A method for evaluating consistency of traditional extraction method and modern extraction method of traditional Chinese medicine compound preparation

Abstract

The present application belongs to the field of evaluation method, and provides a method for evaluating the consistency of traditional extraction method and modern extraction method of traditional Chinese medicine compound preparation. The present application calculates the benchmark correlation degree based on the relative deviation value, and the value threshold of the benchmark correlation degree is in the interval [-∞, 1]. When the relative deviation value is 0, the benchmark correlation degree is 100%, at this time, the quality of the modern extraction sample and the benchmark sample is completely consistent. It can be seen that the benchmark correlation degree can more intuitively and clearly represent the quality difference between the modern extraction sample and the benchmark sample than the relative deviation value. The subjective weighting method and the objective weighting method are combined for comprehensive evaluation, which can reflect the subjective will of the decision maker and not deviate from the actual data, so that the final weighting result is more convincing. The evaluation method of the present application solves the conversion problem of the traditional extraction method and the modern extraction method of the current classic famous prescription, and ensures that the determined modern extraction method can not only meet the quality requirements of the benchmark sample, but also meet the modern process production conditions.

Description

Technical Field

[0001] This invention relates to the field of evaluation methods, and in particular to a method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations. Background Technology

[0002] The "Traditional Chinese Medicine Law of the People's Republic of China," passed in 2016, stipulates that when applying for drug approval numbers, products based on classic prescriptions can provide only non-clinical safety research data. In 2018, the State Administration of Traditional Chinese Medicine released the first batch of the "Directory of Ancient Classic Prescriptions," which includes 100 prescriptions, detailing six aspects: prescription name, source, prescription method, preparation method, usage, and dosage form. Policy encouragement and support for the research and development of ancient classic prescriptions have become a key to unlocking the treasure trove of traditional Chinese medicine, significantly shortening the new drug development cycle, reducing research costs, and lowering research risks.

[0003] According to the "Technical Guidelines for Pharmaceutical Research of Traditional Chinese Medicine Compound Preparations Managed According to the Catalogue of Ancient Classic Prescriptions (Trial Implementation)" issued by the Center for Drug Evaluation of the National Medical Products Administration in August 2021, the reference sample refers to the sample prepared by studying and following the key information of ancient classic prescriptions published by the state and the methods recorded in ancient books. The development of traditional Chinese medicine compound preparations based on classic prescriptions should adhere to the fundamental principle of respecting ancient practices. However, during large-scale production, changes in production equipment, heating methods, and the degree of heating can lead to significant variations in the samples, resulting in inconsistencies in quality between the large-scale production samples and the reference samples. Therefore, ensuring consistency in product quality between the production process and the traditional process is crucial for the successful research of traditional Chinese medicine compound preparations based on classic prescriptions.

[0004] Achieving consistent product quality between traditional and modern processes for classic Chinese herbal medicine compound preparations still faces many challenges, particularly in extraction, concentration, and drying processes, with extraction being a major hurdle. The conversion between traditional and modern extraction methods should be based on benchmark samples, ensuring that the chosen modern extraction method meets both the quality requirements of the benchmark sample and the conditions of modern production processes. Currently, commonly used methods in Chinese herbal medicine extraction research, such as orthogonal experiments, uniform design, and Box-Behnken response surface methodology, aim to maximize the extraction of components. However, benchmark samples are prepared according to key information on ancient classic prescriptions published by the state and methods recorded in ancient texts, thus embodying the efficacy and safety of these ancient prescriptions. Therefore, this approach contradicts the requirement of adhering to ancient practices in classic Chinese herbal medicine compound preparations. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations. The evaluation method provided by this invention can accurately and comprehensively evaluate modern extraction methods.

[0006] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0007] This invention provides a method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations, comprising the following steps:

[0008] (1) Extraction of traditional Chinese medicine compound preparations using modern extraction methods to obtain modern extracted samples;

[0009] (2) Determine the evaluation indicators of the reference sample and the modern extraction sample, and obtain the measurement data of each evaluation indicator of the reference sample and the modern extraction sample; the reference sample is a sample obtained by extracting traditional Chinese medicine compound preparations using traditional extraction methods.

[0010] (3) Using the measurement data of each evaluation index of the reference sample as a benchmark, calculate the relative deviation value of each evaluation index of the modern extraction sample.

[0011] (4) Calculate the benchmark correlation of each evaluation indicator based on the relative deviation value of each evaluation indicator;

[0012] (5) Using subjective weighting and objective weighting methods, based on the measurement data of each evaluation index of the modern extracted sample, determine the subjective weight coefficient and objective weight coefficient of each evaluation index; calculate the comprehensive weight coefficient according to the subjective weight coefficient and objective weight coefficient.

[0013] (6) The benchmark correlation of each evaluation index is comprehensively scored according to the comprehensive weight coefficient to obtain the score of the modern extraction sample;

[0014] There is no specific order between (4) and (5).

[0015] Preferably, the number of evaluation indicators is ≥3; the evaluation indicators include at least one quantitative evaluation indicator and at least two evaluation indicators that can represent the overall quality attributes.

[0016] Preferably, the conventional extraction method includes processing.

[0017] Preferably, the formula for calculating the relative deviation value is as shown in Formula 1:

[0018] Rd i,j =|X i,j -S j | / S j Formula 1;

[0019] In Formula 1, X i,jS represents the measured value of the j-th (j=1,2,...,n) evaluation index of the i-th (i=1,2,...,m) sample from modern extraction samples; j Rd represents the measured value of the j-th (j = 1, 2, ..., n) evaluation index of the reference sample. i,j This represents the relative deviation value of the j-th (j=1,2,...,n) evaluation index for the i-th (i=1,2,...,m) sample.

[0020] Preferably, the formula for calculating the benchmark correlation degree is as shown in Formula 2:

[0021] SR i,j =(1-Rd) i,j )×100% Formula 2;

[0022] In Formula 2, SR i,j It represents the baseline correlation degree under the evaluation of the j-th indicator (j = 1, 2, ..., m) of the i-th (i = 1, 2, ..., m)-th sample.

