Method and system for judging quality uniformity of herba epimedii medicinal material

Abstract

The invention provides a method and a system for judging quality uniformity of herba epimedii medicinal materials, and relates to the technical field of quality uniformity detection of traditional Chinese medicine, the method comprises the following steps: collecting spectral data and physicochemical values of herba epimedii samples, determining a homogenization index range, and dividing a training set and a verification set; preprocessing the training set data, establishing a PCA (Principal Component Analysis) model, and identifying a non-uniform sample through a Hotellag's T2 value; using the model to predict a verification set sample, verifying the validity of the model, and adjusting the model according to a result; and measuring the spectrum of a sample to be homogenized under the same condition, and judging the homogeneity of the sample by using the established model. According to the method, the quality of the herba epimedii medicinal material is accurately controlled and evaluated.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine quality uniformity testing technology, and in particular to a method and system for determining the quality uniformity of Epimedium medicinal material. Background Technology

[0002] With the widespread application of Traditional Chinese Medicine (TCM) globally, the requirements for the uniformity of TCM quality are increasingly stringent. The uniformity of TCM quality not only relates to the stability and safety of drug efficacy but is also a crucial indicator of the modernization and internationalization of TCM. Therefore, developing efficient and accurate technologies for detecting the uniformity of TCM quality has become a hot topic and a challenge in the current field of TCM research.

[0003] Traditional methods for testing the quality of traditional Chinese medicine, such as high-performance liquid chromatography (HPLC) for content determination, while highly accurate, are complex to operate, time-consuming, and resource-intensive, making them unsuitable for large-scale production and online testing. Furthermore, HPLC is affected by various factors such as the source of the medicinal materials and the extraction method, resulting in poor result stability.

[0004] In recent years, with the rapid development of spectral analysis technology, the application of near-infrared spectrometers in the field of traditional Chinese medicine quality testing has gradually increased. Near-infrared spectrometers can quickly acquire the spectral information of medicinal materials, and the uniformity of these materials can be assessed by analyzing the spectral data. This method has advantages such as fast detection speed, no need for complex pretreatment, and simple operation, making it very suitable for online detection and quality control in large-scale production processes. However, the accuracy of near-infrared spectroscopy is relatively low, and it mainly relies on deviation calculations, failing to provide more detailed and comprehensive quality information.

[0005] To address the aforementioned issues, Chinese patents CN105699505A and CN101832921A proposed a method for intelligent analysis of traditional Chinese medicine fingerprint spectra to ensure the uniformity of traditional Chinese medicine quality, and a method for near-infrared online detection to control the uniformity of two-dimensional mixing of traditional Chinese medicine powders, respectively. These methods have improved the efficiency and accuracy of traditional Chinese medicine quality uniformity detection to some extent, but some limitations and challenges still exist.

[0006] Therefore, developing more efficient, accurate, and comprehensive technologies for detecting the uniformity of traditional Chinese medicine quality is one of the important directions in current research on traditional Chinese medicine. Summary of the Invention

[0007] To address this, this invention provides a method and system for determining the quality uniformity of Epimedium medicinal material, which solves the problems of existing methods for analyzing the quality uniformity of traditional Chinese medicine, such as complex operation, time consumption, high cost, low accuracy, or reliance on deviation calculations.

[0008] To address the aforementioned problems, embodiments of the present invention provide a method for determining the quality uniformity of Epimedium medicinal material, the method comprising:

[0009] Step S1: Use near-infrared technology to collect spectral data of Epimedium medicinal material and simultaneously measure the corresponding physicochemical values. Collect multiple groups of Epimedium samples from different production batches, define the homogenization index according to production needs, and obtain training set and validation set.

[0010] Step S2: Preprocess the spectral data in the training set;

[0011] Step S3: Use the preprocessed spectral data to build a homogenization model and determine the optimal number of principal components through interactive validation in order to capture the most important information in the spectral data;

[0012] Step S4: Using the PCA dimensionality reduction results, calculate the Hotelling's T2 value of the sample, set a confidence line to identify non-homogeneous samples, and remove abnormal samples whose Hotelling's T2 value exceeds the confidence line;

[0013] Step S5: Use the established homogenization model to predict the validation set samples, calculate the Hotelling's T2 value of each sample in the validation set, and draw the T-squared test chart to check whether the model can successfully identify non-homogeneous samples. If there are misjudgments, check whether there are any abnormalities in the spectral acquisition and physicochemical testing process of the validation set. If there are no abnormalities, re-select the training set, reconstruct the homogenization model, and perform validity verification again.

