Medicine quality detection method and system based on component analysis

Abstract

The invention provides a medicine quality detection method and system based on component analysis, and the method comprises the steps: S100, carrying out the pretreatment of a to-be-detected medicine sample, and obtaining the pretreated to-be-detected medicine sample; s200, detecting the pretreated to-be-detected drug sample by using a gas chromatograph to obtain a first chromatogram of the to-be-detected drug sample; s300, carrying out data processing on the first chromatogram, and identifying the target component in the to-be-detected drug sample according to the first chromatogram after data processing, through a highly automatic operation process and an accurate data processing mechanism, the personal error is significantly reduced, the accuracy and efficiency of drug quality detection are significantly improved, and the detection efficiency is improved. The intelligent degree, the usability and the reliability of the system are improved to a great extent.

Description

Technical Field

[0001] The present invention relates to the field of drug quality detection, and in particular to a drug quality detection method and system based on component analysis. Background Art

[0002] As the pharmaceutical industry develops rapidly, the types and quantities of drugs are increasing. At the same time, the quality and safety of drugs are becoming more and more the focus of people's attention.

[0003] In recent years, modern analytical techniques such as gas chromatography (GC), liquid chromatography (HPLC), and mass spectrometry (MS) have made remarkable progress. However, in the field of drug ingredient analysis, the existing traditional drug ingredient analysis methods have obvious deficiencies in many aspects. In terms of analysis efficiency, traditional methods are often relatively low and require a lot of time to complete the analysis of drug ingredients. In terms of accuracy, traditional methods are also lacking and it is difficult to meet the high-precision standards required by modern testing, which to a certain extent affects the accurate judgment and control of drug quality. This situation may lead to misjudgment of drug quality, thereby affecting the safety and effectiveness of drugs in the market, and may even pose potential risks to patients' health.

[0004] Therefore, the prior art needs to be further developed. Summary of the invention

[0005] The purpose of the present invention is to overcome the above technical deficiencies and provide a drug quality detection method and system based on component analysis to solve the problems existing in the prior art.

[0006] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides a drug quality detection method based on component analysis, comprising: S100, pre-treating the drug sample to be tested to obtain the pre-treated drug sample to be tested; S200, using a gas chromatograph to detect the pretreated drug sample to be tested, to obtain a first chromatogram of the drug sample to be tested; S300, performing data processing on the first chromatogram, and identifying target components in the drug sample to be tested according to the data-processed first chromatogram.

[0007] Specifically, the pre-processing of the drug sample to be tested to obtain the pre-processed drug sample to be tested includes: An electronic balance is used to weigh a first weight of the drug sample to be tested, and the drug sample to be tested is placed in a first solvent to form a first solution. The first solution is stirred by a magnetic stirrer to dissolve the drug sample to be tested in the first solution. Then, a centrifuge is used to remove insoluble impurities in the first solution. Finally, a rotary evaporator is used to remove the first solvent in the first solution, thereby obtaining a pretreated drug sample to be tested.

[0008] Specifically, the method of using a gas chromatograph to detect the pretreated drug sample to obtain a first chromatogram of the drug sample to be tested includes: Configuring the chromatographic column material, coating and injection method of the gas chromatograph; The pretreated drug sample to be tested is injected into a gas chromatograph, and the pretreated drug sample to be tested is separated in the chromatographic column of the gas chromatograph to obtain a separated drug sample to be tested, and then the separated drug sample to be tested is detected using a high-sensitivity detector connected to the gas chromatograph to obtain a first chromatogram.

[0009] Specifically, the method further includes: Comprehensive evaluation index using chromatogram Evaluate the first chromatogram, chromatogram comprehensive evaluation index The calculation method is as follows: ; Where N is the total number of components in the first chromatogram; and are the retention times of two adjacent component peaks; and are the peak widths of two adjacent component peaks at the baseline; is the time interval between adjacent peaks; is the standard deviation of the time interval; is the signal-to-noise ratio of the ith peak; is the moderating factor for the effect of signal-to-noise ratio; is the first chromatogram signal as a function of time; is the average value of the first chromatogram signal; is the total time length of the first chromatogram.

[0010] Specifically, performing data processing on the first chromatogram and identifying the target component in the drug sample to be tested according to the first chromatogram after data processing includes: Obtaining an established reference spectral database, wherein the reference spectral database includes chromatograms of standard drug components; The first chromatogram is processed using a data analysis module constructed based on a machine learning algorithm, and the processed first chromatogram is compared with the reference spectrum database to identify the target component in the drug sample to be tested.

[0011] Specifically, the data analysis module constructed based on the machine learning algorithm is used to process the data of the first chromatogram, including: performing data preprocessing on the first chromatogram; Feature extraction is performed on the first chromatogram after data preprocessing, and a first feature is extracted from the first chromatogram, where the first feature includes the position, height, and area of ​​the component peak.

[0012] Specifically, the first chromatogram after data processing is compared with the reference spectrum database to identify the target component in the drug sample to be tested, including: Compare the first feature of the first chromatogram with the feature of the chromatogram of the standard drug component in the reference spectral database, and calculate the similarity between the first chromatogram and the chromatogram of the standard drug component , the specific calculation method is as follows: ; in, represents the value of the i-th first feature in the first chromatogram after data preprocessing; Represents the value of the i-th feature in the chromatogram of the standard drug component in the reference spectrum database; n represents the total number of features.

[0013] Specifically, the method further includes: Classifying the target components in the identified drug sample to be tested by a decision tree algorithm, and / or obtaining the concentration of the target components, and predicting the concentration of the target components by a linear regression algorithm; Based on the classification results of the target components in the drug sample to be tested, and / or the results of predicting the concentration of the target components, the algorithm parameters are adaptively adjusted and verified by a cross-validation method.

