibility between adjacent extreme value points.

[0030] Furthermore, the method for obtaining the filtered respiratory volume includes:

[0031] A sliding window is constructed with each data point as the center, and the time length between each data point and the center point in the sliding window is obtained, and a negative correlation mapping is performed as the weight factor of each data point relative to the center point; the product of the weight factor and the corresponding data credibility is calculated and normalized as the weight value of each data point relative to the center point;

[0032] Based on the weight value of each data point relative to the center point, the respiratory volume of the center point is filtered by a weighted filtering algorithm to obtain a filtered respiratory volume.

[0033] The present invention also proposes a patient's respiratory signal monitoring device, including a data acquisition module, a respiratory abnormality factor correction module, a data credibility determination module, and a respiratory curve optimization module:

[0034] Data acquisition module: obtains the patient's respiratory volume at each moment and forms data points of the respiratory signal curve;

[0035] Respiratory abnormality factor correction module: according to the distribution characteristics of the respiratory volume in the neighborhood of each extreme point on the respiratory signal curve, obtain the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points; according to the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments, obtain the noise influence degree between adjacent extreme points; according to the initial respiratory abnormality factor and the noise influence degree between adjacent extreme points, obtain the corrected respiratory abnormality factor between adjacent extreme points;

[0036] Data credibility determination module: For extreme value points of the same type, the possibility of emotional mutation of each extreme value point of the same type is obtained according to the change trend between adjacent extreme value points of the same type and the corresponding respiratory abnormality fluctuation factor; the trend characteristic factor of each extreme value point of the same type is obtained according to the distribution of emotional mutation possibility of extreme value points of the same type and the fluctuation characteristics between extreme value points of the same type; the data credibility between adjacent extreme value points is obtained according to the difference in trend characteristic factors between adjacent extreme value points and the correction of respiratory abnormality factors;

[0037] Respiratory curve optimization module: According to the data credibility between the adjacent extreme value points of each data point on the respiratory signal curve, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve.

[0038] The present invention also proposes a patient's respiratory signal monitoring system, wherein the system stores programs or instructions, and when the programs or instructions are executed by a processor, the steps of the patient's respiratory signal monitoring method as described above are implemented.

[0039] The present invention has the following beneficial effects:

[0040] The present invention obtains the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points according to the distribution characteristics of the respiratory volume in the neighborhood range of each extreme point on the respiratory signal curve, which reflects the degree of change of the respiratory signal between adjacent extreme points and helps to understand the respiratory abnormality; according to the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments, the degree of noise influence between adjacent extreme points is obtained, and the change caused by the noise is further analyzed to obtain the modified respiratory abnormality factor between adjacent extreme points, which more accurately reflects the real change of the respiratory signal; for the same type of extreme points, according to the difference between the adjacent similar extreme points, the difference between the adjacent extreme points is obtained. According to the change trend and the corresponding respiratory abnormality fluctuation factor, the trend characteristic factor of each similar extreme point is obtained, which is helpful to identify the long-term or periodic change law in the respiratory signal; according to the difference in trend characteristic factors between adjacent extreme points and the correction of respiratory abnormality factors, the data credibility between adjacent extreme points is obtained, and the reliability of data between adjacent extreme points is more accurately evaluated, which helps to reduce the impact caused by noise interference; according to the data credibility between adjacent extreme points where each data point on the respiratory signal curve is located, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve, which reduces the interference of noise and abnormal values ​​and more truly reflects the changes in respiratory conditions. The present invention takes noise interference into consideration and improves the accuracy of respiratory monitoring by obtaining relatively accurate respiratory signal data of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A flowchart of a method for monitoring a patient's respiratory signal provided by an embodiment of the present invention;

[0043] Figure 2 A flow chart of a method for obtaining the degree of noise influence provided by an embodiment of the present invention;

[0044] Figure 3 A flow chart of a method for obtaining a trend characteristic factor provided by an embodiment of the present invention;

[0045] Figure 4 This is a structural block diagram of a patient's respiratory signal monitoring device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the method, device and system for monitoring the respiratory signal of a patient proposed by the present invention, its specific implementation, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0047] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0048] The following is a detailed description of a method, device and system for monitoring a patient's respiratory signal provided by the present invention in conjunction with the accompanying drawings.

[0049] See also Figure 1 , which shows a flow chart of a method for monitoring a patient's respiratory signal provided by an embodiment of the present invention, and the specific method includes:

[0050] Step S1: Obtain the patient's respiratory volume at each moment to form data points of the respiratory signal curve.

