A clinical pharmacy blood drug concentration detection method

Abstract

The present invention relates to the technical field of drug concentration detection, and relates to a method for detecting blood drug concentration in clinical pharmacy, including capturing drug molecules in a blood sample based on a microfluidic chip, enabling a sensor to detect the drug molecules and bind to their surface, and collecting highly sensitive blood drug reaction data. In the present invention, principal component analysis is used to identify and utilize the main variables in the data, effectively reducing the complexity of the spectral data while retaining the most critical information for drug concentration calculation, optimizing the data processing process, improving the speed and efficiency of the analysis, making the prediction of drug concentration more accurate, and providing more reliable data support for clinical practice. The moving average algorithm smoothes the data, reducing random fluctuations caused by sample sampling errors or external variables, improving the overall quality and credibility of the data, and helping to significantly reduce errors, thereby supporting doctors to make more accurate drug dosage adjustments and efficacy evaluations.

Description

Technical Field

[0001] The invention belongs to the technical field of drug concentration detection, and relates to a blood drug concentration detection method for clinical pharmacy. Technical Background

[0002] The field of drug concentration monitoring technology focuses on assessing and managing the effectiveness and safety of drug therapies by measuring the concentration of specific drugs in the blood. Drug concentration monitoring helps healthcare professionals determine the optimal dosage of a drug to maximize its effectiveness while minimizing the risk of side effects. This technology is important because it provides valuable information about a patient's drug metabolism rate, thereby supporting the development of personalized drug treatment plans.

[0003] Blood drug concentration testing in clinical pharmacy is a method used to measure drug concentrations in blood samples. Its primary purpose is to ensure the effectiveness and safety of drug treatment. By monitoring blood drug concentrations, doctors can determine whether a drug has reached its therapeutic window (the range between the minimum effective concentration and the maximum tolerated concentration). This method is also used to diagnose drug overdose, assess patient absorption and metabolism of drugs, and adjust dosages to accommodate individual differences.

[0004] Currently, there are still some areas that need to be optimized for drug concentration detection, which are specifically reflected in the following aspects:

[0005] Compared with innovative solutions, traditional blood drug concentration detection methods have certain limitations in data processing and analysis. Traditional methods rely on complex biochemical data and may not be able to effectively distinguish important features in the data, resulting in a lack of necessary accuracy in the calculation of drug concentrations. This lack of analysis may affect the judgment of the therapeutic window and increase the risk of insufficient drug efficacy or excessive side effects during treatment. Due to the lack of effective data smoothing, traditional methods often show shortcomings in dealing with data fluctuations caused by the natural variability of biological samples, which may lead to treatment decisions based on unstable data, affecting patient safety and treatment effects. These shortcomings illustrate the limitations of traditional methods in handling clinical data with high variability and complexity, especially in personalized treatment scenarios that require highly accurate measurements. Summary of the Invention

[0006] In view of the above problems in the prior art, the present invention provides a clinical pharmaceutical blood drug concentration detection method for solving the above technical problems.

[0007] In order to achieve the above-mentioned and other purposes, the technical solutions adopted by the present invention are as follows:

[0008] In one aspect, the present invention provides a method for detecting blood drug concentration in clinical pharmacy, the method comprising:

[0009] Step 1: Based on the microfluidic chip, drug molecules in the blood sample are captured, and the sensor detects the drug molecules and their binding to the surface, collecting highly sensitive blood drug reaction data;

[0010] Step 2: Using real-time spectral analysis technology, the reaction speed and sensitivity of the chemical reagents are improved by adjusting the parameters of the microfluidic chip. A miniaturized spectrometer is used to perform spectral scanning on the chip, record spectral changes, and generate chemical reaction data.

