A clinical blood drug concentration detection method

Abstract

The present invention relates to the technical field of detecting blood drug concentration using optical means, and specifically to a clinical blood drug concentration detection method. The method obtains the delayed response of spectral data to drug concentration and the complexity of changes in bands other than a preset characteristic band; further obtains the credibility of the interference change of drug concentration; obtains the spectral shape distortion of the spectral curve corresponding to each drug concentration within the preset characteristic band based on the morphological characteristics of the spectral curve corresponding to each drug concentration within the preset characteristic band, and the distribution difference characteristics of the data points between the corresponding spectral curve and the preset spectral curve within the preset characteristic band; obtains the processing benefit confidence of the spectral data; filters the spectral data to obtain filtered spectral data; and detects blood drug concentration. The present invention improves the quality of spectral data and the detection accuracy of blood drug concentration by removing interfering components in spectral detection.

Description

Technical Field

[0001] The present invention relates to the technical field of detecting blood drug concentration using optical means, and in particular to a clinical blood drug concentration detection method. Background Art

[0002] In clinical medicine, blood concentration refers to the concentration level of a certain drug in the blood. It is an important indicator for evaluating drug metabolism and efficacy. Too low drug concentration may lead to poor efficacy, while too high drug concentration may cause side effects. The detection of blood drug concentration is helpful to determine the absorption, distribution, metabolism and excretion of drugs in the patient's body, so as to adjust the drug dosage to achieve better therapeutic effects and reduce side effects.

[0003] In the existing technology, near-infrared spectroscopy is used for intelligent detection of blood drug concentration. It has the ability to quickly collect and analyze samples and can monitor the concentration changes of drugs in the body in real time. However, during the detection process, light source fluctuations, instrument vibrations, and temperature changes will cause the intensity of the light signal to change in the optical detection results, introducing a certain degree of interference components. At the same time, the detector of the near-infrared spectroscopy instrument itself may also have electronic interference or amplifier interference. Because the characteristics of various interfering components are not taken into account in the existing filtering, the filtering effect is poor, which causes errors in the analysis of the changing trend of drug concentration in the blood, affecting the detection results of blood drug concentration. Summary of the Invention

[0004] In order to solve the technical problem that the filtering effect of interfering components is poor and affects the detection results of blood drug concentration, the purpose of the present invention is to provide a clinical blood drug concentration detection method, and the technical solution adopted is as follows:

[0005] The present invention proposes a clinical blood drug concentration detection method, which comprises:

[0006] Obtaining spectral data of blood samples at different drug concentrations; the spectral data includes absorption intensities corresponding to different wavelengths, constituting data points of a spectral curve;

[0007] Obtaining a drug concentration fitting curve and a maximum absorption intensity fitting curve of a drug concentration corresponding spectral curve within a preset characteristic band; obtaining a delayed response of the spectral data to the drug concentration based on the correlation coefficient and difference characteristics of the corresponding data between the two fitting curves; obtaining a change complexity of other bands outside the preset characteristic band based on the fluctuation characteristics of data points in other bands outside the preset characteristic band between the spectral curves corresponding to different drug concentrations; and obtaining a credibility of the interference change of the drug concentration based on the delayed response and the change complexity;

[0008] Obtaining the spectral shape distortion of the spectral curve corresponding to each drug concentration within the preset characteristic band based on the morphological characteristics of the spectral curve corresponding to each drug concentration within the preset characteristic band and the distribution difference characteristics of data points between the corresponding spectral curve and the preset spectral curve within the preset characteristic band; obtaining the processing benefit confidence of the spectral data based on the interference change credibility and the spectral shape distortion;

[0009] The spectral data is filtered according to the processing benefit confidence to obtain filtered spectral data; and the drug concentration is detected according to the filtered spectral data.

[0010] Furthermore, the method for obtaining the delayed responsiveness includes:

[0011] The delayed response of the spectral data to the drug concentration was obtained based on the correlation coefficient and mean square error of the corresponding data between the two fitting curves. The correlation coefficient was negatively correlated with the delayed response, while the mean square error was positively correlated with the delayed response.

[0012] Furthermore, the method for obtaining the complexity of the change includes:

[0013] The fluctuation degree differences of data points in bands other than the preset characteristic band between the spectral curves corresponding to different drug concentrations were calculated, and the fluctuation degree differences between the spectral curves corresponding to all drug concentrations were averaged to obtain the change complexity of bands other than the characteristic band.

[0014] Furthermore, the method for obtaining the change credibility includes:

[0015] The product of delayed response and change complexity is calculated to obtain the credibility of the disturbed change.