[0023] Preferably, the subjective weighting method includes the analytic hierarchy process, the least squares method, the G1 method, the expert survey method, the binomial coefficient method, or the chain ratio scoring method.

[0024] Preferably, the objective weighting method includes the information entropy weighting method, principal component analysis method, deviation and mean square error method, or multi-objective programming method.

[0025] Preferably, the subjective weighting method is the analytic hierarchy process (AHP); the objective weighting method is the information entropy weighting method.

[0026] Preferably, the formula for calculating the comprehensive weighting coefficient is shown in Formula 3:

[0027]

[0028] In formula 3, W j o W represents the objective weight coefficient for the j-th (j = 1, 2, ..., n) evaluation index. j s This represents the subjective weight coefficient for the j-th (j = 1, 2, ..., n) evaluation index.

[0029] Preferably, the formula for calculating the comprehensive score is as shown in Formula 4:

[0030]

[0031] In Formula 4, SR i,j W represents the baseline correlation degree under the j-th (j=1,2,...,n) evaluation index for the i-th (i=1,2,...,m) sample; jIt is the comprehensive weight coefficient under the j-th (j=1,2,……,n) evaluation index.

[0032] This invention provides a method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations, comprising the following steps: (1) extracting the traditional Chinese medicine compound preparation using modern extraction methods to obtain a modern extraction sample; (2) determining the evaluation indicators of the benchmark sample and the modern extraction sample, and obtaining the measurement data of each evaluation indicator of the benchmark sample and the modern extraction sample; the benchmark sample is a sample obtained by extracting the traditional Chinese medicine compound preparation using traditional extraction methods; (3) using the measurement data of each evaluation indicator of the benchmark sample as a benchmark, calculating the relative deviation value of each evaluation indicator of the modern extraction sample; (4) calculating the benchmark correlation degree of each evaluation indicator based on the relative deviation value of each evaluation indicator; (5) using subjective weighting method and objective weighting method, determining the subjective weight coefficient and objective weight coefficient of each evaluation indicator based on the measurement data of each evaluation indicator of the modern extraction sample; calculating the comprehensive weight coefficient based on the subjective weight coefficient and objective weight coefficient; (6) comprehensively scoring the benchmark correlation degree of each evaluation indicator based on the comprehensive weight coefficient to obtain the score of the modern extraction sample; there is no order between (4) and (5).

[0033] The quality specification for the reference sample is not a fixed value, but a range, specifically between 70% and 130% of the reference sample's quality, which meets the quality standards of classic formulas. Using this as a benchmark, research on modern extraction methods compares the quality range of samples from modern extraction methods with that of the reference sample to examine the consistency between the two. The evaluation method of this invention calculates the benchmark correlation degree based on the relative deviation value. The threshold value of the benchmark correlation degree is in the range [-∞, 1]. When the relative deviation is 0, the benchmark correlation degree is 100%, indicating that the quality of the modern extraction sample is completely consistent with the reference sample. Therefore, the benchmark correlation degree more intuitively and clearly represents the quality difference between the modern extraction sample and the reference sample than the relative deviation value. A benchmark correlation degree greater than 70% indicates that the modern extraction process sample meets the quality requirements of the reference sample. Using a single value, rather than a confidence interval, to represent the quality consistency between the modern extraction sample and the reference sample is beneficial for using a comprehensive weighting method to screen the parameters of modern extraction methods for classic formulas. This invention employs a comprehensive evaluation method combining subjective and objective weighting approaches. This approach reflects the decision-maker's subjective intentions while remaining grounded in actual data, making the final weighting results more convincing. The evaluation method provided by this invention solves the problem of converting traditional extraction methods from classic formulas to modern extraction methods, ensuring that the determined modern extraction method meets both the quality requirements of the reference sample and complies with modern production processes.

[0034] Furthermore, the present invention uses ≥3 evaluation indicators, including at least one quantitative evaluation indicator and at least two evaluation indicators that can represent the overall quality attributes, which can enrich the information contained in the evaluation results and make them more credible; it avoids the problem of evaluation bias caused by single indicators or incomplete indicator selection.

[0035] Furthermore, the Analytic Hierarchy Process (AHP) is a subjective weighting method, where the importance of evaluation indicators is determined manually. However, different people may have different opinions, so the influence of human factors is significant. The information entropy weighting method is an objective weighting method, which derives the final weight coefficients based on the changing patterns of data, but it cannot reflect the principal, assistant, adjuvant, and guide relationships in traditional Chinese medicine. The evaluation method of this invention uses the AHP-entropy weighting method, which combines subjective and objective weighting methods to comprehensively determine the weight coefficients, resulting in data that is systematic and comprehensive. Detailed Implementation

[0036] This invention provides a method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations, comprising the following steps:

[0037] A method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations includes the following steps:

[0038] (1) Extraction of traditional Chinese medicine compound preparations using modern extraction methods to obtain modern extracted samples;

[0039] (2) Determine the evaluation indicators of the reference sample and the modern extraction sample, and obtain the measurement data of each evaluation indicator of the reference sample and the modern extraction sample; the reference sample is a sample obtained by extracting traditional Chinese medicine compound preparations using traditional extraction methods.

[0040] (3) Using the measurement data of each evaluation index of the reference sample as a benchmark, calculate the relative deviation value of each evaluation index of the modern extraction sample.

[0041] (4) Calculate the benchmark correlation of each evaluation indicator based on the relative deviation value of each evaluation indicator;

[0042] (5) Using subjective weighting and objective weighting methods, based on the measurement data of each evaluation index of the modern extracted sample, determine the subjective weight coefficient and objective weight coefficient of each evaluation index; calculate the comprehensive weight coefficient according to the subjective weight coefficient and objective weight coefficient.