[0051] In the embodiment of the present invention, due to the noise interference of the surrounding environment and medical equipment, the respiratory signal collected from the patient has the problem of low accuracy; and because the normal respiratory signal change characteristics have certain regular characteristics, the abnormal situation is analyzed and corrected by the change characteristics of the respiratory signal. First, an airflow sensor is placed near the patient's nose or mouth to obtain the patient's respiratory volume at each moment to form the data points of the respiratory signal curve, where the time is the horizontal axis of the respiratory signal curve and the respiratory volume is the vertical axis of the respiratory signal curve.

[0052] It should be noted that, in one embodiment of the present invention, the time interval is 0.2s; in other embodiments of the present invention, the size of the time interval can be specifically set according to the specific situation, which is not limited or elaborated here.

[0053] Step S2: According to the distribution characteristics of the respiratory volume in the neighborhood of each extreme point on the respiratory signal curve, the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points are obtained; according to the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments, the degree of noise influence between adjacent extreme points is obtained; according to the initial respiratory abnormality factor and the degree of noise influence between adjacent extreme points, the corrected respiratory abnormality factor between adjacent extreme points is obtained.

[0054] Under normal breathing conditions, the inhaled and exhaled air volumes fluctuate within a certain range. However, if affected by noise interference, the respiratory signal will appear irregular and the fluctuations in respiratory volume will be inconsistent. Extreme points usually represent important turning points on the data curve. Under stable conditions, the more stable the change trend of extreme points of the same type, the closer the respiratory volume. By analyzing the distribution characteristics of the respiratory volume in the neighborhood of the extreme points, the changing trend of the respiratory volume can be reflected. The greater the fluctuation, the greater the respiratory abnormality factor. According to the distribution characteristics of the respiratory volume in the neighborhood of each extreme point on the respiratory signal curve, the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points are obtained.

[0055] Preferably, in one embodiment of the present invention, obtaining the abnormal breathing fluctuation factor of each extreme point and the initial abnormal breathing factor between adjacent extreme points includes:

[0056] The mean of the respiratory volume in the neighborhood of each extreme point is obtained as the average respiratory volume; the standard deviation of the respiratory volume in the neighborhood of each extreme point is obtained;

[0057] The difference between the respiratory volume corresponding to the extreme point and the average respiratory volume is calculated. If the difference result is greater than three times the standard deviation, the difference result is normalized and used as the abnormal respiratory fluctuation factor of each extreme point; if the difference result is less than or equal to three times the standard deviation, the abnormal respiratory fluctuation factor of each extreme point is 0; wherein, the method for comparing the difference result and three times the standard deviation is the three times variance criterion, which is used to evaluate the abnormal distribution of data points. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0058] The mean of the abnormal breathing fluctuation factors between adjacent extreme value points is obtained and normalized as the initial abnormal breathing factor between adjacent extreme value points.

[0059] It should be noted that, in one embodiment of the present invention, The function is normalized, and the specific means are technical means well known to those skilled in the art and will not be described in detail here.

[0060] It should be noted that the extreme points on the respiratory signal curve include two categories: maximum points and minimum points. The acquisition method is: traverse the respiratory signal curve, if the respiratory volume of each data point is greater than the respiratory volume of its adjacent previous and next data points, it is marked as a maximum point; if the respiratory volume of each data point is greater than the respiratory volume of its adjacent previous and next data points, it is marked as a minimum point; in other embodiments of the present invention, Newton's method can also be used to obtain the extreme points on the respiratory signal curve. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0061] It should be noted that, in one embodiment of the present invention, the neighborhood range of each extreme point is a range centered on each extreme point and consisting of a number of adjacent extreme points of the same type, and the number of extreme points within the range is 15. In other embodiments of the present invention, the size of the neighborhood range can be set according to specific circumstances and is not limited or elaborated here.

[0062] Since the respiratory signal curve is relatively smooth, the extreme points are easily affected by noise and move slightly, resulting in changes in the correlation between the time ranges of adjacent extreme points. The signal change rate and stability are reflected by analyzing the time change characteristics between different adjacent extreme points; the local characteristics of the signal are reflected by the change trend of the data points between adjacent moments. The more drastic the change trend of the data points and the greater the fluctuation, the greater the possibility that the signal is interfered by noise; therefore, the degree of noise influence between adjacent extreme points is obtained based on the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments.

[0063] Preferably, in one embodiment of the present invention, the method for obtaining the noise impact degree can be found in Figure 2 , which shows a flow chart of a method for obtaining the degree of noise influence, including:

[0064] Step S201: Obtain monotonic curve segments between adjacent extreme value points, obtain the time length of each monotonic curve segment, and obtain the average of the time lengths between adjacent monotonic curve segments before and after each monotonic curve segment as the comparison time length.