[0016] Furthermore, the method for obtaining the distribution difference feature includes:

[0017] Calculating the standard deviation of all data points of the spectral curve in the preset characteristic band, calculating the product of the standard deviation and the amplitude of each data point of the spectral curve in the preset characteristic band to obtain a weighted amplitude;

[0018] The difference in weighted amplitudes of all data points corresponding to the spectral curve corresponding to each drug concentration and the preset spectral curve in the preset characteristic band is calculated to obtain the distribution difference characteristics.

[0019] Furthermore, the method for obtaining the spectral shape distortion includes:

[0020] For the spectral curve corresponding to each drug concentration, the slopes between all adjacent data points in the preset characteristic band are calculated, and the mean difference between all adjacent slopes is calculated as the spectral shape mutation;

[0021] The spectral shape distortion is obtained based on the spectral shape mutation, the number of breakpoints of the spectral curve in the preset characteristic band, and the distribution difference characteristics. The spectral shape mutation and the number of breakpoints of the spectral curve in the preset characteristic band are positively correlated with the spectral shape distortion, while the distribution difference characteristics are negatively correlated with the spectral shape distortion.

[0022] Furthermore, the method for obtaining the processing benefit confidence includes:

[0023] Calculate the mean value of spectral shape distortion corresponding to all drug concentrations as the distortion mean;

[0024] The confidence level of treatment benefit was obtained based on the disturbed change credibility and the distortion mean. Both the disturbed change credibility and the distortion mean were positively correlated with the treatment benefit confidence level.

[0025] Furthermore, the method for acquiring filtered spectral data includes:

[0026] The processing benefit confidence is used as the standard deviation parameter in the filter to filter the spectral data to obtain filtered spectral data.

[0027] Furthermore, the method for obtaining the preset spectral curve includes:

[0028] Spectral data of the blood sample without drug concentration is obtained to form a corresponding preset spectral curve.

[0029] Furthermore, the correlation coefficient is a Pearson correlation coefficient.

[0030] The present invention has the following beneficial effects:

[0031] The present invention obtains a drug concentration fitting curve and a maximum absorption intensity fitting curve of a drug concentration corresponding spectral curve within a preset characteristic band. The fitting curve can more intuitively show the relationship between drug concentration and spectral data; based on the correlation coefficient and difference characteristics of the corresponding data between the two fitting curves, the delayed response of the spectral data to the drug concentration is obtained, which is helpful to understand the response speed of the spectral data to the change of drug concentration; considering that the spectral data in other bands may also contain information about the drug concentration, but may be affected by more interference factors, based on the fluctuation characteristics of the data points in other bands outside the preset characteristic band between the spectral curves corresponding to different drug concentrations, the change complexity of other bands outside the preset characteristic band is obtained; based on the delayed response and the change complexity, the credibility of the interfered change of the drug concentration is obtained. Taking these two factors into consideration, a more comprehensive credibility assessment can be obtained, which is helpful to understand the degree of interference that the spectral data may suffer when detecting drug concentration; based on each The morphological characteristics of the drug concentration-corresponding spectral curve within a preset characteristic band, as well as the distribution difference characteristics of data points between the corresponding spectral curve and the preset spectral curve within the preset characteristic band, are used to obtain the spectral shape distortion of the spectral curve corresponding to each drug concentration within the preset characteristic band. This helps to understand the changes in the morphology and distribution of spectral data and more accurately judge the impact of different drug concentrations on the spectrum. The processing benefit confidence of the spectral data is obtained based on the interference change credibility and the spectral shape distortion. This comprehensively considers the impact of drug concentration on the spectrum and the impact of interference factors, and is an important indicator for evaluating the reliability of spectral data in detecting drug concentration. The spectral data is filtered according to the processing benefit confidence to obtain filtered spectral data, ensuring that useful information in the original data is retained to the maximum extent while filtering out interference, more accurately removing noise and interference factors in the spectral data, and improving the signal-to-noise ratio and reliability of the data. The blood drug concentration is detected based on the filtered spectral data. The present invention improves the quality of spectral data and the detection accuracy of blood drug concentration by removing interference components in spectral detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. 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 any creative work.

[0033] Figure 1 A flow chart of a clinical blood drug concentration detection method provided by one embodiment of the present invention;

[0034] Figure 2 A schematic diagram of a spectrum curve provided by one embodiment of the present invention;

[0035] Figure 3 A schematic diagram of a filtered spectral curve provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0036] To further illustrate the technical means and efficacy employed by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a clinical blood drug concentration detection method according to the present invention, including its specific implementation, structure, features, and efficacy. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0037] 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.

[0038] The following describes in detail a specific scheme of a clinical blood drug concentration detection method provided by the present invention with reference to the accompanying drawings.