[0043] (6) The benchmark correlation of each evaluation index is comprehensively scored according to the comprehensive weight coefficient to obtain the score of the modern extraction sample;

[0044] There is no specific order between (4) and (5).

[0045] Unless otherwise specified, all raw materials used in this invention are preferably commercially available products.

[0046] This invention utilizes modern extraction methods to extract traditional Chinese medicine compound preparations, obtaining modern extracted samples.

[0047] In this invention, the modern extraction method is obtained by using single-factor experimental design, orthogonal experimental design, uniform design, star point design, Doehlert design, or Box-Behnken experimental design for traditional Chinese medicine compound preparations. In this invention, the preferred number of modern extraction samples is m.

[0048] In this invention, the traditional Chinese medicine compound preparation preferably includes classic prescriptions; the classic prescriptions are those recorded in ancient Chinese medicine classics that comply with the provisions of the "Law of the People's Republic of China on Traditional Chinese Medicine" and are still widely used, have definite curative effects, and have obvious characteristics and advantages.

[0049] After obtaining the modern extraction sample, the present invention determines the evaluation index of the reference sample and the modern extraction sample, and obtains the measurement data of the evaluation index of the reference sample and the modern extraction sample; the reference sample is a sample obtained by extracting traditional Chinese medicine compound preparations using traditional extraction methods.

[0050] In this invention, the traditional extraction method preferably includes processing. In a specific embodiment of this invention, the traditional extraction method is preferably the method disclosed in the "Catalogue of Famous Ancient Prescriptions (First Batch)".

[0051] In this invention, the number of evaluation indicators is preferably ≥3; the evaluation indicators preferably include at least one quantitative evaluation indicator and at least two evaluation indicators that can represent the overall quality attributes. In this invention, the number of evaluation indicators is preferably n.

[0052] After determining the evaluation indicators for the reference sample and the modern extraction sample, the present invention obtains the measurement data of the evaluation indicators for the reference sample and the modern extraction sample.

[0053] The present invention does not specifically limit the method for obtaining the measurement data of the evaluation indicators of the reference sample and the modern extraction sample; any measurement method well known to those skilled in the art can be used.

[0054] After obtaining the measurement data of the evaluation indicators of the reference sample and the modern extraction sample, this invention uses the measurement data of each evaluation indicator of the reference sample as a benchmark to calculate the relative deviation value of each evaluation indicator of the modern extraction sample.

[0055] In this invention, the formula for calculating the relative deviation value is shown in Formula 1:

[0056] Rdi,j =|X i,j -S j | / S j Formula 1;

[0057] In Formula 1, X i,j S represents the measured value of the j-th (j=1,2,...,n) evaluation index of the i-th (i=1,2,...,m) sample from modern extraction samples; j Rd represents the measured value of the j-th (j = 1, 2, ..., n) evaluation index of the reference sample. i,j This represents the relative deviation value of the j-th (j=1,2,...,n) evaluation index for the i-th (i=1,2,...,m) sample.

[0058] After obtaining the relative deviation values ​​of each evaluation index of the modern extracted sample, the present invention calculates the benchmark correlation degree of each evaluation index based on the relative deviation values ​​of each evaluation index.

[0059] In this invention, the benchmark correlation degree is defined as follows: the measured data of each evaluation index of the benchmark sample obtained by the traditional extraction method is the standard value (S), the measured data of each evaluation index of the modern extraction sample obtained by the modern extraction method is the measured value (X), and the value calculated by formula 2 using multiple indicators as evaluation indicators is called the benchmark correlation degree.

[0060] In this invention, the formula for calculating the benchmark correlation degree is shown in Formula 2:

[0061] SR i,j =(1-Rd) i,j )×100% Formula 2;

[0062] In Formula 2, SR i,j It represents the baseline correlation degree of the i-th (i = 1, 2, ..., m) sample under the j-th (j = 1, 2, ..., n) evaluation index.

[0063] After obtaining the baseline correlation of each evaluation index, this invention uses subjective weighting method and objective weighting method to determine the subjective weight coefficient and objective weight coefficient of each evaluation index based on the measurement data of each evaluation index of the modern extracted sample; and calculates the comprehensive weight coefficient according to the subjective weight coefficient and objective weight coefficient.

[0064] In this invention, the subjective weighting method preferably includes the analytic hierarchy process (AHP), the least squares method, the G1 method, the expert survey method, the binomial coefficient method, or the chain ratio scoring method, and is more preferably the AHP. In this invention, when using the AHP to determine the subjective weight coefficients of each evaluation index, the principle of the principal, assistant, adjuvant, and guide herbs in traditional Chinese medicine is preferably used for scoring.

[0065] In this invention, the objective weighting method preferably includes the information entropy weighting method, principal component analysis method, deviation and mean square error method or multi-objective programming method, and is more preferably the information entropy weighting method.

[0066] In this invention, the formula for calculating the comprehensive weighting coefficient is shown in Formula 3:

[0067]

[0068] In formula 3, W j W is the comprehensive weight coefficient under the j-th (j = 1, 2, ..., n) evaluation index. j o W represents the objective weight coefficient for the j-th (j = 1, 2, ..., n) evaluation index. j s This represents the subjective weight coefficient for the j-th (j = 1, 2, ..., n) evaluation index.

[0069] After obtaining the comprehensive weighting coefficient, the present invention performs a comprehensive score on the benchmark correlation of each evaluation index based on the comprehensive weighting coefficient to obtain the score of the modern extraction sample.

[0070] In this invention, the formula for calculating the comprehensive score is shown in Formula 4:

[0071]

[0072] In Formula 4, SR i,j W represents the baseline correlation degree under the j-th (j=1,2,...,n) evaluation index for the i-th (i=1,2,...,m) sample; j It is the comprehensive weight coefficient under the j-th (j=1,2,……,n) evaluation index.