[0011] Step 3: Analyze the chemical reaction data using principal component analysis and regression models, extract key features from the spectral data, calculate drug concentrations, and obtain preliminary blood drug concentration readings;

[0012] Step 4: Based on the preliminary blood drug concentration readings, adjust the calibration curve to match the real-time measurement data and establish the real-time adjusted concentration analysis results;

[0013] Step 5: Based on the real-time adjustment concentration analysis results, the moving average algorithm is used to smooth the data to generate optimized concentration data;

[0014] Step 6: performing data verification on the optimized concentration data and performing a consistency check to obtain verified optimized concentration data;

[0015] Step 7: Based on the validated and optimized concentration data, use the report generation tool to record and integrate all drug concentration data and analysis processes to obtain a comprehensive blood drug concentration test report.

[0016] As a further solution of the present invention, in step 1, based on the microfluidic chip, the width and flow rate of the blood channel in the chip are adjusted to optimize the contact frequency and contact quality between the drug molecules and the sensor surface. At the same time, the chip environmental settings are adjusted, including the temperature and pH value of the internal channel, and the changes in the efficiency of drug molecule capture are tracked. At the same time, the micro-spectrometer on the chip is started, the wavelength and scanning speed of the spectrometer are adjusted, the spectral changes are recorded and the data output format is organized to generate highly sensitive blood drug reaction data, which includes drug absorption rate, drug metabolism rate and drug excretion rate.

[0017] As a further solution of the present invention, in step 2, the flow rate and pressure parameters of the internal channel are adjusted through the microfluidic chip to control the contact efficiency between the chemical reagent and the sensor, and the dynamic changes of the real-time reaction are monitored. At the same time, the light source intensity and detection sensitivity of the micro-spectrometer are adjusted according to the dynamic changes, and the reaction signals at each wavelength are recorded. The integrated chemical reaction process record is output to obtain chemical reaction data, which includes spectral absorbance data, spectral peak data, and spectral change rate data.

[0018] As a further solution of the present invention, in step 3, the chemical reaction data transmission is analyzed and the data is normalized, linear transformation is used to eliminate the scale differences between the data, and data quality is checked to exclude outliers. The components are separated according to the contribution of the variables in the data and a validation analysis is performed. At the same time, principal component analysis and regression model are used, and parameters are adjusted in combination with known drug concentration data. The model is tested by cross-validation to obtain preliminary blood drug concentration readings.

[0019] As a further solution of the present invention, the principal component analysis method is according to the formula:

[0020]

[0021] Where: PCA(X) is the preliminary blood drug concentration data, X is the original data matrix, α is the adjustment coefficient, is the expected value vector of the data matrix X, S is the standard deviation matrix of the data, and W is the eigenvector matrix.

[0022] As a further embodiment of the present invention, the regression model is according to the formula:

[0023] Y=β0+β1X1+β2X2+β3X3+β4X4+β5X5+β6X6+∈

[0024] Where: Y is the initial blood drug concentration reading, β0 is the intercept, β1 to β6 are associated with variables X1 to X6, representing the contribution to drug concentration, X1 and X2 are the main components affecting drug concentration obtained from principal component analysis, X3 is the time length from drug administration to sampling, X4 is the logarithm of the drug administration amount, X5 is the temperature variable, X6 is the pH value, and ∈ represents the error term in the model.

[0025] As a further embodiment of the present invention, in step 4, the preliminary blood drug concentration readings are analyzed, outliers and noise are identified and eliminated, and the data are organized into a unified format. At the same time, the difference between the data and the measurement standard is evaluated, and the data set is adjusted according to the evaluation results to obtain real-time adjusted concentration analysis results.

[0026] As a further solution of the present invention, in step 5, based on the real-time adjustment of the concentration analysis results, the data set is reorganized, and the time series data set is constructed by sorting the time labels of the continuous data blocks. Based on this data set, the average value in the continuous data blocks is calculated, and a smoothed data sequence is generated using a moving average algorithm. The data points in the smoothed data sequence are integrated, and the data are smoothed as a whole to obtain optimized concentration data.

[0027] As a further solution of the present invention, the moving average algorithm is according to the formula:

[0028]

[0029] Among them: MA t represents the average blood drug concentration calculated based on the first N data points at a specific time t, where N is the number of consecutive data points considered in the moving average calculation, and x t-i is the blood drug concentration at time point ti, t is the current time point, i is the index in the cycle, w i is the data point x t-i The weight coefficient of .