[0039] See also Figure 1 , which shows a flow chart of a clinical blood drug concentration detection method provided by one embodiment of the present invention, the specific method includes:

[0040] Step S1: Acquire spectral data of a blood sample at different drug concentrations; the spectral data includes absorption intensities corresponding to different wavelengths, constituting data points of a spectral curve.

[0041] In the embodiments of the present invention, since drugs have a relatively narrow therapeutic concentration range, excessive or insufficient concentrations will lead to treatment failure or serious side effects. Therefore, it is necessary to perform intelligent detection of the drug's blood concentration; in order to ensure the accuracy of the test results, the same sample is tested independently multiple times, the blood sample is placed in the sample chamber of the near-infrared spectrometer, and appropriate parameters such as scanning range, resolution, number of sampling, etc. are set according to existing empirical knowledge; then the instrument is started to collect spectral data of the blood sample at different drug concentrations; the spectral data includes the absorption intensity corresponding to different wavelengths, which constitute the data points of the spectral curve. It should be noted that the horizontal axis of the spectral curve represents the wavelength, and the vertical axis represents the absorption intensity corresponding to the wavelength; as Figure 2 , which shows a schematic diagram of a spectrum curve.

[0042] It should be noted that in clinical pharmacy, the drugs used to detect drug concentrations in the blood may include various types of drugs, which may be antibiotics, anti-epileptic drugs, immunosuppressants, chemotherapy drugs, analgesics, etc.; in one embodiment of the present invention, penicillin among the anti-interference drugs is mainly used as the object of blood drug concentration detection.

[0043] Step S2: Obtain the maximum absorption intensity fitting curve of the drug concentration fitting curve and the drug concentration corresponding spectral curve within the preset characteristic band; obtain the delayed response of the spectral data to the drug concentration based on the correlation coefficient and difference characteristics of the corresponding data between the two fitting curves; obtain the change complexity of other bands outside the preset characteristic band based on the fluctuation characteristics of the data points in other bands outside the preset characteristic band between the spectral curves corresponding to different drug concentrations; obtain the credibility of the disturbed change of drug concentration based on the delayed response and the change complexity.

[0044] Drug concentration fitting curves can help understand how drug concentrations in vivo change over time or dose, which is important for drug efficacy assessment, dose adjustment, and pharmacokinetic studies. In a spectral curve, the maximum absorption intensity typically corresponds to the strongest absorption peak of a substance at a specific wavelength. This peak not only allows for more accurate drug identification but also can be used for quantitative drug analysis. Fitting a curve to the maximum absorption intensity of a drug within a preset characteristic band facilitates more accurate prediction of the drug's absorption intensity at different concentrations. Obtain a drug concentration fitting curve and a maximum absorption intensity fitting curve for the drug concentration-corresponding spectral curve within a preset characteristic band.

[0045] It should be noted that in one embodiment of the present invention, the least squares method can be used to fit the drug concentrations and the maximum absorption intensity of the drug concentration-corresponding spectral curve within a preset characteristic band according to the order in which the drug concentrations were obtained as the horizontal axis, thereby obtaining a drug concentration fitting curve and a maximum absorption intensity fitting curve. The fitted data is then standardized to eliminate the dimension of the data of the two fitting curves, thereby more intuitively displaying the differences and similarities between the fitting curves. In other embodiments of the present invention, existing fitting methods such as polynomial fitting and spline interpolation can also be used for fitting. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0046] It should be noted that the preset characteristic band is a band region where the specific absorption or emission characteristics of the drug molecules show obvious concentration changes. The implementer can obtain it in advance based on existing experience and knowledge. In an embodiment of the present invention, the preset characteristic band is 700nm-1200nm.

[0047] The correlation coefficient is a statistic that measures the strength and direction of the linear relationship between two variables and is used to quantify the degree of association between drug concentration and spectral data. By calculating the correlation coefficient, we can determine whether the spectral data responds significantly to drug concentration, that is, whether the spectral data can serve as an effective indicator of drug concentration. A high correlation coefficient indicates that the spectral data is sensitive to drug concentration, while a low correlation coefficient indicates that the response is insensitive. The difference characteristic refers to the difference in the magnitude of the corresponding data between the two fitted curves. If there is a significant difference between the two fitted curves, it may indicate a delayed response of the spectral data to drug concentration. Understanding the delayed response helps to more accurately predict and evaluate the dynamic behavior of drugs in vivo. Therefore, the delayed response of the spectral data to drug concentration can be obtained based on the correlation coefficient and difference characteristic of the corresponding data between the two fitted curves.