[0073] In this invention, the higher the score of the modern extraction sample, the closer the composition of the modern extraction sample is to the reference sample, which in turn indicates that the modern extraction method for preparing the modern extraction sample is better and can be used as a modern extraction method.

[0074] The following detailed description, in conjunction with embodiments, illustrates the method provided by this invention for evaluating the consistency between traditional and modern extraction methods of traditional Chinese medicine compound preparations. However, these descriptions should not be construed as limiting the scope of protection of this invention.

[0075] Example 1: Consistency Evaluation of Traditional and Modern Extraction Methods for Xiaoxuming Decoction

[0076] 1. Determine the evaluation object

[0077] This embodiment uses the Xiaoxuming Tang (XXMT) recorded in the "Catalogue of Famous Ancient Prescriptions (First Batch)" as an example. This prescription comes from "Essential Prescriptions Worth a Thousand Pieces of Gold for Emergency Use" (by Sun Simiao of the Tang Dynasty). It consists of one liang each of ephedra, fangji, ginseng, scutellaria, cinnamon bark, licorice, peony root, chuanxiong rhizome, and apricot kernel, one aconite root, one and a half liang of saposhnikovia root, and five liang of fresh ginger. The above twelve ingredients are as follows: The herbs are prepared by first boiling the ephedra in 12 liters of water until it boils three times, skimming off the foam, then adding the other herbs and boiling until 3 liters remain. Based on historical medical records and following key information from ancient classic prescriptions published by the state and ancient texts, XXMT reference samples and modern extraction samples were prepared after determining the medicinal material origin, medicinal part, processing specifications, and calculated dosage. The specific methods are as follows:

[0078] (1) Preparation of reference sample (traditional extraction method): Weigh ephedra, fangji, ginseng, scutellaria, cinnamon, licorice, white peony, chuanxiong, apricot kernel, aconite, saposhnikovia, and ginger according to the dosage of the ancient prescription, and then crush them into particles with a particle size of 5-6 mm. Place the ephedra particles in a ceramic decoction pot, add water, and boil over high heat. Then add the remaining medicinal materials, boil over high heat until boiling, then reduce to low heat and keep boiling until the decoction volume is appropriate. Filter while hot to obtain the XXMT reference sample.

[0079] (2) Preparation of orthogonal experimental samples (modern extraction method): Weigh the same batch of medicinal materials with the same prescription amount as the traditional process, and take the material-liquid ratio (A), extraction time (B), and extraction times (C) as the three factors to be investigated. Extract according to Table 1 to obtain m modern extraction samples.

[0080] Table 1 L9(3) of XXMT 4 Orthogonal experimental table

[0081]

[0082]

[0083] 2. Determine evaluation indicators

[0084] The current idea of traditional Chinese medicine quality control is to control the amount of one or several active ingredients or characteristic ingredients in traditional Chinese medicine. Therefore, when conducting the determination of chemical components in this study, substances that are closely related to the functional properties of traditional Chinese medicine, have clear chemical structures, can be qualitatively identified and quantitatively determined, and representative substances of monarch drugs in prescriptions composed according to traditional Chinese medicine compatibility principles, taking into account the representative substances of ministerial, adjuvant, and guiding drugs should be selected for determination. To ensure the authenticity and reliability of the evaluation results, at least one chemical component in the compound prescription should be determined, so the quantitative index should be at least 1. At the same time, fingerprints, extract yield, extractives, etc. that can represent the overall quality should be used as evaluation indicators. The evaluation indicators representing the overall quality should be no less than 2. Therefore, the total number of evaluation indicators n should be at least 3. The screening of evaluation indicators can refer to the document "Technical Guidelines for the Research of Quality Standards for New Traditional Chinese Medicine Drugs (Trial)".

[0085] In this example, the determination data of n (n = 6) evaluation indicators including pseudoephedrine hydrochloride, ginsenoside Rg1, baicalin, paeoniflorin, fingerprint similarity, and extract yield in XXMT orthogonal test samples and reference samples were obtained, as shown in Table 2 for details.

[0086] Table 2 Example of XXMT original data

[0087]

[0088]

[0089] 3. Calculate the relative deviation value

[0090] The basic calculation principle is based on the requirements of the document "Technical Guidelines for the Pharmaceutical Research of Traditional Chinese Medicine Compound Preparations Managed According to the Catalog of Ancient Classic Famous Formulas (Trial)". The content fluctuation range of the index components within 70% - 130% of the mean value of the reference sample meets the quality standards. First, the number of evaluation indicators n of the sample and the number of samples m were determined. Therefore, in this example, X was calculated through the following formula i,j Relative to S j The relative deviation value (Rd i,j ) more scientifically and comprehensively analyzes the similarity between different samples and the reference sample.

[0091] Rd i,j = |X i,j - S j | / S j

[0092] Rd i,j represents the relative deviation value of the i (i = 1, 2,..., m) - th sample under the j (j = 1, 2,..., n) - th evaluation indicator. The calculation results are shown in Table 3. The smaller Rd i,j , the more it represents that X i,j is relative to S jThe smaller the deviation, the higher the similarity between the sample and the reference sample.

[0093] Table 3. Relative deviation values ​​of the XXMT orthogonal experiment

[0094]

[0095] 4. Calculate the baseline correlation degree

[0096] Since relative deviation values ​​cannot be used for comprehensive scoring, and their meaning is not intuitive enough to facilitate understanding and application, a further method was adopted to assign an evaluation value, namely the benchmark correlation degree, to each evaluation object. This method calculates the benchmark correlation degree (SR) based on the relative deviation values ​​between each sample in the orthogonal experiment and the benchmark sample under each index in Table 3, using the following formula. i,j That's all.

[0097] SR i,j =(1-Rd) i,j )×100%

[0098] SR i,j This represents the baseline correlation degree under the j-th (j=1,2,…,n) evaluation index for the i-th (i=1,2,…,m) sample. The closer the baseline correlation degree is to 100%, the higher the similarity between the sample obtained by the modern extraction method and the baseline sample under that evaluation index. The calculation results are shown in Table 4. Therefore, the next step is to use the analytic hierarchy process (AHP)-information entropy weight method to determine the weight coefficients of each index and perform a comprehensive score.

