Correction curve establishment method based on multi-weight parallel fitting in liquid chromatogram data management software

Abstract

The invention provides a calibration curve establishment method based on multi-weight parallel fitting in liquid chromatogram data management software, relates to the cross technical field of a chemometrics algorithm and chromatographic data analysis, and aims to establish and implement a full-automatic technical process of parallel fitting-automatic evaluation-intelligent decision making. According to the invention, the establishment of the correction curve is converted from a process depending on user experience and manual trial and error into a technical optimization process which is automatically completed by the system and based on objective quantitative indexes. According to the method, the data processing efficiency can be improved, the subjectivity and complexity of human intervention can be reduced, and better fitting results can be obtained under different given data sets through comprehensive comparison and automatic optimization of multiple models, so that the accuracy, the reliability and the automation level of liquid chromatogram quantitative analysis are remarkably improved.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of chemometrics algorithms and chromatographic data analysis, and in particular to a method for establishing calibration curves based on multi-weighted parallel fitting in liquid chromatography data management software. Background Technology

[0002] In the field of liquid chromatography (LC), a calibration curve is a quantitative relationship curve established by measuring a series of standard samples of known concentrations, between the analyte concentration and the instrument response value (such as chromatographic peak area or peak height). This curve is the core basis for quantitative chromatographic analysis and is widely used in fields such as pharmaceutical analysis, environmental monitoring, and food safety. The accuracy of the calibration curve directly affects the reliability of the final analytical results; therefore, establishing an accurate calibration curve is a crucial step in LC analysis.

[0003] Currently, liquid chromatography data management software has significant limitations in establishing calibration curves. First, traditional software only supports a few weight types such as 1 / x and 1 / x², which cannot meet the needs of complex analytical scenarios, especially performing poorly in processing nonlinear responses or data with high heteroscedasticity. Second, existing technologies lack support for logarithmic weights (such as Ln x, lg x) and polynomial weights (such as x, x²), resulting in insufficient fitting accuracy for specific concentration-response relationships. Furthermore, when faced with complex data, users often need to manually export data to third-party tools (such as Excel) for advanced weight calculations. This process is not only cumbersome but also prone to introducing human error, reducing analytical efficiency and the reliability of results. These problems severely restrict the further development and application of liquid chromatography analysis technology. Summary of the Invention

[0004] In view of this, the present invention provides a method for establishing calibration curves based on multi-weighted parallel fitting in liquid chromatography data management software, aiming to solve the problems of low efficiency and strong subjectivity of fitting results caused by the reliance on manual selection of a single weight in the prior art.

[0005] Therefore, the present invention provides the following technical solution: This invention provides a method for establishing a calibration curve based on multi-weighted parallel fitting in liquid chromatography data management software, comprising the following steps: Acquire and integrate the standard concentration and corresponding peak area or peak height data output by the liquid chromatograph, and use the standard concentration and corresponding peak area or peak height data as a set of standard data. Multiple pre-stored mathematical weighting functions are called in parallel to simultaneously perform weighted least squares fitting on the same set of input standard data, generating multiple candidate correction curves and their fitting parameters corresponding to each weight type. Calculate the quantized goodness-of-fit parameter for each candidate calibration curve; The system automatically selects the optimal weight type and corresponding correction curve based on preset evaluation rules, and generates the final curve correction result.

[0006] Furthermore, the mathematical weighting function includes at least two of the following: equal weight, x, x², 1 / x, 1 / x², 1 / y, 1 / y², Ln x, and lg x.

[0007] Furthermore, the quantification goodness-of-fit parameters include the coefficient of determination R² and the residual sum of squares (WRSS).

[0008] Furthermore, the preset evaluation rules include at least one of the following: Select the candidate calibration curve with the largest coefficient of determination R²; Select the candidate correction curve with the smallest sum of squared residuals (WRSS); The candidate correction curve with the highest score was selected by using a weighted composite score of R² and WRSS. Multi-parameter evaluation is performed by combining R², WRSS, and curve residual distribution characteristics; the curve residual distribution characteristics include: residual variance, residual trend with concentration, or residual normality test results.

[0009] Furthermore, the method also includes: The selected optimal weight type and its corresponding calibration curve are stored in the database.

[0010] Furthermore, the method also includes: Displays the optimal weight type and the corresponding fitting parameters and goodness-of-fit parameters of the calibration curve.