[0048] Preferably, in one embodiment of the present invention, the method for obtaining delayed responsiveness includes:

[0049] The delayed response of the spectral data to drug concentration is obtained based on the correlation coefficient and mean square error of the corresponding data between the two fitting curves. The correlation coefficient is negatively correlated with the delayed response, while the mean square error is positively correlated with the delayed response. A positive correlation indicates that the dependent variable increases with an increase in the independent variable, or decreases with a decrease in the independent variable. The specific relationship can be multiplication, addition, or the power of an exponential function, and is determined by actual application. A negative correlation indicates that the dependent variable decreases with an increase in the independent variable, or increases with a decrease in the independent variable. It can be a subtraction, division, or other relationship, and is determined by actual application.

[0050] In one embodiment of the present invention, the formula for delayed responsiveness is expressed as:

[0051] ;

[0052] in, represents the delayed response of spectral data to drug concentration; represents the data set on the drug concentration fitting curve; represents the data set on the maximum absorption intensity fitting curve; Represents the data set on the drug concentration fitting curve and the data set on the maximum absorption intensity fitting curve The covariance between Represents the data set on the drug concentration fitting curve The standard deviation of Represents the data set on the maximum absorption intensity fitting curve The standard deviation of Indicates the number of corresponding data between fitting curves; Represents the data set on the drug concentration fitting curve No. individual data; Indicates the maximum absorption intensity of the fitted curve on the data set No. data.

[0053] In the formula of delayed response, Represents the data set on the drug concentration fitting curve and the data set on the maximum absorption intensity fitting curve The Pearson correlation coefficient between them is as follows: the larger the correlation coefficient, the more consistent the data change response is, and the greater the real-time response is. The smaller it is, the smaller the delay response is; Represents the data set on the drug concentration fitting curve and the data set on the maximum absorption intensity fitting curve The larger the mean square error, the greater the difference between the corresponding data, the worse the synchronization, the less likely there is real-time responsiveness, and the greater the delayed responsiveness; on the contrary, the smaller the mean square error, the smaller the difference between the corresponding data, the stronger the synchronization of changes, the greater the real-time responsiveness, and the smaller the delayed responsiveness.

[0054] It should be noted that, in one embodiment of the present invention, the Pearson correlation coefficient can be used to compare the correlation of different data. Regardless of the size of the data itself, a correlation coefficient between -1 and 1 can be obtained; the closer it is to 1, the stronger the positive linear relationship between the two rows of data, the stronger the linear relationship, and the more reference the spectral data has for the detection of drug concentration; in other embodiments of the present invention, the DTW distance of the two curves can be calculated, and the DTW distance can be negatively correlated and mapped to obtain a correlation coefficient; the specific Pearson correlation coefficient and DTW distance are technical means well known to those skilled in the art and will not be elaborated here.

[0055] It should be noted that, in other embodiments of the present invention, The method of multiplying the two parts can also use other basic mathematical operations such as addition to achieve a positive and negative correlation relationship in which the smaller the correlation coefficient, the larger the mean square error, and the greater the delayed response. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0056] Changes in drug concentration directly affect the data points in each band of the spectral curve. By analyzing the fluctuation characteristics of data points in non-characteristic bands, we can more comprehensively understand the relationship between drug concentration and spectral data. Noise and interference may exhibit different characteristics in different bands. By analyzing the variance differences in bands outside the characteristic band, we can identify which bands are more susceptible to noise and interference, thereby evaluating the reliability and stability of the spectral data. Therefore, based on the fluctuation characteristics of data points in bands outside the preset characteristic band between spectral curves corresponding to different drug concentrations, we can obtain the complexity of changes in bands outside the preset characteristic band.

[0057] Preferably, in one embodiment of the present invention, the method for obtaining the complexity of the change includes:

[0058] The fluctuation degree differences of data points in bands other than the preset characteristic band between the spectral curves corresponding to different drug concentrations were calculated, and the fluctuation degree differences between the spectral curves corresponding to all drug concentrations were averaged to obtain the change complexity of bands other than the characteristic band.

[0059] In one embodiment of the present invention, the formula for varying complexity is expressed as:

[0060] ;

[0061] in, Indicates the complexity of changes in other bands outside the preset characteristic bands; Indicates the The degree of fluctuation of the spectral curve at data points in other bands outside the preset characteristic band under different drug concentrations; Indicates the The degree of fluctuation of the spectral curve at data points in other bands outside the preset characteristic band under different drug concentrations; The serial number indicating the difference in the degree of fluctuation between the spectral curves corresponding to the drug concentration; Indicates the number of differences in the degree of fluctuation between the spectral curves corresponding to the drug concentration.