[0099] SR i,j =(1-|X i,j -S j | / S j )×100%

[0100] Table 4. Benchmark Correlation Degree of XXMT Orthogonal Experiment

[0101]

[0102] As can be seen from Table 4: SR i,j The closer the similarity is to 100%, the higher the similarity between the modern extracted sample and the reference sample. The results showed that: when using the content of pseudoephedrine hydrochloride and paeoniflorin as evaluation indicators, test number 8 was similar to the reference sample; when using baicalin and fingerprint spectrum similarity as evaluation indicators, test number 1 was similar to the reference sample; and when using ginsenoside Rg1 and extract yield as evaluation indicators, test number 6 was similar to the reference sample. Therefore, it is evident that a weighted method must be used for comprehensive scoring before determining the relevant extraction process parameters.

[0103] Calculation of combined weights using the 5-level analytic hierarchy process and the information entropy weighting method

[0104] 5.1 Subjective Weighting in the Analytic Hierarchy Process (AHP)

[0105] 5.1.1 Constructing the XXMT Judgment Matrix

[0106] The construction of this judgment matrix follows the principle of monarch, minister, assistant and guide in traditional Chinese medicine. In the XXMT formula, ephedra is the monarch herb, which has the functions of dispelling wind and cold, and inducing sweating to relieve exterior syndrome. It is combined with fangfeng, chuanxiong, cinnamon twig, fangji and aconite to dispel wind. The combined use of these herbs has a strong pungent and dispersing nature. Therefore, the bitterness of scutellaria and the sweetness of ginseng and white peony are added to prevent the pungent and dispersing herbs from damaging body fluids and depleting qi, and to replenish qi and blood, and support the body's resistance to expel pathogens. Based on the formulation of XXMT, the active ingredients of the principal herbs were chosen as the primary indicators for content determination. Considering the correlation with preparation methods and stability, quality standard control indicators were specifically selected and determined. Therefore, pseudoephedrine hydrochloride in ephedra, ginsenoside Rg1 in ginseng, baicalin in scutellaria, and paeoniflorin in peony root were measured. Fingerprint spectroscopy and yield were used as supplementary quality control methods. Although fingerprint spectroscopy can reflect the profile and qualitative characteristics of small molecule components, it contains less quality information compared to the content-determined indicators. Therefore, its importance was considered equal to that of yield. Finally, a judgment matrix was constructed based on the following order: pseudoephedrine hydrochloride > ginsenoside Rg1 > baicalin > paeoniflorin > fingerprint spectroscopy similarity ≈ yield.

[0107] Then, based on the Analytic Hierarchy Process (AHP) theory and the 1-9 scale method, the relative importance of n indicators within the same level was scored, as shown in Table 5. Pseudoephedrine hydrochloride is the main active ingredient of the principal drug in the prescription, and is relatively important compared to ginsenoside Rg1, so it is marked as 2; it is significantly more important than baicalin in Scutellaria baicalensis, so it is marked as 4; it is significantly more important than paeoniflorin in Paeonia lactiflora, so it is marked as 5; it is significantly more important than fingerprint similarity and extract yield, so it is marked as 6; and so on. The judgment matrix A is constructed according to the following formula, with data shown in Table 6, using a... ef This represents the comparison result of factor e (e = 1, 2, ..., n) with factor f (f = 1, 2, ..., n).

[0108]

[0109] Table 5. Criteria for constructing the judgment matrix using the analytic hierarchy process.

[0110]

[0111]

[0112] Table 6. Priority Matrix for Paired Comparison of XXMT Indicators

[0113]

[0114] 5.1.2 Calculate the weights

[0115] First, the geometric mean (square root method) of the judgment matrix A is applied, and the initial weight coefficients W are calculated using the following formula. j ':

[0116]

[0117] a jn This represents the comparison result of the j-th (j = 1, 2, ..., n) evaluation index with the n-th (n = 1, 2, ..., j) evaluation index.

[0118] Then calculate the normalized weighting coefficient W according to the following formula. j s The weight coefficients of the six indicators were obtained as follows: 0.420, 0.238, 0.143, 0.089, 0.055, and 0.055, respectively.

[0119]

[0120] 5.1.3 Consistency Check

[0121] After obtaining the weight coefficients, the matrix needs to be checked for consistency. The consistency index (CI) and consistency ratio (CR) should be calculated, and the largest eigenvalue λ should be calculated using the following formula. max .

[0122]

[0123] Then, CI is calculated using the following formula, resulting in a consistency index CI of 0.02 and a consistency ratio CR (CR = CI / RI, where RI is the degrees of freedom index) of 0.01, both less than 0.1, indicating that the matrix is ​​consistent. If either CI or CR is greater than 0.1, the scoring is unreasonable and needs to be recalculated.

[0124] CI=(λ max -n) / (n-1)

[0125] 5.2 Objective Weighting of Information Entropy

[0126] 5.2.1 Establishing the original data matrix for the XXMT orthogonal experiment

[0127] Given m samples and n evaluation indicators, the original data matrix R = (X... i,j ) m×n The original data matrix R was established using six indicators as evaluation criteria: the content of pseudoephedrine hydrochloride, ginsenoside Rg1, baicalin, paeoniflorin, fingerprint spectrum similarity, and extract yield.

[0128]

[0129] 5.2.2 Transform the original data into a probability matrix

[0130] The original data matrix R is transformed into a probability matrix P using the following formula, P i,j This represents the probability of the i-th (i=1,2,...,m) sample under the j-th (j=1,2,...,n) index.

[0131]

[0132]

[0133] 5.2.3 Calculate the information entropy and weighting coefficients of each indicator in XXMT.