[0014] Preferably, the method further includes:

[0015] Step S6: Under the same environmental, instrumental, and experimental conditions, measure the spectrum of the sample to be homogenized, call the established homogenization model, and calculate Hotelling's T of the spectrum of the sample to be homogenized. 2 Value, according to Hotelling's T 2 By comparing the values ​​with the confidence line, non-homogeneous samples can be identified.

[0016] Preferably, the method for obtaining the training set and validation set is as follows:

[0017] From the collected samples, standard samples covering the range of homogenization indicators are selected as the training set, while ensuring that the training set data has sufficient variability and representativeness.

[0018] The validation set is selected from samples within the homogenization index range and non-homogeneous samples outside the range to comprehensively evaluate the model's performance.

[0019] Preferably, the method for preprocessing the spectral data in the training set includes at least smoothing, derivative calculation, and normalization.

[0020] Preferably, the calculation of the Hotelling's T of the sample 2 The method for setting the value is as follows:

[0021] T 2 =X T PΛ -1 P T X

[0022] in

[0023] Λ=diag{λ1,…λ k}

[0024] In the formula, T 2 Hotelling's T 2 Value; X represents the covariance matrix; P represents the characteristic matrix; λ k These are the eigenvalues ​​of the X covariance matrix.

[0025] Preferably, the confidence line is calculated using the following formula:

[0026]

[0027] In the formula, T α The confidence curve is given by k, which represents the dimension after PCA dimensionality reduction; n represents the sample size, i.e., the product of the number of collection batches and the number of cutoff time points in each batch; F α (k,nk) represents an F-distribution with k as the first degree of freedom and nk as the second degree of freedom; P{F(n1,n2)>F α (n1,n2)}=α, where 1-α is the confidence interval and α is 0.05.

[0028] This invention also provides a system for determining the quality uniformity of Epimedium medicinal materials. This system is used to implement the aforementioned method for determining the quality uniformity of Epimedium medicinal materials, specifically including:

[0029] The data acquisition module is used to acquire spectral data of Epimedium medicinal material using near-infrared technology, and simultaneously measure the corresponding physicochemical values. It collects multiple groups of Epimedium samples from different production batches, defines homogenization indicators according to production needs, and obtains training and validation sets.

[0030] The data preprocessing module is used to preprocess the spectral data in the training set;

[0031] The homogenization model building module is used to build a homogenization model using preprocessed spectral data and determine the optimal number of principal components through interactive verification in order to capture the most important information in the spectral data.

[0032] The calculation and elimination module is used to calculate Hotelling's T of the samples using the PCA dimensionality reduction results. 2Values ​​are set to establish confidence lines for identifying non-homogeneous samples and eliminating Hotelling's T values. 2 Abnormal samples whose values ​​exceed the confidence threshold;

[0033] The validation module is used to predict validation set samples using the established homogenized model and calculate Hotelling's T for each sample in the validation set. 2 The values ​​are calculated, and a T-square test plot is plotted to check whether the model can successfully identify non-homogeneous samples. If there are misjudgments, check whether there are any abnormalities in the spectral acquisition and physicochemical testing process of the validation set. If there are no abnormalities, the training set is re-selected, the homogenized model is reconstructed, and the validity is verified again.

[0034] Preferably, the system further includes:

[0035] The application module is used to measure the spectrum of a sample to be homogenized under the same environmental, instrumental, and experimental conditions, call the established homogenization model, and calculate Hotelling's T of the spectrum of the sample to be homogenized. 2 Value, according to Hotelling's T 2 By comparing the values ​​with the confidence line, non-homogeneous samples can be identified.

[0036] This invention also provides an electronic device, which includes a processor, a memory, and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the above-described method for determining the quality uniformity of Epimedium medicinal materials.

[0037] This invention also provides a computer storage medium storing a computer software product, the computer software product including several instructions for causing a computer device to execute the above-described method for determining the quality uniformity of Epimedium medicinal materials.

[0038] As can be seen from the above technical solutions, this invention application has the following beneficial effects:

[0039] (1) Non-destructive testing: Near-infrared technology is used to collect spectral data without the need for destructive treatment of medicinal materials, which can maintain the integrity and originality of medicinal materials, which is especially important for precious medicinal materials.