[0014] Specifically, the method further includes: Prepare a test report containing information about the drug sample to be tested, wherein the information about the drug sample to be tested includes at least: Basic information of the drug samples to be tested, recorded chromatograms, analytical methods used, analytical results, and conclusions and recommendations based on the analytical results.

[0015] According to a second aspect of the present invention, there is provided a drug quality detection system based on component analysis, comprising: Acquisition module: including an electronic balance, a magnetic stirrer, a centrifuge, and a rotary evaporator, the acquisition module is used to pre-treat the drug sample to be tested to obtain the pre-treated drug sample to be tested; Detection module: comprising a gas chromatograph and a sample introduction device, the detection module is used to detect the pre-treated drug sample to be tested by using the gas chromatograph to obtain a first chromatogram of the drug sample to be tested; Control module: includes a data analysis module constructed based on a machine learning algorithm, the control module is used to process the first chromatogram and identify the target component in the drug sample to be tested based on the first chromatogram after data processing.

[0016] Beneficial effects: The present invention performs pretreatment on a drug sample to be tested to obtain a pretreated drug sample to be tested; uses a gas chromatograph to detect the pretreated drug sample to be tested to obtain a first chromatogram of the drug sample to be tested; performs data processing on the first chromatogram, and identifies a target component in the drug sample to be tested based on the first chromatogram after data processing. Through a highly automated operating process and an accurate data processing mechanism, human errors are significantly reduced, the accuracy and efficiency of drug quality detection are significantly improved, and the intelligence, usability and reliability of the present invention are greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a drug quality detection method based on component analysis provided in a specific embodiment of the present invention; Figure 2 It is a schematic diagram of the composition of a drug quality detection system based on component analysis provided in a specific embodiment of the present invention; Figure 3 It is a schematic diagram of a drug quality detection step based on component analysis provided in a specific embodiment of the present invention; Figure 4 It is a structural schematic diagram of a drug quality detection system based on component analysis provided in a specific embodiment of the present invention. DETAILED DESCRIPTION In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0018] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0019] See also Figure 1 This embodiment provides a drug quality detection method based on component analysis, including: S100, pre-treating the drug sample to be tested to obtain the pre-treated drug sample to be tested; Specifically, the pre-processing of the drug sample to be tested to obtain the pre-processed drug sample to be tested includes: An electronic balance is used to weigh a first weight of the drug sample to be tested, and the drug sample to be tested is placed in a first solvent to form a first solution. The first solution is stirred by a magnetic stirrer to dissolve the drug sample to be tested in the first solution. Then, a centrifuge is used to remove insoluble impurities in the first solution. Finally, a rotary evaporator is used to remove the first solvent in the first solution, thereby obtaining a pretreated drug sample to be tested.

[0020] It can be understood that the present invention uses a high-precision electronic balance to accurately weigh a certain mass of pharmaceutical samples. The accuracy range of the high-precision electronic balance ranges from 0.00001g to 0.1g, which can meet the needs of different experiments and industrial applications, and can further ensure the accuracy and repeatability of the weighing results.

[0021] S200, using a gas chromatograph to detect the pretreated drug sample to obtain a first chromatogram of the drug sample to be tested.

[0022] Specifically, the method of using a gas chromatograph to detect the pretreated drug sample to obtain a first chromatogram of the drug sample to be tested includes: Configuring the chromatographic column material, coating and injection method of the gas chromatograph; The pretreated drug sample to be tested is injected into a gas chromatograph, and the pretreated drug sample to be tested is separated in the chromatographic column of the gas chromatograph to obtain a separated drug sample to be tested, and then the separated drug sample to be tested is detected using a high-sensitivity detector connected to the gas chromatograph to obtain a first chromatogram.

[0023] It should be further explained that the present invention uses a gas chromatograph equipped with a capillary chromatographic column, accurately selects a suitable chromatographic column material and coating according to the chemical and physical properties of the components of the drug sample to be tested, sets an appropriate injection method, and finely adjusts the injection port temperature to prevent the sample from decomposing during the injection process, and configures a high-sensitivity detector and correctly connects it to the gas chromatograph; then the pretreated drug sample to be tested is accurately injected into the gas chromatograph through a precise injector, and the sample is effectively separated in the chromatographic column according to the difference in boiling points of the components, and then detected by the high-sensitivity detector and a detailed chromatogram is recorded, that is, a first chromatogram is obtained; In some specific embodiments, the selection of chromatographic column materials and coatings, as well as the setting of injection methods and injection port temperatures, are all based on an in-depth understanding and scientific analysis of the characteristics of the components of the drug sample to be tested, in order to optimize the separation effect and prevent sample decomposition. In sample analysis, the separation process of the drug sample to be tested in the chromatographic column is based on the difference in boiling points of the components and is precisely controlled by a gas chromatograph. The chromatogram is captured and recorded in real time by a high-sensitivity detector to ensure accurate identification and separation of each component. The separation effect of the chromatographic column is evaluated by calculating the separation degree of adjacent component peaks. The higher the separation degree, the more thorough the separation between the components. The calculation formula is as follows: ; in, and are the retention times of two adjacent component peaks, and are the peak width and resolution of two adjacent component peaks at the baseline. The larger the value, the better the separation effect.