[0134] Calculate the information entropy (H) of each indicator using the following formula. j The values ​​are 0.9799, 0.9912, 0.9919, 0.9956, 0.9690, 0.9930, and H, respectively. j The smaller the value of the j-index, the higher the dispersion of the data, and therefore the greater the amount of information it provides.

[0135]

[0136] Calculate the weight coefficient (W) of each indicator according to the following formula. j o The values ​​are 0.254, 0.110, 0.101, 0.056, 0.391, and 0.088, respectively.

[0137]

[0138] 5.2.4 Determination of XXMT Combination Weights and Comprehensive Scoring

[0139] After obtaining the subjective weight coefficients using the analytic hierarchy process and the objective weight coefficients using the information entropy weighting method, the combined weight coefficients of each indicator were calculated according to the following formula. The results are shown in Table 7.

[0140]

[0141] Table 7 XXMT Overall Weighting Coefficient

[0142]

[0143] The comprehensive weighting coefficients of pseudoephedrine hydrochloride (Q1), ginsenosides (Q2), baicalin (Q3), paeoniflorin (Q4), fingerprint similarity (Q5), and extract yield (Q6) were obtained by the analytic hierarchy process-information entropy weighting method. The comprehensive score was then calculated according to the following formula. The scoring results are shown in Table 8.

[0144]

[0145] Table 8 XXMT Comprehensive Score Results

[0146]

[0147] As shown in Table 8, Experiment 8 has the highest overall score. The results indicate that the modern process parameters most similar to the traditional extraction method are A3B2C1, which means that the sample obtained by boiling with 12 times the amount of water for 1 hour is most consistent with the quality of the reference sample.

[0148] 6. XXMT Verification Test

[0149] Based on the final determined extraction method parameters, three batches of verification experiments were conducted comparing the parameters of the modern extraction method with those of the traditional extraction method. The contents of pseudoephedrine hydrochloride, ginsenoside Rg1, baicalin, and paeoniflorin were determined, along with fingerprint similarity and extract yield. The measured values ​​of the reference samples obtained from the three traditional extraction methods were averaged (S...). i，j The results show that the overall score was calculated and the RSD value was within 5%, as shown in Table 9.

[0150] Table 9 XXMT Validation Experiment Results

[0151]

[0152] As can be seen from Table 9, the modern extraction method for XXMT obtained by the benchmark correlation degree and the analytic hierarchy process-information entropy weight method is stable and feasible, with high repeatability. The content of the index components of all samples is between 70% and 130% of the mean of the benchmark samples, and the fingerprint spectrum similarity is greater than 0.9.

[0153] Example 2: Consistency Evaluation of Ancient and Modern Extraction Processes for Dang Gui Bu Xue Tang (Angelica Blood-Nourishing Decoction)

[0154] 1. Determine the evaluation object

[0155] This embodiment uses the Dang Gui Bu Xue Tang (DBT) formula recorded in the "Catalogue of Famous Ancient Prescriptions (First Batch)" as an example. This formula originates from "Treatise on Differentiation of Internal and External Injuries" (Jin Dynasty, Li Dongyuan). The ingredients are: Astragalus membranaceus 1 liang (approximately 30g), Dang Gui (Angelica sinensis) 2 qian (approximately 6g) (washed with wine). The ingredients are coarsely ground and used as one dose. Boil two cups of water down to one cup, remove the dregs, and drink warm on an empty stomach before meals. Based on historical medical records and following the key information of ancient classic prescriptions published by the state and ancient texts, a reference sample and a modern extraction sample for DBT were prepared after determining the medicinal material origin, medicinal part, processing specifications, and calculated dosage. The specific methods are as follows:

[0156] (1) Preparation of reference sample (traditional extraction method): Weigh the ancient prescription dosage of Astragalus membranaceus and Angelica sinensis, and crush them into particles with a particle size of 5-6 mm; place them in a ceramic decoction pot, add an appropriate amount of water, bring to a boil over high heat, then simmer over low heat until the appropriate volume is reached, filter while hot, and cool to obtain the appropriate DBT reference sample.

[0157] (2) Preparation of orthogonal experimental samples (modern extraction method): Weigh the same batch of medicinal materials with the same prescription amount as the traditional extraction method, and take the material-liquid ratio (A), extraction time (B), and extraction times (C) as the three factors to be investigated. Extraction was carried out according to Table 10 to obtain m samples of modern extraction process.

[0158] Table 10 DBT L9(3) 4 Orthogonal experimental table

[0159]

[0160] 2. Determine evaluation indicators

[0161] When selecting evaluation indicators, at least one quantitative indicator should be chosen. Considering measurability, the effective components of Astragalus membranaceus (the principal ingredient) and Astragaloside A (the secondary ingredient) were selected, along with ferulic acid (the secondary ingredient) from Angelica sinensis. The evaluation was conducted using two overall quality evaluation indicators: fingerprint spectrum and extract yield.

[0162] The determination data of n indicators (n=5) of verbascoside, astragaloside A, ferulic acid, fingerprint spectrum similarity, and extract yield were obtained in DBT reference samples and orthogonal test samples. See Table 11 for details.

[0163] Table 11 Example of DBT raw data

[0164]

[0165] 3. Calculate the relative deviation value

[0166] Refer to Example 1, according to formula Rd i,j =|X i,j -S j | / S jThe relative deviation values ​​of the DBT orthogonal test samples were calculated, and the results are shown in Table 12.

[0167] Table 12 Relative Deviation Values ​​of DBT Orthogonal Experiment

[0168]

[0169]

[0170] 4. Calculate the baseline correlation degree

[0171] Refer to Example 1, and follow the formula SR i,j =(1-Rd) i,j The baseline correlation of the DBT orthogonal test samples was calculated by multiplying the result by 100%, and the results are shown in Table 13.