[0040] (2) Fast and efficient: The acquisition and processing speed of spectral data is fast, and a large number of samples can be homogenized in a short time, which improves work efficiency.

[0041] (3) High accuracy: The optimal number of principal components is determined through interactive verification, and Hotelling's T2 value is calculated using PCA dimensionality reduction results. This method can accurately capture important information in spectral data and effectively distinguish between homogenized and non-homogeneous samples.

[0042] (4) High reliability: By setting a confidence line, abnormal samples can be identified and eliminated, ensuring the reliability and accuracy of the model. At the same time, the validity of the validation set can ensure the accuracy and reliability of the model in practical applications.

[0043] (5) High flexibility: The scheme can set homogenization indicators according to different production needs, and is applicable to the homogenization judgment of medicinal materials from different batches and sources. At the same time, for abnormal situations judged by the model, it can be corrected and optimized by re-screening the training set and rebuilding the model, thus ensuring the flexibility and adaptability of the scheme.

[0044] (6) High repeatability: Spectral measurements and model predictions are performed under the same environment, instruments and test conditions, ensuring the repeatability and consistency of the results, making the scheme more reliable in practical applications. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the embodiments will be briefly described below. Referring to the accompanying drawings will provide a clearer understanding of the features and advantages of the present invention. The drawings are illustrative and should not be construed as limiting the present invention in any way. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort. Wherein:

[0046] Figure 1 This is a flowchart of a method for determining the quality uniformity of Epimedium medicinal material provided in the embodiments;

[0047] Figure 2 The spectrum collected in the example is shown below;

[0048] Figure 3 This is a variance contribution plot for the example;

[0049] Figure 4 This is a schematic diagram of the 95% confidence line results in the example;

[0050] Figure 5 For the example Hotelling's T 2 A diagram illustrating the results of the values;

[0051] Figure 6 This is a block diagram of a system for determining the quality uniformity of Epimedium medicinal material, as provided in the embodiments. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Example 1

[0054] like Figure 1 As shown in the embodiment of the present invention, a method for determining the quality uniformity of Epimedium medicinal material is proposed, the method comprising:

[0055] Step S1: Use near-infrared technology to collect spectral data of Epimedium medicinal material and simultaneously measure the corresponding physicochemical values. Collect multiple groups of Epimedium samples from different production batches, define the homogenization index according to production needs, and obtain training set and validation set.

[0056] Step S2: Preprocess the spectral data in the training set;

[0057] Step S3: Use the preprocessed spectral data to build a homogenization model and determine the optimal number of principal components through interactive validation in order to capture the most important information in the spectral data;

[0058] Step S4: Using the PCA dimensionality reduction results, calculate the Hotelling's T2 value of the sample, set a confidence line to identify non-homogeneous samples, and remove abnormal samples whose Hotelling's T2 value exceeds the confidence line;

[0059] Step S5: Use the established homogenization model to predict the validation set samples, calculate the Hotelling's T2 value of each sample in the validation set, and draw the T-squared test chart to check whether the model can successfully identify non-homogeneous samples. If there are misjudgments, check whether there are any abnormalities in the spectral acquisition and physicochemical testing process of the validation set. If there are no abnormalities, re-select the training set, reconstruct the homogenization model, and perform validity verification again.

[0060] As can be seen from the above technical solution, this invention provides a method for determining the quality uniformity of Epimedium medicinal materials. First, by collecting multiple groups of Epimedium samples from different production batches and clarifying the uniformity index, the diversity and relevance of the model training data are ensured, improving the practical application value of the model. Second, preprocessing the spectral data effectively reduces the influence of noise and interference factors, laying the foundation for establishing a high-quality uniformity model. Using PCA technology to establish a uniformity model and determining the optimal number of principal components through interactive validation not only reduces the data dimensionality but also captures key information in the spectral data, enabling the model to quickly and accurately predict sample uniformity. Simultaneously, calculating Hotelling's T2 value and setting a confidence line provides a quantitative indicator for identifying non-uniform samples, improving identification efficiency and accuracy. The model validation and optimization stages further ensure the reliability and stability of the model. Validating the model performance using a validation set allows for timely identification and improvement of potential problems, enhancing the model's generalization ability. Furthermore, the model reconstruction and optimization are based on actual feedback, making the model more aligned with actual production needs and increasing its practical value. Finally, maintaining the same environment and conditions during the actual sample measurement and prediction stages ensured the accuracy and comparability of the measurement results. Using the established homogenization model for prediction enabled the rapid and accurate identification of non-homogeneous samples, providing strong support for quality control during the production process. The optimization of the entire process not only improved the efficiency and accuracy of the evaluation but also enhanced the robustness and practicality of the model, providing reliable technical support for the homogenization evaluation of Epimedium medicinal materials.