[0062] In the formula for changing complexity, Indicates the At the same drug concentration, the variance of the data points of the spectral curve in other bands outside the preset characteristic band is The fluctuation degree difference of the spectral curve in the data points of other bands outside the preset characteristic band under different drug concentrations is as follows: the smaller the fluctuation degree difference, the more consistent the data change trend in other bands outside the preset characteristic band between different drug concentrations, the higher the synchronization, indicating that the complexity of the change in other bands is smaller; on the contrary, the greater the fluctuation degree difference, the more inconsistent the data change trend in other bands outside the preset characteristic band between different drug concentrations, the worse the synchronization, indicating that the complexity of the change in other bands is greater.

[0063] It should be noted that, in one embodiment of the present invention, the degree of fluctuation can be characterized by calculating the variance. The larger the variance, the greater the degree of fluctuation. In other embodiments of the present invention, the degree of fluctuation can also be characterized by calculating the range or mean absolute deviation between data points. The larger the range and the larger the mean deviation, the greater the degree of fluctuation. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0064] The delayed response reflects the speed at which spectral data respond to changes in drug concentration. A higher delayed response means that the spectral data does not respond quickly enough to changes in drug concentration, which may reduce the real-time and accuracy of concentration monitoring. The variation complexity reflects the degree of fluctuation of spectral data at different drug concentrations. A higher variation complexity may mean that the spectral data is easily affected by various factors, such as environmental noise and sample preparation differences, thus affecting the stability of concentration monitoring. By comprehensively considering the delayed response and variation complexity, the credibility of the interference change of drug concentration can be more accurately assessed, thereby avoiding misjudgments or errors caused by interference factors and helping to improve the accuracy and reliability of concentration monitoring. Therefore, the credibility of the interference change of drug concentration is obtained based on the delayed response and variation complexity.

[0065] Preferably, in one embodiment of the present invention, the method for obtaining the credibility of the disturbed change includes:

[0066] The product of delayed response and change complexity is calculated to obtain the credibility of the disturbed change.

[0067] In one embodiment of the present invention, the formula for the credibility of the disturbed change is expressed as:

[0068] ;

[0069] in, Indicates the credibility of the interference change of drug concentration; represents the delayed response of spectral data to drug concentration; Indicates the complexity of changes in bands other than the preset characteristic bands.

[0070] In the formula for the credibility of disturbed changes, the greater the delayed response, the worse the real-time response between drug concentration and spectral data, that is, the greater the degree of interference; the worse the complexity of changes in other bands outside the preset characteristic band, the greater the degree to which the changes in the preset characteristic band are affected by the components of other bands, that is, the greater the degree of interference, and the greater the credibility of the disturbed changes.

[0071] It should be noted that, in other embodiments of the present invention, other basic mathematical operations such as addition may also be used to construct a positive correlation. The specific means are well known to those skilled in the art and will not be elaborated here.

[0072] Step S3: Obtain the spectral shape distortion of the spectral curve corresponding to each drug concentration within the preset characteristic band based on the morphological characteristics of the spectral curve corresponding to each drug concentration within the preset characteristic band, and the distribution difference characteristics of the data points between the corresponding spectral curve and the preset spectral curve within the preset characteristic band; obtain the processing benefit confidence of the spectral data based on the credibility of the interference change and the spectral shape distortion.

[0073] Changes in drug concentration directly affect intermolecular interactions and molecular vibration modes, which will be reflected in the morphology of the spectral curve; different drugs have different molecular structures, which determine the ability to absorb or scatter light of specific wavelengths, thus forming different spectral curve morphologies; by comparing the spectral curve morphologies at different concentrations, changes in drug concentration can be monitored, which can be used to identify different drugs or drug ingredients, and assist in drug quality control and safety assessment; spectral shape distortion refers to the difference in morphology and distribution between the actual spectral curve and the expected or standard spectral curve; by analyzing the spectral shape distortion, the degree of interference can be more accurately assessed; the spectral shape distortion of the spectral curve corresponding to each drug concentration in the preset characteristic band is obtained based on the morphological characteristics of the spectral curve corresponding to each drug concentration in the preset characteristic band, and the distribution difference characteristics of the data points between the corresponding spectral curve and the preset spectral curve in the preset characteristic band.

[0074] Preferably, in one embodiment of the present invention, the method for obtaining distribution difference features includes:

[0075] Calculating the standard deviation of all data points of the spectral curve in the preset characteristic band, calculating the product of the standard deviation and the amplitude of each data point of the spectral curve in the preset characteristic band to obtain a weighted amplitude;

[0076] The difference in weighted amplitudes of all data points corresponding to the spectral curve corresponding to each drug concentration and the preset spectral curve in the preset characteristic band is calculated to obtain the distribution difference characteristics.