[0024] Preferably, the chromatographic column in this embodiment can be made of stainless steel and adopt a polar coating suitable for analyzing the components of antibiotic drugs, such as polyethylene glycol (PEG) or polydimethylsiloxane (PDMS), to ensure that the components in the sample can be effectively separated on the chromatographic column. The chromatographic column specifications can be set to: 30 meters in length, 0.32 mm in inner diameter, and 0.25 microns in film thickness to meet the analysis requirements of high resolution and high sensitivity. The split injection method is adopted to ensure that the injection volume is stable and suitable for analysis by accurately controlling the split ratio (such as 1:100). According to the thermal stability and volatility of the antibiotic sample, the injection port temperature can be set to 250°C to prevent the sample from decomposing during the injection process.

[0025] Preferably, the high-sensitivity detector may be a high-sensitivity flame ionization detector (FID), which is suitable for the detection of most organic compounds. Ensure that the detector is correctly connected to the gas chromatograph, and perform necessary debugging and calibration to ensure the accuracy of the detection results.

[0026] Specifically, the method further includes: Comprehensive evaluation index using chromatogram Evaluate the first chromatogram, chromatogram comprehensive evaluation index The calculation method is as follows: ; Where N is the total number of components in the first chromatogram; and are the retention times of two adjacent component peaks; and are the peak widths of two adjacent component peaks at the baseline; is the time interval between adjacent peaks; is the standard deviation of the time interval; is the signal-to-noise ratio of the ith peak; is the moderating factor for the effect of signal-to-noise ratio; is the first chromatogram signal as a function of time; is the average value of the first chromatogram signal; is the total time length of the first chromatogram.

[0027] It can be understood that the comprehensive evaluation index using the chromatogram ( ≥0) is used to evaluate the first chromatogram to quantify the quality of the chromatogram and the accuracy of the analysis. The comprehensive evaluation index of the chromatogram is used to evaluate the overall quality of the chromatogram, including the separation effect, the stability of the time interval, the signal-to-noise ratio, and the overall rate of change of the signal. The larger the value of the comprehensive evaluation index of the chromatogram, the higher the quality of the chromatogram, the better the separation effect, and the higher the accuracy of the analysis. The calculated value can further guide the subsequent steps of drug quality testing, including optimizing chromatographic conditions, adjusting analysis parameters, and verifying the accuracy of analysis results.

[0028] In some specific embodiments, the specific implementation steps for comprehensive evaluation of chromatograms are as follows: (1) Data collection Use gas chromatograph to analyze the pre-treated drug samples to obtain detailed chromatogram data; Record key parameters such as retention time, peak width, signal-to-noise ratio, etc. of each component in the chromatogram.

[0029] (2) Chromatogram preprocessing De-noising the original chromatogram to reduce the interference of background noise on the analysis results; Standardize the signal intensity of chromatograms to ensure comparability of chromatograms produced from different batches or different equipment.

[0030] (3) Calculation of comprehensive evaluation index of chromatogram Apply the above mathematical formula to calculate the comprehensive evaluation index of the chromatogram .

[0031] This formula takes into account the separation effect, the stability of the time interval, the signal-to-noise ratio, and the overall rate of change of the signal.

[0032] (4) Evaluation and analysis according to The value of evaluates the overall quality of the chromatogram. The larger the value, the higher the quality of the chromatogram, the better the separation effect, and the higher the analysis accuracy. Review low-quality chromatograms to check for instrument failure, improper operation, or sample contamination. Based on the evaluation results, the chromatographic conditions are optimized, such as adjusting the chromatographic column, changing the injection volume, modifying the temperature program, etc., to improve the quality of the chromatogram.

[0033] (5) Result feedback and optimization Incorporate the comprehensive evaluation results of the chromatogram into the test report as an important reference for analytical accuracy; Utilize the self-learning and optimization functions of the data analysis module to automatically adjust algorithm parameters based on accumulated historical data to improve the ability to analyze components of different types of drugs; Regularly review the comprehensive evaluation results of chromatograms, summarize experiences and lessons, and continuously optimize the detection process.

[0034] Furthermore, in order to quantitatively evaluate the chromatogram, a reasonable evaluation threshold needs to be set. The above evaluation threshold can be set according to historical data, industry standards or experimental requirements. The present invention does not further limit the value of the evaluation threshold. For example, the evaluation threshold can be set to 0.8. Only when the comprehensive evaluation index of the chromatogram is greater than or equal to 0.8, its quality is considered to be qualified. According to the value range of the comprehensive evaluation index of the chromatogram, the chromatogram can be divided into the following categories: High-quality chromatogram: When the comprehensive evaluation index of the chromatogram is much higher than the set evaluation threshold (such as above 0.95), it can be considered that the quality of the chromatogram is very high, the separation effect is good, and the analysis accuracy is high; Qualified chromatogram: When the comprehensive evaluation index of the chromatogram is near the set evaluation threshold (such as 0.8-0.95), the quality of the chromatogram can be considered qualified, but there may be some room for improvement; Low-quality chromatogram: When the comprehensive evaluation index of the chromatogram is lower than the set evaluation threshold (such as below 0.8), it can be considered that the quality of the chromatogram is low, and there may be problems such as incomplete separation, low signal-to-noise ratio or unstable signal.

[0035] The following are the treatment measures for abnormal chromatograms: Review the chromatogram: When the comprehensive evaluation index of the chromatogram is found to be lower than the evaluation threshold, the chromatogram should be reviewed immediately to check whether there are problems such as instrument failure, improper operation or sample contamination; Optimize chromatographic conditions: Based on the review results, if the chromatogram is qualified, the chromatographic conditions can be optimized, such as adjusting the chromatographic column, changing the injection volume, modifying the temperature program, etc., to improve the quality of the chromatogram; Re-testing: After optimizing the chromatographic conditions, the drug samples to be tested need to be retested and the comprehensive evaluation index of the chromatogram needs to be recalculated to ensure that the quality of the chromatogram meets the requirements.