[0061] In this embodiment, in step S1, near-infrared technology is used to collect spectral data of Epimedium medicinal material and simultaneously measure the corresponding physicochemical values. Multiple groups of Epimedium samples from different production batches are collected. According to production needs, homogenization indicators are defined, and training sets and validation sets are obtained.

[0062] Specifically, from the collected samples, standard samples covering the range of the homogenization index are selected as the training set, while ensuring that the training set data has sufficient variability and representativeness; samples within the range of the homogenization index and non-homogeneous samples outside the range are selected as the validation set to comprehensively evaluate the performance of the model.

[0063] In this embodiment, in step S2, the spectral data in the training set is preprocessed.

[0064] Specifically, the training set spectral data undergoes a detailed examination to identify and eliminate potential interference factors such as noise and baseline drift. Appropriate preprocessing methods (such as smoothing, derivative calculation, and normalization) are applied to improve the quality and stability of the spectral data.

[0065] In this embodiment, in step S3, a homogenization model is established using the preprocessed spectral data, and the optimal number of principal components is determined through interactive validation to maximize the model's predictive power and generalization performance. This ensures that the homogenization model can effectively capture key information and variations in the spectral data.

[0066] In this embodiment, in step S4, the Hotelling's T2 value of the sample is calculated using the PCA dimensionality reduction result, and a confidence line is set to identify non-homogeneous samples and remove abnormal samples whose Hotelling's T2 value exceeds the confidence line.

[0067] The Hotelling's T of the sample is calculated. 2 Value method:

[0068] T 2 =X T PΛ -1 P T X

[0069] in

[0070] Λ=diag{λ1,…λ k}

[0071] In the formula, T 2 Hotelling's T 2 Value; X represents the covariance matrix; P represents the characteristic matrix; λ k These are the eigenvalues ​​of the X covariance matrix.

[0072] The formula for calculating the confidence line is as follows:

[0073]

[0074] In the formula, T α The confidence curve is given by k, which represents the dimension after PCA dimensionality reduction; n represents the sample size, i.e., the product of the number of collection batches and the number of cutoff time points in each batch; F α (k,nk) represents an F-distribution with k as the first degree of freedom and nk as the second degree of freedom; P{F(n1,n2)>F α (n1,n2)}=α, where 1-α is the confidence interval and α is 0.05.

[0075] In this embodiment, in step S5, the established homogenized model is used to predict the validation set samples, and Hotelling's T for each sample in the validation set is calculated. 2The values ​​are calculated, and a T-square test plot is plotted to check whether the model can successfully identify non-homogeneous samples. If there are misjudgments, check whether there are any abnormalities in the spectral acquisition and physicochemical testing process of the validation set. If there are no abnormalities, the training set is re-selected, the homogenized model is reconstructed, and the validity is verified again.

[0076] Furthermore, the present invention also includes step S6: under the same environment, the same instruments and experimental conditions, measuring the spectrum of the sample to be homogenized, calling the established homogenization model, and calculating the Hotelling's T of the spectrum of the sample to be homogenized. 2 Value, according to Hotelling's T 2 By comparing the values ​​with the confidence line, non-homogeneous samples can be identified.

[0077] To further verify the advantages of the method, the following section combines... Figures 2 to 5 To elaborate.

[0078] First, near-infrared spectral data of Epimedium was collected using near-infrared spectroscopy. 120 Epimedium samples from different production batches were collected. Moisture content was used as a homogenization index. Spectra were acquired using a Thermo Fisher Scientific near-infrared spectrometer, with each sample scanned three times and the average value taken. The moisture content was then determined using a drying method. The collected spectra are shown below. Figure 2 As shown.

[0079] Then, based on the physicochemical test results and production requirements, the moisture content range was set to 8.8%-11.73%. From 120 samples, 75 samples meeting the moisture content index were selected as the training set, and 30 samples outside the training set were selected as the validation set. In this case, the SNV preprocessing method was used to process the spectra. A PCA (homogenization model) was established using the preprocessed training set spectra. Based on the cross-validation results, the number of principal components was set to 4, and the variance contribution was as follows: Figure 3 As shown.