[0077] Preferably, in one embodiment of the present invention, the method for obtaining spectral shape distortion includes:

[0078] For the spectral curve corresponding to each drug concentration, the slopes between all adjacent data points in the preset characteristic band are calculated, and the mean difference between all adjacent slopes is calculated as the spectral shape mutation;

[0079] Spectral shape distortion is determined based on the spectral shape mutation, the number of breakpoints in the spectral curve within a preset characteristic band, and the distribution difference characteristics. Spectral shape mutation and the number of breakpoints in the spectral curve within the preset characteristic band are positively correlated with spectral shape distortion, while the distribution difference characteristics are negatively correlated with spectral shape distortion. A positive correlation indicates that the dependent variable increases as the independent variable increases, or decreases as the independent variable decreases. The specific relationship can be multiplication, addition, or the power of an exponential function, and is determined by actual application. A negative correlation indicates that the dependent variable decreases as the independent variable increases, or increases as the independent variable decreases. The specific relationship can be subtraction, division, or other relationships, and is determined by actual application.

[0080] In one embodiment of the present invention, the formula for spectral shape distortion is expressed as:

[0081] ;

[0082] ;

[0083] in, Indicates the The spectral shape distortion of the spectral curve corresponding to the drug concentration in the preset characteristic band; Indicates the distribution difference characteristics; Indicates the number of breakpoints of the spectral curve within the preset characteristic band; Represents a non-zero constant between 0 and 0.1 to avoid the formula being 0; Indicates the The slope between adjacent data points in a group; Indicates Group adjacent The slope between adjacent data points in a group; A serial number indicating the difference between adjacent slopes; represents the amount of difference between adjacent slopes; Indicates the standard deviation of the data points in the preset characteristic band of the spectral curve corresponding to the drug concentration; Indicates the standard deviation of the data points of the preset spectral curve within the preset characteristic band; Indicates the number of data points in the preset characteristic band; Indicates that the drug concentration corresponds to the spectral curve within the preset characteristic band. The amplitude of the data point; Indicates that the preset spectrum curve is the first The amplitude of the data point.

[0084] In the formula for spectral shape distortion, It indicates the mean difference of all adjacent slopes of the calculated spectral curve within the preset characteristic band, that is, the spectral shape is sudden, the difference between adjacent slopes is large, and the spectral shape has a large suddenness; Indicates the calculated drug concentration corresponding to the spectral curve The standard deviation and the The product of the amplitudes of the data points is the weighted amplitude; Indicates calculation of preset spectral curve The standard deviation and the The product of the amplitudes of the data points; It represents the distribution difference characteristics of all data points in the preset characteristic band between the spectral curve corresponding to the drug concentration and the preset spectral curve. The larger the distribution difference characteristics, the greater the separation between the spectral curve corresponding to the drug concentration and the preset spectral curve, and the smaller the spectral shape distortion; conversely, the worse the distribution difference characteristics, the higher the possibility of background interference; therefore, the greater the mutation of the spectral shape, the more breakpoints, the smaller the distribution difference characteristics, the smaller the separation between the spectral curve corresponding to the drug concentration and the preset spectral curve, the more likely it is to be affected by interfering components, the greater the interference intensity of the interfering components, and the greater the spectral shape distortion.

[0085] It should be noted that in order to eliminate or correct spectral changes caused by non-drug influences such as light source fluctuations, instrument vibrations and temperature changes, and to more accurately identify spectral changes caused by drugs; in one embodiment of the present invention, the method for obtaining a preset spectral curve includes: obtaining spectral data of a blood sample when it does not contain drug concentration, and forming a corresponding preset spectral curve.

[0086] It should be noted that, in other embodiments of the present invention, The method of performing positive correlation mapping by multiplication can also be constructed through other basic mathematical operations such as addition. The specific means are technical means well known to those skilled in the art and will not be described in detail here.

[0087] By analyzing the credibility of interference variations, we assess the extent to which spectral data is affected by external factors during the measurement process. This helps identify and eliminate abnormal data caused by interference, thereby improving the reliability of spectral data. Spectral shape distortion reflects the inherent characteristics of spectral data. By analyzing spectral shape variations, we can gain a deeper understanding of the sample's optical properties, chemical composition, and molecular structure, allowing for more accurate interpretation of spectral data. The confidence level of spectral data processing benefit is determined based on the credibility of interference variations and spectral shape distortion.

[0088] Preferably, in one embodiment of the present invention, the method for obtaining the useful confidence level includes:

[0089] The mean of the spectral shape distortion corresponding to all drug concentrations was calculated as the distortion mean. The treatment benefit confidence was obtained based on the disturbed change credibility and the distortion mean. Both the disturbed change credibility and the distortion mean were positively correlated with the treatment benefit confidence.