[0030] As a further solution of the present invention, in step 6, based on the optimized concentration data, data batch processing is performed, the data is divided into batches containing the same number of records in chronological order, each batch of data is independently time series labeled, the changes in each batch of data are tracked and compared, and deviation analysis is performed at the same time. The average value of each batch of data is calculated, and the difference between the average value of each batch and the average value of all data is compared. The stability of the data is evaluated by calculating the standard deviation of each batch of data, the deviation analysis data of each batch is summarized, and the overall consistency of the data set is confirmed by comprehensively evaluating the error of the data to obtain verified optimized concentration data.

[0031] As a further solution of the present invention, in step 7, the verification and optimization concentration data are summarized and the data format is unified. At the same time, the key information of each blood sample, including sampling time, concentration value and change trend, is captured and recorded. All analysis data and processes are organized according to a preset format, and a comprehensive blood drug concentration detection report is output.

[0032] As described above, the present invention provides a clinical pharmaceutical blood drug concentration detection method, which has at least the following beneficial effects:

[0033] In the present invention, principal component analysis is used to identify and utilize the main variables in the data, effectively reducing the complexity of spectral data while retaining the most critical information for drug concentration calculation, optimizing the data processing process, improving the speed and efficiency of analysis, making the prediction of drug concentration more accurate, and providing more reliable data support for clinical practice. The moving average algorithm smoothes the data, reduces random fluctuations caused by sample sampling errors or external variables, improves the overall quality and credibility of the data, and helps to significantly reduce errors, thereby supporting doctors to make more accurate drug dosage adjustments and efficacy evaluations, greatly improving data processing capabilities and the stability of analysis results, and providing a solid data foundation for individualized drug treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. 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 creative work.

[0035] Figure 1 It is a schematic diagram of the connection of each step of the method of the present invention. DETAILED DESCRIPTION

[0036] The above contents described below in conjunction with the implementation of the present invention are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they shall fall within the scope of protection of the present invention.

[0037] See also Figure 1 As shown, a clinical pharmaceutical blood drug concentration detection method, the method comprising:

[0038] Step 1: Based on the microfluidic chip, drug molecules in the blood sample are captured, and the sensor detects the drug molecules and their binding to the surface, collecting highly sensitive blood drug reaction data;

[0039] Step 2: Using real-time spectral analysis technology, the reaction speed and sensitivity of the chemical reagents are improved by adjusting the parameters of the microfluidic chip. A miniaturized spectrometer is used to perform spectral scanning on the chip, record spectral changes, and generate chemical reaction data.

[0040] Step 3: Analyze the chemical reaction data using principal component analysis and regression models, extract key features from the spectral data, calculate drug concentrations, and obtain preliminary blood drug concentration readings;

[0041] Step 4: Based on the preliminary blood drug concentration readings, adjust the calibration curve to match the real-time measurement data and establish the real-time adjusted concentration analysis results;

[0042] Step 5: Based on the real-time adjustment concentration analysis results, the moving average algorithm is used to smooth the data to generate optimized concentration data;

[0043] Step 6: performing data verification on the optimized concentration data and performing a consistency check to obtain verified optimized concentration data;

[0044] Step 7: Based on the validated and optimized concentration data, use the report generation tool to record and integrate all drug concentration data and analysis processes to obtain a comprehensive blood drug concentration test report.

[0045] In step 1, based on the microfluidic chip, the width and flow rate of the blood channel within the chip are adjusted to optimize the contact frequency and quality between the drug molecules and the sensor surface. The chip environmental settings, including the temperature and pH value of the internal channel, are adjusted, and the changes in the efficiency of drug molecule capture are tracked. At the same time, the on-chip micro-spectrometer is activated, the wavelength and scanning speed of the spectrometer are adjusted, the spectral changes are recorded, and the data output format is organized to generate highly sensitive blood drug reaction data, which includes drug absorption rate, drug metabolism rate, and drug excretion rate.