[0036] Recording and feedback: The comprehensive evaluation results of the chromatogram are included in the test report as an important reference for analysis accuracy. At the same time, the self-learning and optimization functions of the data analysis module are used to automatically adjust the algorithm parameters according to the accumulated historical data to improve the analysis capabilities of different types of drug components.

[0037] It can be understood that by introducing the comprehensive evaluation method of chromatograms, multiple factors in the chromatogram are comprehensively considered, including the total number of components, retention time of adjacent component peaks, peak width, stability of time intervals, signal-to-noise ratio, and overall rate of change of the signal, etc. These factors together reflect the overall quality of the chromatogram and the accuracy of the analysis, and successfully realize the quantitative evaluation of the quality of chromatograms in the process of drug quality testing, which not only improves the accuracy and reliability of the test results, but also provides strong data support for subsequent optimization work. At the same time, the automation and intelligent characteristics of this method also significantly reduce human errors and improve detection efficiency.

[0038] S300, performing data processing on the first chromatogram, and identifying target components in the drug sample to be tested according to the data-processed first chromatogram.

[0039] Specifically, performing data processing on the first chromatogram and identifying the target component in the drug sample to be tested according to the first chromatogram after data processing includes: Obtaining an established reference spectral database, wherein the reference spectral database includes chromatograms of standard drug components; The first chromatogram is processed using a data analysis module constructed based on a machine learning algorithm, and the processed first chromatogram is compared with the reference spectrum database to identify the target component in the drug sample to be tested.

[0040] Preferably, the above-mentioned machine learning algorithm can select linear regression, logistic regression, support vector machine (SVM) and other methods, and the specific principles and applications are as follows: (1) Linear regression algorithm Principle: The linear regression algorithm describes the linear relationship between the input variables and the output variables by finding the specific weights of the input variables. In chromatogram analysis, the input variables can be various features of the chromatogram (such as peak height, peak width, retention time, etc.), and the output variable can be the concentration or presence of the target component.

[0041] Application: By training the model, linear regression can predict the concentration of the target component in a new chromatogram or determine whether it exists. (2) Logistic regression Principle: Logistic regression is a special method for binary classification problems. It uses a logistic function to convert the output of linear regression into a probability value between 0 and 1, and classifies it according to the set threshold.

[0042] Application: In chromatogram analysis, logistic regression can be used to determine whether the target component exists (i.e., binary classification problem), such as determining whether a specific peak represents the target component; (3) Support Vector Machine (SVM) Principle: SVM is a linear classifier based on maximizing the classification interval (without considering the kernel function). It separates the points in the input variable space by their categories by finding a hyperplane.

[0043] Application: In chromatogram analysis, SVM can be used for multi-classification problems, such as identifying multiple target components. By training the model, SVM can learn how to map the features of the chromatogram to different categories.

[0044] It should be noted here that the data analysis module constructed based on the machine learning algorithm can automatically and accurately process the acquired first chromatogram, and quickly and accurately identify and quantitatively analyze the target components in the sample by real-time and precise comparison with the established reference spectrum database containing the chromatograms of standard drug components, thereby significantly improving the accuracy and efficiency of qualitative and quantitative analysis. At the same time, the data analysis module constructed based on the machine learning algorithm also has self-learning and optimization capabilities, and can automatically adjust the algorithm parameters based on the continuously accumulated data and analysis results to continuously improve the analysis performance.

[0045] Specifically, the data analysis module constructed based on the machine learning algorithm is used to process the data of the first chromatogram, including: performing data preprocessing on the first chromatogram; In some specific embodiments, performing data preprocessing and denoising on the first chromatogram may include removing noise in the chromatogram by a filtering method, and normalizing the signal intensity of the chromatogram to the same scale to facilitate subsequent processing; Feature extraction is performed on the first chromatogram after data preprocessing, and a first feature is extracted from the first chromatogram, where the first feature includes the position, height, and area of ​​the component peak.

[0046] Specifically, the first chromatogram after data processing is compared with the reference spectrum database to identify the target component in the drug sample to be tested, including: Compare the first feature of the first chromatogram with the feature of the chromatogram of the standard drug component in the reference spectral database, and calculate the similarity between the first chromatogram and the chromatogram of the standard drug component , the specific calculation method is as follows: ; in, represents the value of the i-th first feature in the first chromatogram after data preprocessing; Represents the value of the i-th feature in the chromatogram of the standard drug component in the reference spectrum database; Represents the total number of features.

[0047] Understandably, Represents the total number of features, that is, each chromatogram is converted into a vector of values, the similarity The closer the value of is to 1, the higher the similarity between the first chromatogram and the chromatogram of the standard drug component is; It should be further explained that the threshold selection principle of similarity is as follows: High accuracy: To ensure the accuracy of recognition, the similarity threshold should be set relatively high to reduce the risk of misidentification; Adaptability: The selection of thresholds should take into account the differences between different drug components, as well as the impact of chromatogram data preprocessing and feature extraction methods; Consistency: Under the same analytical conditions, the similarity between chromatograms of the same drug component from different batches or sources should remain consistent.

[0048] The specific steps for setting the threshold of the specific similarity are as follows: Preliminary threshold: Based on historical data and experimental experience, a similarity threshold can be preliminarily set, such as 0.90 or higher, which means that only when the similarity between the first chromatogram and the chromatogram of the standard drug component reaches or exceeds 0.90, they are considered to have a high degree of similarity; Dynamic adjustment: In practical applications, the threshold can be dynamically adjusted according to the specific situation. For example, if it is found that the chromatograms of certain drug components have large variability during preprocessing or feature extraction, the similarity threshold can be appropriately lowered to improve the recognition rate; Multi-threshold strategy: In order to more finely control the recognition results, a multi-threshold strategy can be adopted. For example, a higher threshold (such as 0.95) is set for preliminary screening, and then the screening results are further verified at a lower threshold (such as 0.90).