[0065] The monotonic curve segments between adjacent extreme points show different breathing trends during the breathing process. Since the time range ratio of inhalation and exhalation in the breathing signal fluctuates within a certain range, if there is noise, it will affect the time relationship between inhalation and exhalation. By taking the average value to obtain the mean of the time length between the adjacent monotonic curve segments before and after, the average time length level can be quantified for comparison.

[0066] Step S202: obtaining the average value of the slopes of the data points between all adjacent moments on each monotonic curve segment as the first slope; obtaining the average value of the first slopes between the adjacent monotonic curve segments before and after each monotonic curve segment as the second slope.

[0067] The slope of the data points between adjacent moments can determine whether the change in the data points is larger or smaller; the general trend of the change in the data points between the adjacent curve segments before and after is quantified by taking the average value, which serves as the basis for subsequent comparisons.

[0068] It should be noted that the method for obtaining the slope is to calculate the ratio of the ordinate difference to the abscissa difference between the data points to obtain the slope. The specific means are well known to those skilled in the art and will not be elaborated here.

[0069] Step S203: If the monotonic curve segment between adjacent extreme points is monotonically increasing, obtain the length difference between the preset multiple time length and the comparison time length, obtain the slope difference between the first slope and the preset multiple second slope; if the monotonic curve segment between adjacent extreme points is monotonically decreasing, obtain the length difference between the time length and the preset multiple comparison time length, obtain the slope difference between the preset multiple first slope and the second slope.

[0070] Since the air pressure in the chest cavity is lower than the external atmospheric pressure during inhalation in the breathing process, the external air will flow into the lungs through the respiratory tract, and the respiratory volume will continue to increase; during exhalation, the chest cavity shrinks, and the air pressure in the lungs is higher than the external atmospheric pressure, so the air in the lungs is expelled through breathing, and the respiratory volume gradually decreases; therefore, if the curve segment is monotonically increasing, that is, it rises from the minimum point to the maximum point, the respiratory volume continues to increase, corresponding to the inhalation process, and the adjacent monotonic curve segments are the exhalation process; if the curve segment is monotonically decreasing, that is, it drops from the maximum point to the minimum point, the respiratory volume gradually decreases, corresponding to the exhalation process, and the adjacent monotonic curve segments are the inhalation process.

[0071] It should be noted that, under normal circumstances, since inhalation is relatively short and exhalation is relatively long during breathing, the time length ratio of inhalation and exhalation is between 1:1 and 1:2. In one embodiment of the present invention, in order to express the average ratio level, the time length ratio of inhalation and exhalation is set to 1:1.5, the preset multiple is 1.5, and the time length of the monotonically increasing curve segment is 1.5 times the time length of the monotonically decreasing curve segment; in other embodiments of the present invention, the preset multiple can be set according to the specific situation, which is not limited or elaborated here.

[0072] Step S204: Calculate the product of the length difference and the slope difference, and perform normalization to obtain the degree of noise influence between adjacent extreme value points.

[0073] In one embodiment of the present invention, the formula for the noise impact degree is expressed as:

[0074] ;

[0075] ;in, Indicates The extreme point and The degree of noise influence of the monotonic curve segment increasing between the extreme points; Indicates The extreme point and The noise influence degree of the monotonic curve segment decreasing between the extreme points; Indicates The extreme point and The time length of the monotonic curve segment between the extreme points; Indicates The extreme point and The time length of the monotonic curve segment between the extreme points; Indicates The extreme point and The time length of the monotonic curve segment between the extreme points; Indicates The extreme point and The first slope of the monotonic curve segment between the extreme points; Indicates The extreme point and The first slope of the monotonic curve segment between the extreme points; Indicates The extreme point and The first slope of the monotonic curve segment between the extreme points; represents the normalization function; Indicates the preset multiple.

[0076] In the formula for the degree of noise impact, It means calculating the mean of the time lengths between adjacent monotonic curve segments before and after the monotonic curve segment, that is, comparing the time lengths; It represents the average of the first slopes of adjacent monotonic curve segments before and after the calculated monotonic curve segment, that is, the second slope; the smaller the length difference and the smaller the slope difference, the more stable the time range experienced by different breathing processes, the more consistent the change amplitude of data points, and the less affected by noise; the larger the length difference and the larger the slope difference, the more fluctuating the time range experienced by different breathing processes, the more inconsistent the change amplitude of data points, and the greater the impact of noise.

[0077] The initial abnormal breathing factor reflects the degree of influence of noise interference on the respiratory signal. The larger the initial abnormal breathing factor, the greater the fluctuation of the respiratory volume between adjacent extreme points, and the more abnormal the breathing. The noise influence degree can understand the interference of noise on the respiratory signal. The greater the noise influence, the greater the interference to breathing, and the greater the possibility of abnormal breathing. The initial abnormal breathing factor and the noise influence degree are combined to more comprehensively evaluate the patient's respiratory condition. Therefore, according to the initial abnormal breathing factor between adjacent extreme points and the noise influence degree, the modified abnormal breathing factor between adjacent extreme points is obtained.