[0080] Next, Hotelling's T is calculated based on the PCA modeling results score. 2 The value and 95% confidence line, the 95% confidence line is 11.697, the result is as follows. Figure 4 As shown.

[0081] Next, the normalization model is used to predict the spectrum of the validation set, and Hotelling's T is calculated. 2 Value, result as Figure 5 As shown.

[0082] Finally, predictions were made based on the established model, and non-homogeneous samples were successfully detected by using the set confidence line. Comparison of the predicted and actual results showed complete consistency with each other, with no misclassification. Therefore, this method can effectively verify the homogeneity of Epimedium samples.

[0083] Example 2

[0084] like Figure 6 As shown, this invention provides a system for determining the quality uniformity of Epimedium medicinal materials. This system is used to implement the method for determining the quality uniformity of Epimedium medicinal materials in Embodiment 1 above, specifically including:

[0085] The data acquisition module 100 is used to acquire spectral data of Epimedium medicinal material using near-infrared technology, and simultaneously measure the corresponding physicochemical values. It collects multiple groups of Epimedium samples from different production batches, defines homogenization indicators according to production needs, and obtains training and validation sets.

[0086] The data preprocessing module 200 is used to preprocess the spectral data in the training set;

[0087] The homogenization model building module 300 is used to build a homogenization model using preprocessed spectral data and determine the optimal number of principal components through interactive verification in order to capture the most important information in the spectral data.

[0088] The calculation and elimination module 400 is used to calculate the Hotelling's T2 value of the sample using the PCA dimensionality reduction result, set the confidence line, and identify non-homogeneous samples to eliminate abnormal samples whose Hotelling's T2 value exceeds the confidence line.

[0089] The validation module 500 is used to predict validation set samples using the established homogenization model, calculate Hotelling's T2 value for each sample in the validation set, and plot the T-squared test chart to check whether the model can successfully identify non-homogeneous samples. If there are misjudgments, it checks whether there are any abnormalities in the spectral acquisition and physicochemical testing process of the validation set. If there are no abnormalities, it re-selects the training set, reconstructs the homogenization model, and performs validity validation again.

[0090] In addition, the system also includes an application module 600, which is used to measure the spectrum of the sample to be homogenized under the same environment, the same instruments and test conditions, call the established homogenization model, calculate the Hotelling's T2 value of the spectrum of the sample to be homogenized, and identify non-homogenized samples by comparing the Hotelling's T2 value with the confidence line.

[0091] This embodiment provides a system for determining the quality uniformity of Epimedium medicinal materials, used to implement the aforementioned method for determining the quality uniformity of Epimedium medicinal materials. Therefore, the specific implementation of this system can be found in the embodiments section of the Epimedium medicinal material quality uniformity determination method described above. For example, the data acquisition module 100, data preprocessing module 200, uniformization model establishment module 300, calculation and elimination module 400, verification module 500, and application module 600 are respectively used to implement steps S1, S2, S3, S4, S5, and S6 in the aforementioned Epimedium medicinal material quality uniformity determination method. Therefore, its specific implementation can be referred to the descriptions of the corresponding embodiments. To avoid redundancy, further details are omitted here.

[0092] Example 3

[0093] This invention also provides an electronic device, which includes a processor, a memory, and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the above-described method for determining the quality uniformity of Epimedium medicinal materials.

[0094] Example 4

[0095] This invention also provides a computer storage medium storing a computer software product, the computer software product including several instructions for causing a computer device to execute the above-described method for determining the quality uniformity of Epimedium medicinal materials.

[0096] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0097] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0098] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0099] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for determining the quality uniformity of Epimedium medicinal material, characterized in that, include: Step S1: Use near-infrared technology to collect spectral data of Epimedium medicinal material and simultaneously measure the corresponding physicochemical values. Collect multiple groups of Epimedium samples from different production batches, define the homogenization index according to production needs, and obtain training set and validation set. Step S2: Preprocess the spectral data in the training set; Step S3: Use the preprocessed spectral data to build a homogenization model and determine the optimal number of principal components through interactive validation in order to capture the most important information in the spectral data; Step S4: Using the PCA dimensionality reduction results, calculate the Hotelling's T2 value of the sample, set a confidence line to identify non-homogeneous samples, and remove abnormal samples whose Hotelling's T2 value exceeds the confidence line; Step S5: Use the established homogenization model to predict the validation set samples, calculate the Hotelling's T2 value of each sample in the validation set, and draw the T-squared test plot to check whether the model can successfully identify non-homogeneous samples. If there are misjudgments, check whether there are any abnormalities in the validation set spectral acquisition and physicochemical testing process. If there are no abnormalities, re-select the training set, reconstruct the homogenization model, and perform validity verification again.