[0049] The specific evaluation process of similarity is as follows: High similarity: When the similarity between the first chromatogram and the chromatogram of the standard drug component is close to 1 (such as above 0.95), it can be considered that there is a high degree of similarity between them, thereby confirming that the drug sample to be tested contains the standard drug component; Moderate similarity: When the similarity is between 0.90 and 0.95, further analysis and verification may be required. For example, you can check whether the peak shape, retention time and other characteristics of the chromatogram are consistent with the target component, or use other analytical methods for confirmation; Low similarity: When the similarity is lower than 0.90, it can be considered that the difference between the first chromatogram and the chromatogram of the standard drug component is large, and therefore it cannot be confirmed that the drug sample to be tested contains the standard drug component.

[0050] Specifically, the method further includes: Classifying the target components in the identified drug sample to be tested by a decision tree algorithm, and / or obtaining the concentration of the target components, and predicting the concentration of the target components by a linear regression algorithm; It is understandable that high performance liquid chromatography (HPLC) can be used to determine the concentration of the target component. The specific principle is: utilize the distribution phenomenon of the solution between the stationary phase and the mobile phase to separate the drug mixture and determine the concentration of its components; by optimizing the chromatographic column, mobile phase and other conditions, inject the sample to be tested into the chromatograph and record the chromatogram; calculate the concentration of each component based on the proportional relationship between the peak area or peak height of the chromatographic peak and the standard.

[0051] Based on the classification results of the target components in the drug sample to be tested, and / or the results of predicting the concentration of the target components, the algorithm parameters are adaptively adjusted and verified by a cross-validation method.

[0052] Furthermore, the decision tree algorithm performs well in classification tasks. For the target components in the drug samples to be tested, it can effectively classify the target components according to different characteristics, which helps to identify different types of target components; the linear regression algorithm is used to predict the concentration of the target components. This model based on data relationships can more accurately estimate the concentration value. The combination of the two algorithms can analyze the target components in the drug samples from different angles, thereby improving the comprehensiveness and accuracy of data analysis; the algorithm parameters are adaptively adjusted according to the classification results or concentration prediction results of the target components. This is a way of dynamically optimizing the algorithm. With the feedback of the drug sample data analysis results, the algorithm can automatically adjust its own parameters to better adapt to the characteristics of different drug samples and improve the algorithm's analysis capabilities for different drugs; the cross-validation method is used for verification, which can effectively evaluate the generalization ability of the model. When analyzing the drug samples to be tested, the model can be avoided from overfitting by dividing the data set for training and verification multiple times, thereby ensuring that the algorithm can maintain a high reliability on different drug sample data, thereby improving the credibility of the entire analysis method. In summary, the self-learning and optimization functions of the data analysis module built on the machine learning algorithm automatically adjust the algorithm parameters based on the continuously accumulated historical data and analysis results to better adapt to the analysis needs of different drug components and improve the accuracy and reliability of the analysis.

[0053] Specifically, the method further includes: Prepare a test report containing information about the drug sample to be tested, wherein the information about the drug sample to be tested includes at least: Basic information of the drug samples to be tested, recorded chromatograms, analytical methods used, analytical results, and conclusions and recommendations based on the analytical results.

[0054] Taking the sample drug to be tested as Amoxicillin Capsules as an example, the main process of preparing the test report is explained as follows: 1. Basic information of samples Sample name: Amoxicillin capsules (a specific batch); Sample batch: 20230510; Sample source: YY Pharmaceutical Factory; Sample specification: 0.25g / pill; Sample quantity: 50 pieces; Storage conditions: sealed, stored in a cool and dry place; 2. The obtained chromatogram The pretreated amoxicillin capsule sample was tested by gas chromatography to obtain the first chromatogram. The chromatogram records in detail the separation of each component in the sample over time, including key information such as the retention time, peak height, and peak area of ​​each component. The chromatogram is not directly displayed here, but it can be analyzed based on the information on the chromatogram; 3. Analytical methods used (1) Sample pretreatment: A certain mass of amoxicillin capsule sample (e.g. 0.1 g) is weighed using a high-precision electronic balance, and the sample is placed in an appropriate amount of methanol solvent to form a solution; Stir thoroughly with a magnetic stirrer to completely dissolve the sample; Use a centrifuge to remove insoluble impurities from a solution; Finally, the solvent was removed by a rotary evaporator to obtain the pretreated sample; (2) Gas chromatograph configuration: Use a gas chromatograph equipped with a capillary column; According to the chemical and physical properties of the components of amoxicillin capsules, select the appropriate chromatographic column material and coating; Set up appropriate injection mode and fine-tune the injection port temperature; Configure a high-sensitivity detector and properly connect it to the gas chromatograph; (3) Chromatographic analysis: The pretreated sample is accurately injected into the gas chromatograph through a precision injector; The sample is effectively separated in the chromatographic column based on the differences in boiling points of the components; It is detected by a highly sensitive detector and records a detailed chromatogram; (4) Data analysis: The chromatograms were processed using a data analysis module based on machine learning algorithms; Extract characteristic information from the chromatogram, such as the position, height, area, etc. of the component peak; Compare the processed chromatogram with the established reference spectrum database to identify the target components in the sample; 4. Analysis results Through the processing and comparison of the data analysis module, the target components in the amoxicillin capsule sample are identified, including the main drug component of amoxicillin, impurity components, etc. At the same time, the concentration information of each component can also be obtained.