[0078] Preferably, in one embodiment of the present invention, the method for obtaining the corrected abnormal breathing factor includes:

[0079] The product of the initial abnormal breathing factor and the noise influence degree between adjacent extreme value points is obtained as the corrected abnormal breathing factor between adjacent extreme value points.

[0080] It should be noted that in other embodiments of the present invention, the initial abnormal breathing factor and the degree of noise influence may be added to construct a positive correlation relationship in which the larger the initial abnormal breathing factor, the greater the degree of noise influence, and the larger the corrected abnormal breathing factor. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0081] Step S3: For extreme value points of the same type, the possibility of emotional mutation of each extreme value point of the same type is obtained according to the change trend between adjacent extreme value points of the same type and the corresponding respiratory abnormality fluctuation factor; the trend characteristic factor of each extreme value point of the same type is obtained according to the distribution of emotional mutation possibility of extreme value points of the same type and the fluctuation characteristics between extreme value points of the same type; the data credibility between adjacent extreme value points is obtained according to the difference in trend characteristic factors between adjacent extreme value points and the corrected respiratory abnormality factor.

[0082] When the patient is nervous or excited, the respiratory rate may rise sharply in a short period of time, and then gradually decrease due to the body's adaptation or relaxation after a period of time. The larger the abnormal breathing fluctuation factor, the smaller and smoother the waveform amplitude of the patient's respiratory signal curve is when the patient is at normal rest. When the patient is excited, the greater the change in waveform amplitude. Therefore, for the same extreme point, the possibility of emotional mutation of each extreme point of the same type is obtained according to the change trend between adjacent extreme points of the same type and the corresponding abnormal breathing fluctuation factor.

[0083] Preferably, in one embodiment of the present invention, the method for obtaining the possibility of sudden change of emotion includes:

[0084] For extreme value points of the same type, the difference in slope between each extreme value point of the same type and the adjacent extreme value points of the same type before and after is obtained as the first slope difference;

[0085] The product of the first slope difference and the corresponding respiratory abnormality fluctuation factor is calculated as the possibility of emotional mutation at each similar extreme point.

[0086] In one embodiment of the present invention, the formula for the possibility of sudden change in emotion is expressed as:

[0087] ;

[0088] in, Indicates The possibility of emotional mutation at similar extreme points; Indicates The same extreme point and The slope between similar extreme points; Indicates The same extreme point and The slope between similar extreme points; Indicates The abnormal respiratory fluctuation factor corresponding to the same extreme point.

[0089] In the formula for the probability of a sudden change in mood, Indicates The same extreme point and The slope between the same extreme points and the The same extreme point and The difference in the slopes between the same extreme points, that is, the first slope difference, the larger the difference, the greater the change amplitude between the same extreme points, and the more likely it is to correspond to a sudden change in emotion; the larger the abnormal breathing fluctuation factor, the greater the possibility of a sudden change in emotion.

[0090] When emotions are excited, the respiratory signal curve may show mutation characteristics, which usually contain important characteristic information. By analyzing the possibility distribution of emotion mutations at similar extreme points, we can capture the points with more significant emotion mutations, and then analyze the changes in the overall emotion trend; the fluctuation characteristics between similar extreme points can reflect the stability and change trend of the emotional state within a certain period of time, which is conducive to analyzing the abnormal fluctuations of extreme points; therefore, according to the possibility distribution of emotion mutations at similar extreme points and the fluctuation characteristics between similar extreme points, the trend characteristic factor of each similar extreme point is obtained.

[0091] Preferably, in one embodiment of the present invention, the method for obtaining the trend characteristic factor can be found in Figure 3 , which shows a flow chart of a method for obtaining a trend characteristic factor, including:

[0092] Step S301: clustering all extreme value points of the same type according to the possibility of emotional mutation of extreme value points of the same type to obtain multiple clusters of the same type.

[0093] By clustering similar extreme points with similar emotional mutation possibilities, it is helpful to identify abnormal similar extreme points.

[0094] It should be noted that, in the embodiment of the present invention, the K-means algorithm is used for clustering. Since the emotional state is generally divided into emotional stability and emotional excitement, the value of K is set to 2 to obtain two similar clusters. The specific K-means algorithm is a technical means well known to those skilled in the art and will not be described in detail here.

[0095] Step S302: Obtain the mean value of the emotional mutation probability of all similar extreme value points in each similar cluster as the overall emotional mutation probability; select the extreme value point in the corresponding similar cluster with the largest overall emotional mutation probability value as the emotional mutation point.