[0011] Furthermore, the method also includes: The optimal weighting type and the corresponding calibration curve are applied to quantitative analysis by liquid chromatography.

[0012] Furthermore, the method also includes: After generating the final curve correction results, a complete report is output containing the optimal weight type, the correction curve equation, and the goodness-of-fit parameters.

[0013] The above technical solution has the following beneficial effects: By constructing and implementing the fully automated technical process of "parallel fitting-automatic evaluation-intelligent decision-making," this invention transforms the establishment of calibration curves from a process dependent on user experience and manual trial and error into a technical optimization process automatically completed by the system based on objective quantitative indicators. This invention can improve data processing efficiency, reduce the subjectivity and tediousness of human intervention, and, through comprehensive comparison and automatic optimization of multiple models, ensures that better fitting results can be obtained under different types of datasets, thereby significantly improving the accuracy, reliability, and automation level of liquid chromatography quantitative analysis. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart of a method for establishing a calibration curve based on multi-weighted parallel fitting in liquid chromatography data management software according to an embodiment of the present invention. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] like Figure 1As shown in the figure, a method for establishing a calibration curve based on multi-weighted parallel fitting in liquid chromatography data management software in an embodiment of the present invention specifically includes the following steps: S1. Automatically acquire and integrate the standard concentration (x) and corresponding peak area / peak height (y) data output by the liquid chromatograph; S2, Multi-weighted Parallel Fitting Step: Automatically and in parallel call multiple mathematical weighting functions pre-stored in the system (including equal weight, x, x... 2 1 / x, 1 / x 2 , 1 / y, 1 / y 2 The weighted least squares method is simultaneously applied to a set of standard sample data (concentration x and response value y) after S1 processing, thereby generating multiple candidate calibration curves and their fitting parameters corresponding to each weight type at once; This step can be parallelized by creating multiple computation threads or using vector operation instructions. It generates a set of 9 sets of fitting results, each set including: weight type, equation of the calibration curve (slope a, intercept b), and coefficient of determination R0. 2 WRSS (Residual Sum of Squares)

[0019] In practice, nine weight functions are predefined in the mathematical weight library (as shown in Table 1), allowing users to manually select them. The weight formula engine can parse and calculate functions such as w. i =lg(x i ) or w i =1 / y i Expressions such as...

[0020] The liquid chromatography data management software has a weight selection panel: the software drop-down menu lists 9 weight types, and also provides "formula explanation", which explains the applicable scenarios for each weight in the software manual (such as "lg x is suitable for wide concentration range analysis").

[0021] Table 1

[0022] Generate weight coefficients w based on the selected weight type. i The system performs weighted least squares (WLS) fitting; the calibration curve fitting module integrates weighted least squares (WLS), supports linear models, and outputs fitting parameters (slope, intercept, R²) and residual analysis reports. Specifically, data preprocessing is performed before weight calculation, including noise filtering and concentration-response alignment; noise filtering specifically involves smoothing the raw data using median filtering and calculating the peak area and peak height as the response value (y). Concentration-response alignment specifically involves ensuring that each standard concentration point (x) is aligned with the target concentration. i ) and the corresponding peak x response value (y iStrict matching is performed to remove invalid data (such as undetected peaks).

[0023] Calculate the weight coefficient w for each data point based on the weight type selected by the user. i (See Table 2).

[0024] Table 2

[0025] S3, Automatic evaluation of fitting effect: Calculate the quantitative goodness-of-fit parameter for each candidate calibration curve generated in S2.

[0026] Among them, the parameters for quantifying goodness of fit include the coefficient of determination R² and the residual sum of squares WRSS.

[0027] S4. Optimal Result Automatic Decision and Output Steps: Based on preset evaluation rules (e.g., using R...). 2 The system automatically selects the optimal weight type and corresponding correction curve based on the highest composite score of the WRSS, and generates the final curve correction result from the result output module.

[0028] The preset evaluation rules include at least one of the following: selecting the candidate correction curve with the largest coefficient of determination R²; selecting the candidate correction curve with the smallest sum of squared residuals (WRSS); using a weighted comprehensive score of R² and WRSS, for example: comprehensive score = R² - 0.001 * WRSS, selecting the candidate correction curve with the highest score; and performing multi-parameter evaluation by combining R², WRSS, and curve residual distribution characteristics; the curve residual distribution characteristics include: the variance of the residuals, the trend of residuals changing with concentration, or the normality test results of the residuals. In specific implementation, when using "maximizing R²", 2 "The principle serves as a preset rule, and the system automatically compares 9 Rs." 2 Value, select R 2 The set of fitting results with the largest value is taken as the final output.