[0046] It should be noted that based on the microfluidic chip, the width of the blood channel in the chip is adjusted to 0.5 mm and the flow rate is adjusted to 1 ml per second. The contact frequency between the drug molecules and the sensor surface is optimized, and the contact quality is detected by an optical sensor. At the same time, the chip environment settings are adjusted, the temperature of the internal channel is set to 37 degrees Celsius, and the pH value is adjusted to 7.4. A real-time monitoring system is used to track changes in the capture efficiency of drug molecules. Using a miniature spectrometer, the wavelength range is set to 200-800 nanometers and the scanning speed is adjusted to 100 nanometers per second. The spectral changes are recorded and the data output format is organized through the data management system to generate highly sensitive blood drug reaction data.

[0047] It should be noted that the microfluidic chip first receives a small amount of blood sample. The chip is designed with specific channels and surfaces that have been chemically modified. The drug molecules in the blood are captured through molecular recognition technology, including the binding of antibodies and antigens or other affinity interactions, to ensure that only the target drug molecules are bound to the chip surface. When the drug molecules are captured on the chip, they will react to generate detectable signals. Finally, the data captured and converted into signals are read by the sensor on the chip and converted into electronic data.

[0048] In step 2, the flow rate and pressure parameters of the internal channel are adjusted through the microfluidic chip to control the contact efficiency between the chemical reagent and the sensor, and the dynamic changes of the real-time reaction are monitored. At the same time, the light source intensity and detection sensitivity of the micro-spectrometer are adjusted according to the dynamic changes, and the reaction signals at each wavelength are recorded. The integrated chemical reaction process record is output to obtain chemical reaction data. The chemical reaction data includes spectral absorbance data, spectral peak data, and spectral change rate data.

[0049] In step 3, the chemical reaction data transmission is analyzed and normalized, linear transformation is used to eliminate scale differences between the data, and data quality is checked to exclude outliers. Components are separated according to the contribution of variables in the data and validation analysis is performed. At the same time, principal component analysis and regression model are used, and parameters are adjusted in combination with known drug concentration data. The model is tested by cross-validation to obtain preliminary blood drug concentration readings;

[0050] It should be noted that, based on the chemical reaction data, a linear transformation method was used to eliminate the scale differences between the data. The specific adjustment coefficient of the method was 0.01 to adapt to data at different concentration levels, and data standardization was performed. Statistical software was used to check data quality, and the Z-score method was used to identify and exclude outliers in the data. The coefficient was set to 3, indicating that data exceeding three standard deviations from the mean was considered abnormal. The principal component analysis method was used to separate the components of the main variables in the data, and the separated components were verified and analyzed. The load vector in the principal component analysis was adjusted in combination with known drug concentration data, and the parameters were set so that each principal component was not less than 10% of the total variation. The chemical reaction data was analyzed using a regression model, and the model parameters were adjusted to fit using the least squares method. The cross-validation method was used, and the fold number of cross-validation was set to 5. The predictive performance of the model was tested to obtain preliminary blood drug concentration readings.

[0051] The principal component analysis method is based on the formula:

[0052]

[0053] Where: PCA(X) is the preliminary blood drug concentration data, X is the original data matrix, α is the adjustment coefficient, is the expected value vector of the data matrix X, S is the data standard deviation matrix, and W is the eigenvector matrix;

[0054] Execution process: First, the original data matrix X is centrally processed. The calculation method is Where α is the adjustment factor used to optimize the degree of data concentration, and It is the expected value vector of the data matrix X. Then, by calculating the standard deviation of each variable, the data standard deviation matrix S is constructed, and the centralized data is standardized to eliminate the scale difference between the data. Then, the covariance matrix of the standardized data is constructed, and the eigenvalue decomposition is used to obtain the eigenvector matrix W. Finally, W is used to standardize the data. Perform transformation to extract the main components of the data.