[0055] 5. Conclusion According to the analysis results, the content of the main drug component in the amoxicillin capsule sample meets the prescribed standards, and the content of the impurity component is within the acceptable range. Therefore, the quality of this batch of amoxicillin capsules is qualified.

[0056] 6. Suggestions Continue monitoring: It is recommended that the pharmaceutical factory continue to monitor the quality of this batch of amoxicillin capsules to ensure that it remains qualified during use.

[0057] Optimize the production process: Although the test results are qualified, the pharmaceutical factory can consider further optimizing the production process to reduce the production of impurities and improve the purity of the drug.

[0058] Strengthen raw material management: Strengthen the management and testing of raw materials to ensure that the quality of raw materials meets production requirements, thereby producing higher quality medicines.

[0059] Furthermore, the basic information of the drug samples to be tested, such as drug name, batch, source, etc., is helpful in providing basic information in drug management, traceability, etc. The recorded chromatogram, that is, the first chromatogram, is an important raw data in drug analysis. It intuitively reflects the drug ingredient information, provides a visual basis for subsequent qualitative and quantitative analysis, and ensures the integrity of the test report data; clearly defines the analysis method used. Different analysis methods may affect the results. Recording the analysis method helps to verify the repeatability of the results; it also includes qualitative and quantitative analysis results, which can comprehensively reflect the nature of the drug and the content of each component. Qualitative analysis determines the ingredients in the drug, and quantitative analysis determines the content of each component. The specific content of the ingredients, the combination of the two is crucial for accurately evaluating the quality of the drug; the conclusions and suggestions based on the analysis results are an important part of the test report. The conclusions make it clear whether the drug meets the relevant standards or expectations, which is directly related to the further processing of the drug; the suggestions part can propose improvement measures for the problems existing in the drug, such as production process adjustments, storage conditions optimization, etc., which will help improve the quality and safety of the drug; throughout the life cycle of the drug, this detailed test report facilitates quality traceability. If a problem occurs in the subsequent use of the drug, it can be traced back to the drug testing link based on the detailed information in the report, which greatly improves the reliability and usability of the present invention.

[0060] Furthermore, the present invention covers the complete process from sample pretreatment to report preparation. Through high-precision instrument configuration and artificial intelligence-assisted data analysis, it realizes the precise separation and rapid identification of drug components. The chromatogram is comprehensively evaluated by the introduced chromatogram comprehensive evaluation index, and the accuracy of the quantitative analysis and the quality of the chromatogram are provided, which provides a scientific basis for optimizing the detection conditions and verifying the accuracy of the results. The whole process not only improves the accuracy and efficiency of drug quality detection, but also demonstrates self-learning and optimization capabilities, ensuring the continuous improvement of analytical performance.

[0061] See also Figure 3 , the working steps of the present invention are described below by means of specific examples: Step 1: Sample pretreatment A certain mass of the drug sample is accurately weighed using a high-precision electronic balance, and then placed in a suitable solvent, and fully stirred by a magnetic stirrer to ensure that the sample is completely dissolved, and then a centrifuge is used to remove insoluble impurities in the solution, and finally a rotary evaporator is used to completely remove the solvent, thereby obtaining a pretreated drug sample; Step 2: Instrument Configuration Choose a gas chromatograph equipped with a capillary column, accurately select the appropriate column material and coating according to the chemical and physical properties of the drug components to be tested, set the appropriate injection method, and finely adjust the injection port temperature to prevent the sample from decomposing during the injection process. At the same time, configure a highly sensitive detector and correctly connect it to the gas chromatograph; Step 3: Sample Analysis The pretreated drug sample is accurately injected into the gas chromatograph through a precise injector. The sample is effectively separated in the chromatographic column according to the boiling point differences of each component, and then detected by a high-sensitivity detector and a detailed chromatogram is recorded; Step 4: AI-assisted data analysis The data analysis and interpretation software developed based on the machine learning algorithm is used as the data analysis module to automatically and accurately process the chromatogram obtained in step 3. By real-time and accurate comparison with the established reference spectrum database containing the chromatograms of standard drug components, the target components in the sample can be quickly and accurately identified and quantified, thereby significantly improving the accuracy and efficiency of qualitative analysis. At the same time, the data analysis and interpretation software also has self-learning and optimization capabilities, and can automatically adjust the algorithm parameters based on the continuously accumulated data and analysis results to continuously improve the analysis performance. The specific steps are as follows: (1) Data preprocessing Denoising: removing noise from the chromatogram by filtering methods; Normalization: Normalize the signal intensity of the chromatogram to the same scale to facilitate subsequent processing; Feature extraction: Extract key features (i.e., first features) from the chromatogram, including the position, height, and area of ​​the component peaks; (2) Pattern recognition Feature comparison: compare the extracted features with those in the reference spectrum database; Similarity calculation: Calculate the similarity between the sample chromatogram and the reference chromatogram using the cosine similarity calculation formula; the closer the cosine similarity value is to 1, the higher the similarity between the two chromatograms; (3) Classification and regression Classification: When it is necessary to identify specific components in a sample, a decision tree algorithm is used for classification; Regression: When the concentration of a component needs to be quantitatively determined, a linear regression algorithm is used for prediction; (4) Result optimization and feedback Self-learning: Automatically adjust algorithm parameters to optimize performance based on new analysis results and feedback; Result verification: The accuracy and reliability of the analysis results were verified by cross-validation method.