[0096] The possibility of emotional mutation in each cluster is quantified by taking the average value, showing the average level of emotional mutation in clusters of the same type.

[0097] The greater the possibility of overall mood changes, the more likely it is that the mood will change most dramatically, reflecting the characteristics of the patient's mood changes.

[0098] Step S303: Obtain the absolute value of the slope between adjacent emotion mutation points. If the absolute value of the slope is greater than or equal to the preset slope threshold, obtain the absolute value of the accumulated slopes between all adjacent extreme value points of the same type between adjacent emotion mutation points, and normalize them to obtain the trend characteristic factors corresponding to the extreme value points of the same type between adjacent emotion mutation points.

[0099] The absolute value of the slope between adjacent emotion mutation points reflects the changing trend between adjacent emotion mutation points. The larger the absolute value of the slope, the greater the changing trend between emotion mutation points. By analyzing the absolute value of the accumulated slope between adjacent extreme points of the same type, it can be reflected whether the changing trends between adjacent extreme points of the same type are consistent. If the absolute value of the accumulated slope is larger, the changing trend is more consistent, the overall trend may be an upward trend or a downward trend, the more likely it is a normal situation caused by emotion changes, the more obvious and continuous the trend is, and the larger the trend characteristic factor is.

[0100] It should be noted that, in one embodiment of the present invention, the preset slope threshold is 0.5; in other embodiments of the present invention, the preset slope threshold may be set according to specific circumstances, which will not be limited or elaborated herein.

[0101] Step S304: If the absolute value of the slope is less than the preset slope threshold, the average of the abnormal breathing fluctuation factors of all similar extreme value points between adjacent emotional mutation points is obtained, and negative correlation mapping is performed to obtain the trend characteristic factors of the corresponding similar extreme value points between adjacent emotional mutation points.

[0102] The smaller the absolute value of the slope, the smaller the change trend between the emotion mutation points. If the respiratory abnormality fluctuation factor of the same extreme point is smaller, the data change is more normal, and the trend characteristic factor is larger.

[0103] In one embodiment of the present invention, for a type of extreme value point, the formula of the trend characteristic factor is expressed as:

[0104] ;

[0105] in, Indicates Emotional mutation point and The trend characteristic factor of the same type of extreme value points between the emotional mutation points; Indicates Emotional mutation point and The number of similar extreme value points between emotion mutation points; Indicates Abnormal respiratory fluctuation factors corresponding to the same extreme value points; Indicates Emotional mutation point and The slope between the emotional mutation points; Indicates Emotional mutation point and The slope between the emotional mutation points; represents an exponential function with a natural constant as base; represents the normalization function; Indicates taking the absolute value; Indicates the preset slope threshold.

[0106] In the formula of the trend characteristic factor, if the absolute value of the slope is less than the preset abnormal threshold, it means that the change between the emotional mutation points is more stable, and the exponential function with the natural constant as the base is used to convert Perform negative correlation mapping, It means calculating the mean of the abnormal breathing fluctuation factors of all similar extreme value points between adjacent emotional mutation points. The larger the mean, the larger the abnormal breathing fluctuation factors of similar extreme value points, the greater the influence of noise interference, the more abnormal the breathing state, and the smaller the trend characteristic factor; if the absolute value of the slope is greater than or equal to the preset abnormal threshold, it means that the changes between emotional mutation points are more volatile, It means calculating the absolute value of the accumulated slopes between all adjacent extreme points of the same type between adjacent emotion mutation points. The larger the absolute value of the accumulated slopes, the more it changes toward a trend, and the more likely it is that the trend will occur when emotions are excited, and the larger the trend characteristic factor.

[0107] The difference in trend characteristic factors between adjacent extreme points reflects the stability and continuity of data changes. The smaller the difference in trend characteristic factors, the more consistent the changes in adjacent extreme points and the more real the data. The larger the difference in trend characteristic factors, the more inconsistent the changes in adjacent extreme points, the more the data is affected by noise and the more abnormal it is. The corrected breathing abnormality factor reflects the fluctuations caused by noise during the breathing process. The larger the corrected breathing abnormality factor, the lower the data credibility. The data credibility between adjacent extreme points is obtained based on the difference in trend characteristic factors between adjacent extreme points and the corrected breathing abnormality factor.

[0108] Preferably, in one embodiment of the present invention, the method for obtaining data credibility includes:

[0109] The difference in trend characteristic factors between adjacent extreme value points is obtained as the factor difference; the product of the factor difference and the corrected respiratory abnormality factor is calculated, and negative correlation normalization mapping is performed to obtain the data credibility between adjacent extreme value points.