[0055] The regression model is based on the formula:

[0056] Y=β0+β1X1+β2X2+β3X3+β4X4+β5X5+β6X6+∈

[0057] Where: Y is the initial blood drug concentration reading, β0 is the intercept, β1 to β6 are associated with variables X1 to X6, representing the contribution to drug concentration, X1 and X2 are the main components affecting drug concentration obtained from principal component analysis, X3 is the time length from drug administration to sampling, X4 is the logarithm of the drug administration amount, X5 is the temperature variable, X6 is the pH value, ∈ represents the error term in the model;

[0058] Implementation process: First, each X i That is, X1, X2, …, X n represents the key components extracted from principal component analysis. The linear transformation eliminates the scale differences between the original data. i That is, β1, β2,…, β n is the coefficient of each variable, calculated by minimizing the difference between the predicted drug concentration and the actual drug concentration, and β0 is the intercept of the model, which provides the model with the i The baseline value is zero, ∈ represents the error term, which includes random variations that the model cannot explain. Finally, the model is tested using the cross-validation method, and the drug concentration value Y is obtained according to the regression model.

[0059] In step 4, the preliminary blood drug concentration readings are analyzed, outliers and noise are identified and eliminated, and the data are organized into a unified format. At the same time, the difference between the data and the measurement standard is evaluated, and the data set is adjusted according to the evaluation results to obtain real-time adjusted concentration analysis results.

[0060] In step 5, based on the real-time adjustment of the concentration analysis results, the data set is reorganized, and the time series data set is constructed by sorting the time labels of the continuous data blocks. Based on this data set, the average value in the continuous data blocks is calculated, and the moving average algorithm is used to generate a smooth data sequence. The data points in the smooth data sequence are integrated, and the data is smoothed as a whole to obtain optimized concentration data.

[0061] The moving average algorithm is based on the formula:

[0062]

[0063] Among them: MA t represents the average blood drug concentration calculated based on the first N data points at a specific time t, where N is the number of consecutive data points considered in the moving average calculation, and x t-i is the blood drug concentration at time point ti, t is the current time point, i is the index variable, w i is the data point x t-i The weight coefficient of .

[0064] Execution process: First determine each data point x t-i , represents the actual blood drug concentration value at time ti, and then set N, which determines the number of data points to be considered in the moving average calculation, through Calculate the weight coefficient w i , where γ is the decay coefficient, which is used to adjust the weight according to the distance between the data point and the current time point to ensure that recent data has a greater impact on the result. Finally, the weighted data w i ·x t-iSummarize and divide by N to get the weighted moving average MA at time point t t ,The algorithm not only calculates the average value of the data, but also takes into account the timeliness and relevance of the data points, thereby generating a data series that is smooth and reflects the recent trend.

[0065] In step 6, based on the optimized concentration data, data batch processing is performed, and the data is divided into batches containing the same number of records in chronological order. Each batch of data is independently labeled with a time series, and the changes in the data of each batch are tracked and compared. At the same time, deviation analysis is performed, and the average value of each batch of data is calculated. The difference between the average value of each batch and the average value of all data is compared, and the stability of the data is evaluated by calculating the standard deviation of each batch of data. The deviation analysis data of each batch is summarized, and the overall consistency of the data set is confirmed by comprehensively evaluating the errors of the data to obtain verified optimized concentration data.

[0066] In step 7, the validation and optimization concentration data are summarized and the data format is unified. At the same time, the key information of each blood sample, including sampling time, concentration value and change trend, is captured and recorded. All analysis data and processes are organized according to the preset format, and a comprehensive blood drug concentration test report is output.