[0172] Table 13. Benchmark Correlation of DBT Orthogonal Experiments

[0173]

[0174] Table 13 shows that when using isoflavone glucoside as the evaluation index, test number 4 is similar to the benchmark sample; when using astragaloside A and fingerprint spectrum similarity as evaluation indexes, test number 8 is similar to the benchmark sample; when using ferulic acid as the evaluation index, test number 4 is similar to the benchmark sample; and when using extract yield as the evaluation index, test number 4 is similar to the benchmark sample. Therefore, a weighted method is needed for comprehensive scoring before the phase extraction method parameters can be determined.

[0175] Calculation of combined weights using the 5-level analytic hierarchy process and the information entropy weighting method

[0176] 5.1 Subjective Weighting in the Analytic Hierarchy Process (AHP)

[0177] A judgment matrix was constructed based on the principles of principal, assistant, adjuvant, and guide herbs in traditional Chinese medicine. In DBT, the ratio of Astragalus membranaceus to Angelica sinensis is 5:1. Astragalus membranaceus is used to tonify Qi, while Angelica sinensis is used to nourish blood. Since tangible blood cannot be generated quickly, intangible Qi should be consolidated urgently. Tangible blood is generated from intangible Qi. Therefore, tonifying Qi generates blood, so the dosage of Astragalus membranaceus is twice that of Angelica sinensis, making it the principal herb, while Angelica sinensis is the assistant herb. The original judgment matrix was constructed based on the importance of astragaloside A > verbascoside > ferulic acid > fingerprint similarity ≈ yield rate. The scoring was performed according to the method in Example 1. Verascoside and astragaloside A are the main active ingredients of Astragalus membranaceus, the principal herb in the prescription. Astragaloside A is slightly more important than verbascoside, so it is marked as 2; slightly more important than ferulic acid in Angelica sinensis, so it is marked as 3; significantly more important than fingerprint similarity and yield rate, so it is marked as 4; and so on. The judgment matrix was constructed accordingly, and the results are shown in Table 14.

[0178] Table 14 Priority Matrix for Paired Comparison of DBT Indicators

[0179]

[0180] The weight coefficients of the analytic hierarchy process are calculated according to the formula in Example 1. Calculate the subjective weighting coefficients (W) for astragaloside A, verbascoside, ferulic acid, fingerprint similarity, and extract yield. j s The values ​​were 0.414, 0.258, 0.153, 0.088, and 0.088, respectively. Finally, a consistency test was performed. Both the CI (0.009) and CR (0.008) values ​​were less than 0.1, indicating that the judgment matrix was reasonably constructed and the resulting data was authentic and reliable.

[0181] 5.2 Objective Weighting of Information Entropy

[0182] Following the method in Example 1, the original data matrix of the DBT orthogonal experiment was established. There are m samples and n evaluation indicators in the existing DBT orthogonal experiment, and the original data matrix R = (X... i,j ) m×n The original data matrix R was established using five indicators as evaluation indicators: astragaloside A, verbascoside, ferulic acid content, fingerprint spectrum similarity, and extract yield.

[0183]

[0184] The original data matrix is ​​transformed into a probability matrix P as follows:

[0185]

[0186] Refer to the formula in Example 1 The information entropy values ​​of each evaluation index of DBT were calculated to obtain the information entropy values ​​(H) of verbascoside, astragaloside A, ferulic acid, fingerprint similarity, and extract yield. j The values ​​are 0.9699, 0.9800, 0.9937, 0.9673, and 0.9859 respectively, according to the formula. Finally, the objective weighting coefficients (W) for verbascoside, astragaloside A, ferulic acid, fingerprint similarity, and extract yield were calculated. j o The values ​​are 0.194, 0.294, 0.061, 0.316, and 0.137, respectively.

[0187] 5.3 Determination of DBT Combination Weights and Comprehensive Scoring

[0188] The comprehensive weighting coefficient of DBT was calculated according to the method in Example 1, and the results are shown in Table 15.

[0189] Table 15 DBT Comprehensive Weighting Coefficient

[0190]

[0191] The comprehensive weighting coefficients of DBT astragaloside A (Q1), verbenafiloside A (Q2), ferulic acid (Q3), fingerprint spectrum similarity (Q4), and extract yield (Q5) obtained by the analytic hierarchy process-information entropy weighting method are used to calculate the comprehensive score according to the following formula. The scoring results are shown in Table 16.

[0192]

[0193] Table 16 DBT Comprehensive Score

[0194]

[0195] As shown in Table 16, Experiment 8 had the highest overall score. The results indicate that the modern process parameters most similar to the ancient method are A2B1C2, which means that the sample obtained by decocting with 8 times the amount of water for 0.5 hours and decocting twice is most consistent with the quality of the reference sample.

[0196] 6DBT Validation Trial

[0197] Based on the finalized extraction method parameters, three batches of verification experiments were conducted comparing the parameters of the modern extraction method with those of the traditional extraction method. The contents of astragaloside A, verbascoside, and ferulic acid were determined, along with fingerprint similarity and extract yield. The measured values ​​of the reference samples obtained from the three traditional extraction methods were averaged. The results show that the overall score was calculated and the RSD value was within 5%, as shown in Table 17.

[0198] Table 17 Results of DBT Verification Experiment

[0199]

[0200] As can be seen from Table 17, the DBT modern extraction process obtained based on the benchmark correlation degree and the analytic hierarchy process-information entropy weight method is stable and feasible, with high repeatability. The content of the index components of all samples is between 70% and 130% of the mean of the benchmark samples, and the fingerprint spectrum similarity is greater than 0.9.

[0201] According to ancient texts, the decoction methods for the classic formulas Xiaoxumingtang and Dangguibuxuetang require the following medicinal materials: Traditional extraction methods involve crushing raw medicinal materials into particles of a specific size, using water as the extraction solvent, and determining the extraction time based on water loss. However, modern processes often use raw medicinal materials with a closed-loop reflux extraction method, which is inconsistent with industrial production practices. Therefore, traditional extraction methods should be thoroughly compared with actual industrial production practices. The quality of a benchmark sample should be used as an indicator to comprehensively evaluate the extraction process, and the extraction time, number of extractions, and amount of water added should be examined in detail to determine modern extraction process parameters that are essentially consistent with traditional methods.