[0110] In one embodiment of the present invention, the formula for data credibility is expressed as:

[0111] ;

[0112] in, Indicates the data credibility between adjacent extreme points; Indicates the modified respiratory abnormality factor between adjacent extreme points; Indicates Trend characteristic factors of extreme points; Indicates Trend characteristic factors of extreme points; Represents an exponential function with a natural constant as its base.

[0113] In the formula for data credibility, the exponential function with the natural constant as the base will be Perform negative correlation mapping, It means calculating the product of the difference in trend characteristic factors between adjacent extreme points and the corrected respiratory abnormality factor. The larger the product, the greater the difference in trend characteristic factors, the greater the difference in changes between adjacent extreme points, the larger the corrected respiratory abnormality factor, the greater the possibility of abnormality of the data between extreme points, and the lower the data credibility.

[0114] It should be noted that, in one embodiment of the present invention, by right Negative correlation mapping is performed, and the greater the difference in the trend characteristic factor, the greater the corrected respiratory abnormality factor, and the smaller the negative correlation relationship of the data credibility; in other embodiments of the present invention, The reciprocal is taken to realize negative correlation, wherein, in order to avoid the denominator of the formula being 0 and the formula being meaningless, an artificially set threshold, such as 0.01, needs to be added; the specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0115] Step S4: According to the data credibility between the adjacent extreme value points of each data point on the respiratory signal curve, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve.

[0116] By analyzing the data credibility between adjacent extreme points, reducing the characteristics of data points with low data credibility, providing high overall data accuracy, and more accurately reflecting the patient's true breathing state after filtering. Therefore, according to the data credibility between the adjacent extreme points where each data point on the respiratory signal curve is located, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve.

[0117] Preferably, in one embodiment of the present invention, the method for obtaining the filtered respiratory volume includes:

[0118] A sliding window is constructed with each data point as the center, and the time length between each data point and the center point in the sliding window is obtained, and a negative correlation mapping is performed as the weight factor of each data point relative to the center point; the product of the weight factor and the corresponding data credibility is calculated and normalized as the weight value of each data point relative to the center point;

[0119] Based on the weight value of each data point relative to the center point, the respiratory volume of the center point is filtered by a weighted filtering algorithm to obtain a filtered respiratory volume.

[0120] It should be noted that, in one embodiment of the present invention, the length of the sliding window is 9; in other embodiments of the present invention, the length of the sliding window can be set according to specific circumstances, which is not limited or elaborated here.

[0121] It should be noted that the process of the weighted filtering algorithm is: for each sliding window, the product of the weight value of each data point in the window relative to the center point and the respiratory volume of the corresponding data point is calculated as the first product; the ratio of the cumulative value of the first product corresponding to all data points in the window and the cumulative value of the weight value is calculated as the filtered respiratory volume of the center point in the corresponding window, the filtered respiratory volume of each data point is obtained, the respiratory signal curve is updated, the influence of noise is reduced, and the accuracy of respiratory signal monitoring is improved. The specific weighted filtering algorithm is a technical means well known to those skilled in the art and will not be elaborated here.

[0122] In summary, the present invention analyzes the distribution characteristics of extreme points on the respiratory signal curve to obtain the initial respiratory abnormality factor between adjacent extreme points; according to the time change characteristics between different adjacent extreme points and the change trend of data points between adjacent moments, the noise influence degree between adjacent extreme points is obtained, and the modified respiratory abnormality factor between adjacent extreme points is obtained; according to the change trend between adjacent extreme points of the same type and the corresponding respiratory abnormality fluctuation factor, the trend characteristic factor of each extreme point of the same type is obtained; and then the data credibility between adjacent extreme points is obtained; the respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve. The present invention takes into account noise interference and improves the accuracy of respiratory monitoring by obtaining relatively accurate respiratory signal data of the patient.

[0123] Based on the same application concept as the method for monitoring the respiratory signal of a patient provided in the embodiment of the present application, the present embodiment also proposes a device for monitoring the respiratory signal of a patient, such as Figure 4 As shown, it includes: a data acquisition module 401, a respiratory abnormality factor correction module 402, a data credibility determination module 403, and a respiratory curve optimization module 404:

[0124] Data acquisition module 401: acquires the patient's respiratory volume at each moment to form data points of a respiratory signal curve;

[0125] The respiratory abnormality factor correction module 402: obtains the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points according to the distribution characteristics of the respiratory volume in the neighborhood range of each extreme point on the respiratory signal curve; obtains the noise influence degree between adjacent extreme points according to the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments; obtains the corrected respiratory abnormality factor between adjacent extreme points according to the initial respiratory abnormality factor and the noise influence degree between adjacent extreme points;

[0126] Data credibility determination module 403: for similar extreme value points, according to the change trend between adjacent similar extreme value points and the corresponding respiratory abnormality fluctuation factor, the possibility of emotional mutation of each similar extreme value point is obtained; according to the distribution of emotional mutation possibility of similar extreme value points and the fluctuation characteristics between similar extreme value points, the trend characteristic factor of each similar extreme value point is obtained; according to the difference of trend characteristic factors between adjacent extreme value points and the modified respiratory abnormality factor, the data credibility between adjacent extreme value points is obtained;

[0127] Respiratory curve optimization module 404: according to the data credibility between the adjacent extreme value points of each data point on the respiratory signal curve, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve.