[0090] In one embodiment of the present invention, the formula for processing the beneficial confidence is expressed as:

[0091] ;

[0092] in, Indicates the confidence level that the processing of spectral data is beneficial; Indicates the credibility of the interference change of drug concentration; Indicates the The spectral shape distortion of the spectral curve corresponding to the drug concentration in the preset characteristic band; Indicates the type of drug concentration; Indicates square root.

[0093] In the formula that deals with useful confidence, It represents the mean value of the spectral shape distortion corresponding to all drug concentrations, that is, the distortion mean. The greater the spectral shape distortion, the higher the processing intensity required for subsequent spectral data processing, and the greater the confidence in the beneficial effect of the treatment; It represents the Euclidean norm of the credibility of the disturbed changes in drug concentration and the mean of the distortion, and is a quantitative index value; the smaller the credibility of the disturbed changes in drug concentration and the mean of the distortion, the smaller the Euclidean norm, the smaller the possibility of being affected by the interfering components, and the smaller the confidence in the beneficial treatment; conversely, the larger the credibility of the disturbed changes in drug concentration and the mean of the distortion, the larger the Euclidean norm, the greater the possibility of being affected by the interfering components, and the greater the confidence in the beneficial treatment.

[0094] It should be noted that in other embodiments of the present invention, the disturbed change credibility and the distortion mean can also be directly added to perform a positive correlation mapping to construct a relationship in which the greater the disturbed change credibility and the distortion mean of the drug concentration, the greater the confidence of the beneficial treatment. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0095] Step S4: filtering the spectral data according to the processing benefit confidence to obtain filtered spectral data; detecting the drug concentration according to the filtered spectral data.

[0096] Processing benefit confidence refers to the degree of confidence in specific information or features in spectral data. Filtering based on a preset confidence level can smooth fluctuations and outliers in the spectral data, retaining only those data points deemed reliable within the confidence level. This improves the quality of the spectral data and makes subsequent drug concentration analysis more accurate and reliable. Filtering can also reduce irrelevant components and interference in spectral data, making useful signals more prominent and easier to identify. Therefore, spectral data is filtered based on the processing benefit confidence level to obtain filtered spectral data.

[0097] Preferably, in one embodiment of the present invention, the method for acquiring filtered spectral data includes:

[0098] The processing benefit confidence is used as the standard deviation parameter in the filter to filter the spectral data to obtain filtered spectral data. Figure 3 As shown, a schematic diagram of a filtered spectrum curve is shown.

[0099] It should be noted that the greater the confidence level of the beneficial treatment, the greater the possibility of being affected by interference components, and the more necessary it is to smooth and remove interference components; the smaller the confidence level of the beneficial treatment, the smaller the possibility of being affected by interference components, and the more necessary it is to reduce the degree of smoothing to retain more details; because in filter design, the standard deviation is one of the parameters of the filter, which is used to control the smoothness of the filter; the larger the standard deviation, the higher the smoothness of the filter for the signal; the smaller the standard deviation, the lower the smoothness of the filter; based on this, the confidence level of the beneficial treatment is used as the standard deviation parameter in the filter to effectively remove interference components in spectral detection and improve data accuracy. In one embodiment of the present invention, the filter can adopt a Gaussian filter. The specific Gaussian filter is a technical means well known to those skilled in the art and will not be described in detail here.

[0100] After obtaining filtered spectral data, the noise and interference in spectral detection can be effectively reduced, making the spectral data clearer and more accurate, and more accurately identifying the spectral characteristics of drug molecules, thereby obtaining more accurate concentration values. Blood drug concentration is detected based on the filtered spectral data.

[0101] It should be noted that in other embodiments of the present invention, since the characteristic band usually corresponds to the specific absorption or emission behavior of the drug molecules at a specific wavelength, parameters such as the height, width and position of the absorption peak are usually related to the drug concentration; by comparing the absorption peak characteristics at different drug concentrations, a mathematical model or relationship between the drug concentration and the absorption peak characteristics can be established to estimate the drug concentration in the blood sample.

[0102] In summary, the present invention obtains the delayed response of spectral data to drug concentration and the complexity of changes in bands other than the preset characteristic band; further obtains the credibility of the interference change of drug concentration; obtains the spectral shape distortion of the spectral curve corresponding to each drug concentration within the preset characteristic band based on the morphological characteristics of the spectral curve corresponding to each drug concentration within the preset characteristic band, and the distribution difference characteristics of the data points between the corresponding spectral curve and the preset spectral curve within the preset characteristic band; obtains the processing benefit confidence of the spectral data; filters the spectral data to obtain filtered spectral data; and detects blood drug concentration. The present invention improves the quality of spectral data and the detection accuracy of blood drug concentration by removing interfering components in spectral detection.