[0062] Step 5: Report Preparation Based on the analysis results obtained by the artificial intelligence-assisted data analysis and interpretation software in step 4, prepare a test report containing detailed information. The report includes basic information of the sample, the analysis method used, the recorded chromatograms, qualitative and quantitative analysis results, and conclusions and recommendations based on the analysis results.

[0063] Step 6: Comprehensive evaluation of chromatograms The chromatogram comprehensive evaluation index mathematical formula is used to comprehensively evaluate the chromatogram to quantify the quality of the chromatogram and the accuracy of the analysis: The comprehensive evaluation index of the chromatogram is used to evaluate the overall quality of the chromatogram, including separation effect, stability of time intervals, signal-to-noise ratio, and overall rate of change of the signal; the calculated value can further guide the subsequent steps of drug quality testing, including optimizing chromatographic conditions, adjusting analysis parameters, and verifying the accuracy of analysis results.

[0064] It should be noted here that this embodiment provides a drug quality detection method based on component analysis, which significantly improves the accuracy and efficiency of drug quality detection through high-precision preprocessing, precise instrument configuration, efficient sample analysis, intelligent data analysis and detailed report preparation. This embodiment uses modern scientific and technological means, such as high-precision electronic balance, gas chromatograph and artificial intelligence algorithm, to ensure the refinement and automation of the detection process, and effectively reduce human errors; by introducing the chromatogram evaluation index formula to comprehensively evaluate the chromatogram, quantify the analysis accuracy and chromatogram quality, and provide a scientific basis for optimizing the detection conditions. It can comprehensively evaluate the separation effect, time interval stability, signal-to-noise ratio, etc. of the chromatogram, so as to guide the detection personnel to adjust the parameters to improve the quality of the chromatogram and further improve the reliability of the analysis results. And this embodiment adopts advanced artificial intelligence technology, including data preprocessing, pattern recognition, classification and regression, etc., to achieve accurate analysis and rapid interpretation of the chromatogram. At the same time, the data analysis and interpretation software has strong self-learning and optimization capabilities, can continuously adapt to new data, greatly improves the analysis accuracy and reliability, and largely ensures the efficiency, accuracy and reliability of drug quality detection.

[0065] See also Figure 2 This embodiment provides a drug quality detection system based on component analysis, the system comprising: Acquisition module 100: includes an electronic balance, a magnetic stirrer, a centrifuge, and a rotary evaporator. The acquisition module is used to pre-treat the drug sample to be tested to obtain the pre-treated drug sample to be tested; The detection module 200 includes a gas chromatograph and a sample introduction device. The detection module is used to detect the pre-treated drug sample to be tested by using the gas chromatograph to obtain a first chromatogram of the drug sample to be tested. The control module 300 includes a data analysis module constructed based on a machine learning algorithm, and is used to perform data processing on the first chromatogram and identify target components in the drug sample to be tested based on the first chromatogram after data processing.

[0066] See also Figure 4 The present embodiment provides a drug quality detection system based on component analysis, which is suitable for the above-mentioned drug quality detection method based on component analysis, and is equipped with advanced equipment such as a high-precision electronic balance, a magnetic stirrer, a centrifuge, a rotary evaporator, and a gas chromatograph, wherein the gas chromatograph includes a chromatographic column and a high-sensitivity detector, and the injection device includes an injection needle. It is also equipped with a data analysis module built based on a machine learning algorithm. The components of the drug quality detection system based on component analysis are seamlessly integrated through precise control and data exchange to automatically complete the entire process from sample pretreatment to data analysis. Through highly automated operating procedures and precise data processing mechanisms, human errors are significantly reduced, and detection efficiency and accuracy are greatly improved.

[0067] In some specific embodiments, the chromatographic column is a core component of the gas chromatograph, which is located inside the gas chromatograph. Usually, one end of the chromatographic column is connected to the injection port, and the other end is connected to the inlet of the detector. The chromatographic column is fixed in a specially designed column box inside the gas chromatograph, and the column box can provide a precise temperature control environment to meet different analysis requirements. As an important component of the gas chromatograph, the high-sensitivity detector can be directly integrated with the gas chromatograph. It has a specific installation position in the gas chromatograph and is connected to the outlet end of the chromatographic column to receive the components flowing out after separation from the chromatographic column for detection. The injection needle is usually connected to the front end of the injection device so that the sample can be accurately injected into the injection device. Specifically, the injection device provides a fixed and operational interface for the injection needle, and the injection needle is connected to the injection device through this interface, so that the operator can control the injection action of the injection needle through the relevant operating structure of the injection device.

[0068] It should be noted here that the present invention performs pretreatment on the drug sample to be tested to obtain a pretreated drug sample to be tested; uses a gas chromatograph to detect the pretreated drug sample to be tested to obtain a first chromatogram of the drug sample to be tested; performs data processing on the first chromatogram, and identifies the target component in the drug sample to be tested based on the first chromatogram after data processing. Through a highly automated operating process and a precise data processing mechanism, human errors are significantly reduced, the accuracy and efficiency of drug quality detection are significantly improved, and the intelligence, usability and reliability of the present invention are greatly improved.