[0029] Furthermore, the automatically selected optimal correction curve (including its weight type, curve equation, R) can be used for... 2 The values ​​(e.g.) are visualized and a quantitative analysis report is generated.

[0030] The liquid chromatography data management software features a visual main window that displays calibration curves, overlays raw data points, and fit curves in corresponding modules. It also includes a parameter display area that updates slope (a), intercept (b), R² value, and other curve-related parameter calculation results in real time.

[0031] To facilitate understanding, the following section will use the detection of concentrations of intermediate components in pharmaceutical production (linear response) as an example to explain in detail the above method for establishing calibration curves.

[0032] Data range: concentration x = [1, 2, 5, 10, 20, 80] ppm, peak area y = [450, 1400, 3700, 9800, 30000, 95500].

[0033] 1. After the system loads the data, it automatically completes parallel fitting. The core indicators of the fitting results for the nine weighted methods are shown in the following example:

[0034] 2. The system's automatic evaluation found that the weight "x²" had the highest R² value (0.9969).

[0035] 3. The system automatically makes a decision, determining the calibration curve with the "x²" weight as the optimal result and outputting a final report. The final data selection weight type is "x²" (emphasizing the dominance of high-concentration data).

[0036] Compared to traditional methods, users do not need to manually select or repeatedly try different solutions; the system can automatically and quickly provide the optimal fitting solution for the current data, significantly improving both accuracy and efficiency. Furthermore, this invention incorporates traditional correction (equal weighting) into the automatic optimization process, avoiding blindly recommending complex weighted models. When the data itself is of extremely high quality and equal weighting is the optimal solution, the system will recommend using equal weighting based on the data calculation results. This ensures the scientific validity and reliability of the entire solution.

[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for establishing a calibration curve based on multi-weighted parallel fitting in liquid chromatography data management software, characterized in that, Includes the following steps: Acquire and integrate the standard concentration and corresponding peak area or peak height data output by the liquid chromatograph, and use the standard concentration and corresponding peak area or peak height data as a set of standard data. Multiple pre-stored mathematical weighting functions are called in parallel to simultaneously perform weighted least squares fitting on the same set of input standard data, generating multiple candidate correction curves and their fitting parameters corresponding to each weight type. Calculate the quantized goodness-of-fit parameter for each candidate calibration curve; The system automatically selects the optimal weight type and corresponding correction curve based on preset evaluation rules, and generates the final curve correction result.

2. The method for establishing a calibration curve according to claim 1, characterized in that, The mathematical weighting function includes at least two of the following: equal weight, x, x², 1 / x, 1 / x², 1 / y, 1 / y², Ln x, and lg x.

3. The method for establishing a calibration curve according to claim 1, characterized in that, The quantification goodness-of-fit parameters include the coefficient of determination R² and the residual sum of squares (WRSS).

4. The method for establishing a calibration curve according to claim 3, characterized in that, The preset evaluation rules include at least one of the following: Select the candidate calibration curve with the largest coefficient of determination R²; Select the candidate correction curve with the smallest sum of squared residuals (WRSS); The candidate correction curve with the highest score was selected by using a weighted composite score of R² and WRSS. Multi-parameter evaluation is performed by combining R², WRSS, and curve residual distribution characteristics; The residual distribution characteristics of the curve include: the variance of the residuals, the trend of residuals changing with concentration, or the normality test results of the residuals.

5. The method for establishing a calibration curve according to claim 1, characterized in that, The method further includes: The selected optimal weight type and its corresponding calibration curve are stored in the database.

6. The method for establishing a calibration curve according to claim 1, characterized in that, The method further includes: Displays the optimal weight type and the corresponding fitting parameters and goodness-of-fit parameters of the calibration curve.

7. The method for establishing a calibration curve according to claim 1, characterized in that, The method further includes: The optimal weighting type and the corresponding calibration curve are applied to quantitative analysis by liquid chromatography.

8. The method for establishing a calibration curve according to claim 1, characterized in that, The method further includes: After generating the final curve correction results, a complete report is output containing the optimal weight type, the correction curve equation, and the goodness-of-fit parameters.

Images