[0103] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0104] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A clinical blood drug concentration detection method, characterized in that: The method comprises: Obtaining spectral data of blood samples at different drug concentrations; the spectral data includes absorption intensities corresponding to different wavelengths, constituting data points of a spectral curve; Obtaining a drug concentration fitting curve and a maximum absorption intensity fitting curve of a drug concentration corresponding spectral curve within a preset characteristic band; obtaining a delayed response of the spectral data to the drug concentration based on the correlation coefficient and difference characteristics of the corresponding data between the two fitting curves; obtaining a change complexity of other bands outside the preset characteristic band based on the fluctuation characteristics of data points in other bands outside the preset characteristic band between the spectral curves corresponding to different drug concentrations; and obtaining a credibility of the interference change of the drug concentration based on the delayed response and the change complexity; Obtaining the spectral shape distortion of the spectral curve corresponding to each drug concentration within the preset characteristic band based on the morphological characteristics of the spectral curve corresponding to each drug concentration within the preset characteristic band and the distribution difference characteristics of data points between the corresponding spectral curve and the preset spectral curve within the preset characteristic band; obtaining the processing benefit confidence of the spectral data based on the interference change credibility and the spectral shape distortion; The spectral data is filtered according to the processing benefit confidence to obtain filtered spectral data; and the drug concentration is detected according to the filtered spectral data.

2. A clinical blood drug concentration detection method according to claim 1, characterized in that, The method for obtaining the delayed responsiveness includes: The delayed response of the spectral data to the drug concentration was obtained based on the correlation coefficient and mean square error of the corresponding data between the two fitting curves. The correlation coefficient was negatively correlated with the delayed response, while the mean square error was positively correlated with the delayed response.

3. A clinical blood drug concentration detection method according to claim 1, characterized in that: The method for obtaining the complexity of the change includes: The fluctuation degree differences of data points in bands other than the preset characteristic band between the spectral curves corresponding to different drug concentrations were calculated, and the fluctuation degree differences between the spectral curves corresponding to all drug concentrations were averaged to obtain the change complexity of bands other than the characteristic band.

4. A clinical blood drug concentration detection method according to claim 1, characterized in that: The method for obtaining the change credibility includes: The product of delayed response and change complexity is calculated to obtain the credibility of the disturbed change.

5. A clinical blood drug concentration detection method according to claim 1, characterized in that, The method for obtaining the distribution difference feature includes: Calculating the standard deviation of all data points of the spectral curve in the preset characteristic band, calculating the product of the standard deviation and the amplitude of each data point of the spectral curve in the preset characteristic band to obtain a weighted amplitude; The difference in weighted amplitudes of all data points corresponding to the spectral curve corresponding to each drug concentration and the preset spectral curve in the preset characteristic band is calculated to obtain the distribution difference characteristics.

6. A clinical blood drug concentration detection method according to claim 5, characterized in that: The method for obtaining the spectral shape distortion includes: For the spectral curve corresponding to each drug concentration, the slopes between all adjacent data points in the preset characteristic band are calculated, and the mean difference between all adjacent slopes is calculated as the spectral shape mutation; The spectral shape distortion is obtained based on the spectral shape mutation, the number of breakpoints of the spectral curve in the preset characteristic band, and the distribution difference characteristics. The spectral shape mutation and the number of breakpoints of the spectral curve in the preset characteristic band are positively correlated with the spectral shape distortion, while the distribution difference characteristics are negatively correlated with the spectral shape distortion.

7. A clinical blood drug concentration detection method according to claim 1, characterized in that: The method for obtaining the processing benefit confidence includes: Calculate the mean value of spectral shape distortion corresponding to all drug concentrations as the distortion mean; The confidence level of treatment benefit was obtained based on the disturbed change credibility and the distortion mean. Both the disturbed change credibility and the distortion mean were positively correlated with the treatment benefit confidence level.

8. A clinical blood drug concentration detection method according to claim 1, characterized in that: The method for obtaining filtered spectrum data includes: The processing benefit confidence is used as the standard deviation parameter in the filter to filter the spectral data to obtain filtered spectral data.

9. A clinical blood drug concentration detection method according to claim 1, characterized in that: The method for obtaining the preset spectral curve includes: Spectral data of the blood sample without drug concentration is obtained to form a corresponding preset spectral curve.

10. A clinical blood drug concentration detection method according to claim 1, characterized in that: The correlation coefficient is the Pearson correlation coefficient.

Images