[0202] This invention introduces the concept of benchmark correlation for the first time. Following this method, data is transformed into an optimal vector for comprehensive evaluation, namely the benchmark correlation, establishing key evaluation parameters between traditional and modern extraction methods. This evaluation method is simple and easy to understand, bridging the gap between traditional and modern extraction methods and representing a crucial step in the transformation of traditional Chinese medicine compound preparations into modern production processes.

[0203] The Analytic Hierarchy Process (AHP) is a subjective weighting method that manually determines the importance of evaluation indicators. However, different people may have different opinions, so the human factor has a significant influence. The entropy weighting method is an objective weighting method that derives the final weight coefficients based on the changing patterns of data, but it cannot reflect the principal, assistant, adjuvant, and guide relationships in traditional Chinese medicine. The AHP-entropy weighting method used in this invention combines subjective and objective weighting methods to comprehensively determine the weight coefficients, resulting in data that is systematic and comprehensive.

[0204] As can be seen from the above two embodiments, the present invention has the following advantages:

[0205] 1. The benchmark correlation degree pioneered in this invention solves the problem of selecting the most similar rather than the optimal alternative when considering multiple attributes simultaneously in traditional Chinese medicine compound preparations. It overcomes the problem that traditional Chinese medicine compound preparations cannot organically combine multi-index evaluation with research and cannot assign corresponding weights and comprehensive scores to influencing factors according to their importance.

[0206] 2. Evaluation biases can arise due to the use of a single indicator or incomplete indicator selection. Evaluating something based on a single indicator is often inadequate. Therefore, it is necessary to collect and summarize relevant information based on the characteristics of the thing itself to obtain a comprehensive indicator, which can reflect whether the thing has met the expected requirements as a whole. Therefore, this invention stipulates the use of at least one quantitative evaluation indicator and two indicators that can represent the overall quality attributes for evaluation, which enriches the information contained in the evaluation results and makes them more credible.

[0207] 3. This embodiment employs a combination of subjective and objective weighting methods for comprehensive evaluation, as applying either method alone will result in some information gaps. Therefore, reducing information gaps involves organically combining the two methods, reflecting the decision-maker's subjective intentions while remaining grounded in actual data, thus making the final weighting results more convincing.

[0208] 4. The benchmark correlation degree combined with the subjective and objective weighting method can comprehensively study traditional Chinese medicine compound preparations, thus providing better ideas for their process exploration and quality evaluation.

[0209] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for evaluating the consistency between traditional and modern extraction methods for traditional Chinese medicine compound preparations, characterized in that, Includes the following steps: (1) Extraction of traditional Chinese medicine compound preparations using modern extraction methods to obtain modern extracted samples; (2) Determine the evaluation indicators of the reference sample and the modern extraction sample, and obtain the measurement data of each evaluation indicator of the reference sample and the modern extraction sample; The reference sample is a sample obtained by extracting traditional Chinese medicine compound preparations using traditional extraction methods; the number of evaluation indicators is ≥3; the evaluation indicators include at least one quantitative evaluation indicator and at least two evaluation indicators that can represent the overall quality attributes; (3) Using the measurement data of each evaluation index of the reference sample as a benchmark, calculate the relative deviation value of each evaluation index of the modern extraction sample. (4) Calculate the benchmark correlation of each evaluation indicator based on the relative deviation value of each evaluation indicator; (5) Using subjective weighting and objective weighting methods, based on the measurement data of each evaluation index of the modern extracted sample, determine the subjective weight coefficient and objective weight coefficient of each evaluation index; calculate the comprehensive weight coefficient according to the subjective weight coefficient and objective weight coefficient. (6) The benchmark correlation of each evaluation index is comprehensively scored according to the comprehensive weight coefficient to obtain the score of the modern extraction sample; There is no specific order between (4) and (5); The formula for calculating the relative deviation value is shown in Formula 1: Rd i,j =|X i,j -S j | / S j Formula 1; In Formula 1, X i,j S represents the measured value of the j-th (j=1,2,...,n) evaluation index of the i-th (i=1,2,...,m) sample from modern extraction samples; j Rd represents the measured value of the j-th (j = 1, 2, ..., n) evaluation index of the reference sample. i,j This represents the relative deviation value of the j-th (j=1,2,...,n) evaluation index for the i-th (i=1,2,...,m) sample; The formula for calculating the baseline correlation degree is shown in Formula 2: SR i,j = (1 - Rd i,j ) × 100% Formula 2; In Formula 2, SR i,j This represents the baseline correlation degree under the evaluation of the j-th indicator for the i-th (i = 1, 2, ..., m) sample; The formula for calculating the comprehensive weighting coefficient is shown in Formula 3: Official 3; In formula 3, W j o W represents the objective weight coefficient for the j-th (j = 1, 2, ..., n) evaluation index. j s This represents the subjective weight coefficient for the j-th (j = 1, 2, ..., n) evaluation indicator; The formula for calculating the comprehensive score is shown in Formula 4: Official 4; In Formula 4, SR i,j W represents the baseline correlation degree under the j-th (j=1,2,...,n) evaluation index for the i-th (i=1,2,...,m) sample; j It is the comprehensive weight coefficient under the j-th (j=1,2,……,n) evaluation index.

2. The method according to claim 1, characterized in that, The subjective weighting methods include the analytic hierarchy process, the least squares method, the G1 method, the expert survey method, the binomial coefficient method, or the chain ratio scoring method.

3. The method according to claim 1, characterized in that, The objective weighting methods include information entropy weighting, principal component analysis, deviation and mean square error method, or multi-objective programming.

4. The method according to claim 1, 2, or 3, characterized in that, The subjective weighting method is the analytic hierarchy process (AHP); the objective weighting method is the information entropy weighting method.

Images