[0067] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A method for detecting blood drug concentration in clinical pharmacy, characterized in that: The method includes: Step 1: Based on a microfluidic chip, drug molecules in a blood sample are captured and highly sensitive blood-drug reaction data is collected. Based on the microfluidic chip, the width and flow rate of the blood channel within the chip are adjusted to optimize the contact frequency and quality between the drug molecules and the sensor surface. The chip environmental settings, including the temperature and pH value of the internal channel, are also adjusted, and changes in the efficiency of drug molecule capture are tracked. Simultaneously, an on-chip micro-spectrometer is activated, the wavelength and scanning speed of the spectrometer are adjusted, spectral changes are recorded, and data output format is organized to generate highly sensitive blood-drug reaction data, which includes drug absorption rate, drug metabolism rate, and drug excretion rate. Step 2: Using real-time spectral analysis technology, and adjusting the parameters of the microfluidic chip to improve the reaction speed and sensitivity of the chemical reagent, a miniaturized spectrometer is used to perform spectral scanning on the chip, record spectral changes, and generate chemical reaction data. The flow rate and pressure parameters of the internal channel of the microfluidic chip are adjusted to control the contact efficiency between the chemical reagent and the sensor, and the dynamic changes of the real-time reaction are monitored. The light source intensity and detection sensitivity of the micro-spectrometer are adjusted according to the dynamic changes, and the reaction signals at each wavelength are recorded. The integrated chemical reaction process record is output to obtain chemical reaction data, which includes spectral absorbance data, spectral peak data, and spectral change rate data; Step 3: Analyze the chemical reaction data using principal component analysis and regression models, extract key features from the spectral data, calculate drug concentrations, and obtain preliminary blood drug concentration readings. Analyze and normalize the transmitted chemical reaction data, use linear transformations to eliminate scale differences between the data, perform data quality checks, exclude outliers, separate components based on the contribution of variables in the data, and perform validation analysis. Simultaneously, use principal component analysis and regression models, combine with known drug concentration data to adjust parameters, and test the model through cross-validation to obtain preliminary blood drug concentration readings. Step 4: Based on the preliminary blood drug concentration readings, adjust the calibration curve to match the real-time measurement data and establish the real-time adjusted concentration analysis results; Step 5: Based on the real-time adjustment concentration analysis results, the moving average algorithm is used to smooth the data to generate optimized concentration data; Step 6: performing data verification on the optimized concentration data and performing a consistency check to obtain verified optimized concentration data; Step 7: Based on the validated and optimized concentration data, use the report generation tool to record and integrate all drug concentration data and analysis processes to obtain a comprehensive blood drug concentration test report. The principal component analysis method is based on the formula: in: This is the preliminary blood drug concentration data. is the original data matrix, is the adjustment coefficient, is the data matrix The expected value vector of is the standard deviation matrix of the data, is the eigenvector matrix; The regression model is based on the formula: in: For preliminary blood drug concentration readings, is the intercept, to With variables to is associated with, indicating the contribution to drug concentration, and is the main component affecting drug concentration obtained from principal component analysis, is the time from drug administration to sampling, is the logarithm of the dose, is the temperature variable, is the pH value, represents the error term in the model.

2. A clinical pharmaceutical blood drug concentration detection method according to claim 1, characterized in that: In step 4, the preliminary blood drug concentration readings are analyzed, outliers and noise are identified and eliminated, and the data are organized into a unified format. At the same time, the difference between the data and the measurement standard is evaluated, and the data set is adjusted according to the evaluation results to obtain real-time adjusted concentration analysis results.

3. A clinical pharmaceutical blood drug concentration detection method according to claim 1, characterized in that: In step 5, based on the real-time adjustment of the concentration analysis results, the data set is reorganized, and the time series data set is constructed by sorting the time labels of the continuous data blocks. Based on this data set, the average value in the continuous data blocks is calculated, and the moving average algorithm is used to generate a smooth data sequence. The data points in the smooth data sequence are integrated, and the data is smoothed as a whole to obtain optimized concentration data.

4. A clinical pharmaceutical blood drug concentration detection method according to claim 1, characterized in that: The moving average algorithm is based on the formula: in: Represents a specific moment Based on the previous The average blood drug concentration calculated from the data points is is the number of consecutive data points considered in the moving average calculation, For Blood drug concentration at time point, For the current time point, is the index in the loop, For data points The weight coefficient of .

5. A clinical pharmaceutical blood drug concentration detection method according to claim 1, characterized in that: In step 7, the validation and optimization concentration data are summarized and the data format is unified. At the same time, the key information of each blood sample, including sampling time, concentration value and change trend, is captured and recorded. All analysis data and processes are organized according to the preset format, and a comprehensive blood drug concentration test report is output.

Images