[0069] In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the drug quality detection method based on component analysis is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory may provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices through a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0070] It should be noted here that the drug sample to be tested is pretreated to obtain the pretreated drug sample to be tested; the pretreated drug sample to be tested is tested using a gas chromatograph to obtain a first chromatogram of the drug sample to be tested; data processing is performed on the first chromatogram, and the target component in the drug sample to be tested is identified based on the first chromatogram after data processing. Through a highly automated operating process and a precise data processing mechanism, human errors are significantly reduced, the accuracy and efficiency of drug quality detection are significantly improved, and the intelligence, usability and reliability of the present invention are greatly improved. The standard parts or devices used in the present invention can be purchased from the market, and the special-shaped parts can be customized according to the description and the drawings. The specific connection methods of each part adopt conventional means such as mature bolts, rivets, welding, etc. in the prior art. The machinery, parts and equipment all adopt conventional models in the prior art, and the circuit connection adopts the conventional connection method in the prior art, which will not be described in detail here. The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

[0071] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0072] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0073] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A drug quality detection method based on component analysis, characterized in that: include: S100, pre-treating the drug sample to be tested to obtain the pre-treated drug sample to be tested; S200, using a gas chromatograph to detect the pretreated drug sample to be tested, to obtain a first chromatogram of the drug sample to be tested; S300, performing data processing on the first chromatogram, and identifying target components in the drug sample to be tested according to the data-processed first chromatogram.

2. The drug quality detection method based on component analysis according to claim 1 is characterized in that: The method of pre-treating the drug sample to be tested to obtain the pre-treated drug sample to be tested comprises: An electronic balance is used to weigh a first weight of the drug sample to be tested, and the drug sample to be tested is placed in a first solvent to form a first solution. The first solution is stirred by a magnetic stirrer to dissolve the drug sample to be tested in the first solution, and then a centrifuge is used to remove insoluble impurities in the first solution. Finally, a rotary evaporator is used to remove the first solvent in the first solution, thereby obtaining a pretreated drug sample to be tested.

3. The drug quality detection method based on component analysis according to claim 1 is characterized in that: The method of using a gas chromatograph to detect the pretreated drug sample to obtain a first chromatogram of the drug sample to be tested includes: Configuring the chromatographic column material, coating and injection method of the gas chromatograph; The pretreated drug sample to be tested is injected into a gas chromatograph, and the pretreated drug sample to be tested is separated in the chromatographic column of the gas chromatograph to obtain a separated drug sample to be tested, and then the separated drug sample to be tested is detected using a high-sensitivity detector connected to the gas chromatograph to obtain a first chromatogram.

4. The drug quality detection method based on component analysis according to claim 3 is characterized in that: The method further comprises: Comprehensive evaluation index using chromatogram Evaluate the first chromatogram, chromatogram comprehensive evaluation index The calculation method is as follows: ； Where N is the total number of components in the first chromatogram; and are the retention times of two adjacent component peaks; and are the peak widths of two adjacent component peaks at the baseline; is the time interval between adjacent peaks; is the standard deviation of the time interval; is the signal-to-noise ratio of the ith peak; is the moderating factor for the effect of signal-to-noise ratio; is the first chromatogram signal as a function of time; is the average value of the first chromatogram signal; is the total time length of the first chromatogram.

5. The drug quality detection method based on component analysis according to claim 1 is characterized in that: The step of performing data processing on the first chromatogram and identifying a target component in the drug sample to be tested according to the first chromatogram after the data processing includes: Obtaining an established reference spectral database, wherein the reference spectral database includes chromatograms of standard drug components; The first chromatogram is processed using a data analysis module constructed based on a machine learning algorithm, and the processed first chromatogram is compared with the reference spectrum database to identify the target component in the drug sample to be tested.

6. The drug quality detection method based on component analysis according to claim 5 is characterized in that: The method of performing data processing on the first chromatogram using a data analysis module constructed based on a machine learning algorithm comprises: performing data preprocessing on the first chromatogram; Feature extraction is performed on the first chromatogram after data preprocessing, and a first feature is extracted from the first chromatogram, where the first feature includes the position, height, and area of ​​the component peak.

7. The drug quality detection method based on component analysis according to claim 6 is characterized in that: The step of comparing the first chromatogram after data processing with the reference spectrum database to identify the target component in the drug sample to be tested includes: Compare the first feature of the first chromatogram with the feature of the chromatogram of the standard drug component in the reference spectral database, and calculate the similarity between the first chromatogram and the chromatogram of the standard drug component , the specific calculation method is as follows: ； in, represents the value of the i-th first feature in the first chromatogram after data preprocessing; Represents the value of the i-th feature in the chromatogram of the standard drug component in the reference spectrum database; n represents the total number of features.

8. The drug quality detection method based on component analysis according to claim 7 is characterized in that: The method further comprises: Classifying the target components in the identified drug sample to be tested by a decision tree algorithm, and / or obtaining the concentration of the target components, and predicting the concentration of the target components by a linear regression algorithm; Based on the classification results of the target components in the drug sample to be tested, and / or the results of predicting the concentration of the target components, the algorithm parameters are adaptively adjusted and verified by a cross-validation method.

9. The drug quality detection method based on component analysis according to claim 8, characterized in that: The method further comprises: Prepare a test report containing information about the drug sample to be tested, wherein the information about the drug sample to be tested includes at least: Basic information of the drug samples to be tested, recorded chromatograms, analytical methods used, analytical results, and conclusions and recommendations based on the analytical results.

10. A drug quality detection system based on component analysis, characterized in that: include: An acquisition module (100): comprising an electronic balance, a magnetic stirrer, a centrifuge, and a rotary evaporator, the acquisition module being used to pre-treat the drug sample to be tested to obtain the pre-treated drug sample to be tested; A detection module (200): comprising a gas chromatograph and a sample introduction device, wherein the detection module is used to detect the pre-treated drug sample to be tested by using the gas chromatograph to obtain a first chromatogram of the drug sample to be tested; A control module (300): comprising a data analysis module constructed based on a machine learning algorithm, the control module being used to perform data processing on the first chromatogram and identify target components in the drug sample to be tested based on the first chromatogram after data processing.

Images