[0128] It should be understood that the patient's respiratory signal monitoring device provided in this embodiment is used to execute the above-mentioned patient's respiratory signal monitoring method, and therefore has the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0129] The present invention also provides a patient's respiratory signal monitoring system, wherein the system stores programs or instructions, and when the programs or instructions are executed by a processor, the steps of the patient's respiratory signal monitoring method as described above are implemented.

[0130] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0131] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for monitoring a patient's respiratory signal, characterized in that: The method comprises: Obtain the patient's respiratory volume at each moment to form data points of the respiratory signal curve; According to the distribution characteristics of the respiratory volume in the neighborhood of each extreme point on the respiratory signal curve, the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points are obtained; according to the time change characteristics between different adjacent extreme points and the change trend of the data points between adjacent moments, the noise influence degree between adjacent extreme points is obtained; according to the initial respiratory abnormality factor and the noise influence degree between adjacent extreme points, the modified respiratory abnormality factor between adjacent extreme points is obtained; For extreme value points of the same type, the possibility of emotional mutation of each extreme value point of the same type is obtained according to the change trend between adjacent extreme value points of the same type and the corresponding respiratory abnormality fluctuation factor; the trend characteristic factor of each extreme value point of the same type is obtained according to the distribution of emotional mutation possibility of the extreme value points of the same type and the fluctuation characteristics between the extreme value points of the same type; the data credibility between adjacent extreme value points is obtained according to the difference in trend characteristic factors between adjacent extreme value points and the corrected respiratory abnormality factor; According to the data credibility between the adjacent extreme value points of each data point on the respiratory signal curve, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve; The step of obtaining the abnormal breathing fluctuation factor of each extreme point and the initial abnormal breathing factor between adjacent extreme points includes: The mean of the respiratory volume in the neighborhood of each extreme point is obtained as the average respiratory volume; the standard deviation of the respiratory volume in the neighborhood of each extreme point is obtained; Calculate the difference between the respiratory volume at the corresponding extreme point and the average respiratory volume. If the difference is greater than three times the standard deviation, normalize the difference and use it as the abnormal respiratory fluctuation factor for each extreme point. If the difference is less than or equal to three times the standard deviation, the abnormal respiratory fluctuation factor for each extreme point is 0. The mean of the abnormal breathing fluctuation factors between adjacent extreme value points is obtained and normalized as the initial abnormal breathing factor between adjacent extreme value points.

2. The method for monitoring a patient's respiratory signal according to claim 1, characterized in that: The method for obtaining the noise impact degree includes: Obtain the monotonic curve segments between adjacent extreme value points, obtain the time length of each monotonic curve segment, and obtain the average of the time lengths between adjacent monotonic curve segments before and after each monotonic curve segment as the comparison time length; obtain the average of the slopes of the data points between all adjacent moments on each monotonic curve segment as the first slope; obtain the average of the first slopes between adjacent monotonic curve segments before and after each monotonic curve segment as the second slope; If the monotonic curve segment between adjacent extreme value points is monotonically increasing, the length difference between the preset multiple of the time length and the comparison time length is obtained, and the slope difference between the first slope and the preset multiple of the second slope is obtained; if the monotonic curve segment between adjacent extreme value points is monotonically decreasing, the length difference between the time length and the preset multiple of the comparison time length is obtained, and the slope difference between the preset multiple of the first slope and the second slope is obtained; The product of the length difference and the slope difference is calculated and normalized to obtain the degree of noise influence between adjacent extreme points.

3. The method for monitoring a patient's respiratory signal according to claim 1, characterized in that: The method for obtaining the corrected abnormal breathing factor includes: The product of the initial abnormal breathing factor and the noise influence degree between adjacent extreme value points is obtained as the corrected abnormal breathing factor between adjacent extreme value points.

4. The method for monitoring a patient's respiratory signal according to claim 1, characterized in that: The method for obtaining the possibility of sudden change of emotion includes: For extreme value points of the same type, the difference in slope between each extreme value point of the same type and the adjacent extreme value points of the same type before and after is obtained as the first slope difference; The product of the first slope difference and the corresponding respiratory abnormality fluctuation factor is calculated as the possibility of emotional mutation at each similar extreme point.