2. The method for determining the quality uniformity of Epimedium medicinal material according to claim 1, characterized in that, The method further includes: Step S6: Under the same environment, instruments and test conditions, measure the spectrum of the sample to be homogenized, call the established homogenization model, calculate the Hotelling's T2 value of the spectrum of the sample to be homogenized, and identify non-homogenized samples by comparing the Hotelling's T2 value with the confidence line.

3. The method for determining the quality uniformity of Epimedium medicinal material according to claim 1, characterized in that, The method for obtaining the training set and validation set is as follows: From the collected samples, standard samples covering the range of homogenization indicators are selected as the training set, while ensuring that the training set data has sufficient variability and representativeness. The validation set is selected from samples within the homogenization index range and non-homogeneous samples outside the range to comprehensively evaluate the model's performance.

4. The method for determining the quality uniformity of Epimedium medicinal material according to claim 1, characterized in that, The method for preprocessing the spectral data in the training set includes at least smoothing, derivative calculation, and normalization.

5. The method for determining the quality uniformity of Epimedium medicinal material according to claim 1, characterized in that, The calculated sample's Hotelling's T 2 The method for setting the value is as follows: T 2 =X T PΛ -1 P T X in Λ=diag{λ1,…λ k } In the formula, T 2 Hotelling's T 2 Value; X represents the covariance matrix; P represents the characteristic matrix; λ k These are the eigenvalues ​​of the X covariance matrix.

6. The method for determining the quality uniformity of Epimedium medicinal material according to claim 1, characterized in that, The formula for calculating the confidence line is as follows: In the formula, T α The confidence curve is given by k, which represents the dimension after PCA dimensionality reduction; n represents the sample size, i.e., the product of the number of collection batches and the number of cutoff time points in each batch; F α (k,nk) represents an F-distribution with k as the first degree of freedom and nk as the second degree of freedom; P{F(n1,n2)>F α (n1,n2)}=α, where 1-α is the confidence interval and α is 0.

05.

7. A system for determining the quality uniformity of Epimedium medicinal material, characterized in that, The system is used to implement the method for determining the quality uniformity of Epimedium medicinal materials according to any one of claims 1 to 6, specifically including: The data acquisition module is used to acquire spectral data of Epimedium medicinal material using near-infrared technology, and simultaneously measure the corresponding physicochemical values. It collects multiple groups of Epimedium samples from different production batches, defines homogenization indicators according to production needs, and obtains training and validation sets. The data preprocessing module is used to preprocess the spectral data in the training set; The homogenization model building module is used to build a homogenization model using preprocessed spectral data and determine the optimal number of principal components through interactive verification in order to capture the most important information in the spectral data. The calculation and elimination module is used to calculate the Hotelling's T2 value of the sample using the PCA dimensionality reduction results, set the confidence line, and identify non-homogeneous samples to eliminate abnormal samples whose Hotelling's T2 value exceeds the confidence line. The validation module is used to predict validation set samples using the established homogenized model, calculate Hotelling's T2 value for each sample in the validation set, and plot the T-squared test chart to check whether the model can successfully identify non-homogeneous samples. If there are misclassifications, it checks whether there are any abnormalities in the spectral acquisition and physicochemical testing process of the validation set. If there are no abnormalities, it re-selects the training set, reconstructs the homogenized model, and performs validity validation again.

8. The system for determining the quality uniformity of Epimedium medicinal materials according to claim 7, characterized in that, The system also includes: The application module is used to measure the spectrum of the sample to be homogenized under the same environment, instruments and test conditions, call the established homogenization model, calculate the Hotelling's T2 value of the spectrum of the sample to be homogenized, and identify non-homogenized samples by comparing the Hotelling's T2 value with the confidence line.

9. An electronic device, characterized in that, The electronic device includes a processor, a memory, and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the method for determining the quality uniformity of Epimedium medicinal materials as described in any one of claims 1 to 6.

10. A computer storage medium, characterized in that, The computer storage medium stores a computer software product, which includes several instructions for causing a computer device to execute the method for determining the quality uniformity of Epimedium medicinal material as described in any one of claims 1 to 6.

Images