5. The method for monitoring a patient's respiratory signal according to claim 4, characterized in that: The method for obtaining the trend characteristic factor includes: According to the possibility of emotional mutation of similar extreme value points, all similar extreme value points are clustered to obtain multiple similar clusters; Obtain the average of the likelihood of emotional mutation of all similar extreme value points in each similar cluster as the overall emotional mutation likelihood; select the extreme value point in the corresponding similar cluster with the largest overall emotional mutation likelihood as the emotional mutation point; Obtain the absolute value of the slope between adjacent emotion mutation points. If the absolute value of the slope is greater than or equal to the preset slope threshold, obtain the absolute value of the accumulated slopes between all adjacent extreme value points of the same type between adjacent emotion mutation points, and perform normalization to obtain the trend characteristic factor of each extreme value point of the same type between adjacent emotion mutation points. If the absolute value of the slope is less than the preset slope threshold, the mean of the abnormal breathing fluctuation factors of all similar extreme points between adjacent emotional mutation points is obtained, and negative correlation mapping is performed to obtain the trend characteristic factor of each similar extreme point between adjacent emotional mutation points.

6. The method for monitoring a patient's respiratory signal according to claim 1, characterized in that: The method for obtaining the data credibility includes: The difference in trend characteristic factors between adjacent extreme value points is obtained as the factor difference; the product of the factor difference and the corrected respiratory abnormality factor is calculated, and negative correlation normalization mapping is performed to obtain the data credibility between adjacent extreme value points.

7. The method for monitoring a patient's respiratory signal according to claim 1, characterized in that: The method for obtaining the filtered respiratory volume comprises: A sliding window is constructed with each data point as the center, and the time length between each data point and the center point in the sliding window is obtained, and a negative correlation mapping is performed as the weight factor of each data point relative to the center point; the product of the weight factor and the corresponding data credibility is calculated and normalized as the weight value of each data point relative to the center point; Based on the weight value of each data point relative to the center point, the respiratory volume of the center point is filtered by a weighted filtering algorithm to obtain a filtered respiratory volume.

8. A patient's respiratory signal monitoring device, characterized in that: It includes data acquisition module, respiratory abnormality factor correction module, data credibility determination module and respiratory curve optimization module: Data acquisition module: obtains the patient's respiratory volume at each moment and forms data points of the respiratory signal curve; Respiratory abnormality factor correction module: according to the distribution characteristics of the respiratory volume in the neighborhood of each extreme point on the respiratory signal curve, obtain the respiratory abnormality fluctuation factor of each extreme point and the initial respiratory abnormality factor between adjacent extreme points; According to the time variation characteristics between different adjacent extreme value points and the variation trend of data points between adjacent moments, the noise influence degree between adjacent extreme value points is obtained; According to the initial abnormal breathing factor between adjacent extreme value points and the noise influence degree, obtaining a modified abnormal breathing factor between adjacent extreme value points; The step of obtaining the abnormal breathing fluctuation factor of each extreme point and the initial abnormal breathing factor between adjacent extreme points includes: The mean of the respiratory volume in the neighborhood of each extreme point is obtained as the average respiratory volume; the standard deviation of the respiratory volume in the neighborhood of each extreme point is obtained; Calculate the difference between the respiratory volume at the corresponding extreme point and the average respiratory volume. If the difference is greater than three times the standard deviation, normalize the difference and use it as the abnormal respiratory fluctuation factor for each extreme point. If the difference is less than or equal to three times the standard deviation, the abnormal respiratory fluctuation factor for each extreme point is 0. The average of the abnormal breathing fluctuation factors between adjacent extreme value points is obtained and normalized to serve as the initial abnormal breathing factor between adjacent extreme value points; Data credibility determination module: For extreme value points of the same type, the possibility of emotional mutation of each extreme value point of the same type is obtained according to the change trend between adjacent extreme value points of the same type and the corresponding respiratory abnormality fluctuation factor; the trend characteristic factor of each extreme value point of the same type is obtained according to the distribution of emotional mutation possibility of extreme value points of the same type and the fluctuation characteristics between extreme value points of the same type; the data credibility between adjacent extreme value points is obtained according to the difference in trend characteristic factors between adjacent extreme value points and the correction of respiratory abnormality factors; Respiratory curve optimization module: According to the data credibility between the adjacent extreme value points of each data point on the respiratory signal curve, the corresponding respiratory volume is filtered to obtain the filtered respiratory volume and the filtered respiratory curve.

9. A patient's respiratory signal monitoring system, characterized in that: The system stores programs or instructions, and when the programs or instructions are executed by the processor, the steps of a method for monitoring a patient's respiratory signal as described in any one of claims 1 to 7 are implemented.

Images