A quality detection device for oral liquid of Chinese patent medicine

Abstract

The present invention relates to a quality detection device for oral liquid of Chinese patent medicine, which integrates pretreatment, infrared spectrum, fingerprint spectrum, ultraviolet visible spectrum detection module and cloud analysis feedback system. Intelligent pretreatment is used to ensure sample suitability, multi-spectral technology is used to comprehensively analyze components, cloud database and random forest algorithm are combined, and types are quickly subdivided according to absorbance, peak position shift and area change are quantified in detail, and finally the quality grade is determined and feedback is given to users in real time. The system improves detection efficiency and accuracy, and strengthens the intelligent level of quality control of Chinese medicine.

Description

Technical Field

[0001] The invention relates to the technical field of Chinese patent medicine detection, and in particular to a quality detection device for Chinese patent medicine oral liquid. Background Art

[0002] With the rapid development of the modernization of traditional Chinese medicine, Chinese patent medicines are widely popular due to their unique efficacy and fewer side effects. However, the quality control of Chinese patent medicine oral liquid is directly related to the safety and effectiveness of the drug. Traditional detection methods often rely on chemical analysis and biological activity testing, which is not only time-consuming, but also requires high skills of operators, making it difficult to achieve rapid and large-scale quality monitoring. Therefore, it is particularly important to develop an efficient and accurate quality detection device and method for Chinese patent medicine oral liquid.

[0003] Currently, spectral analysis technology has been widely used in the field of drug analysis due to its non-destructive, rapid response and high sensitivity. Among them, infrared spectroscopy, fingerprint (especially high performance liquid chromatography fingerprint) and ultraviolet visible spectroscopy are powerful tools for evaluating the diversity and complexity of Chinese patent medicine ingredients. However, the use of any one technology alone may not be able to fully cover the complex chemical composition of Chinese patent medicines, and the data analysis is complex and difficult to be directly used for quality grading.

[0004] At the same time, among the current Chinese patent medicine oral liquids, due to differences in process, even the same oral liquid produced by different manufacturers may have color depth, concentration or other differences. This is mainly due to differences in production process and does not represent the quality of the oral liquid. Therefore, direct use of fingerprint or infrared spectrum analysis is prone to cause certain influences due to the different colors of the oral liquid itself. It is more accurate to conduct a horizontal evaluation of oral liquids with similar colors. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a quality detection device for oral liquid of Chinese patent medicine, comprising a preprocessing module, an infrared spectrum detection module, a fingerprint spectrum detection module and an ultraviolet-visible spectroscopic spectrum detection module; a result analysis module and a cloud server, a database and a result feedback module are also provided;

[0006] The pre-processing module is connected to the infrared spectrum detection modules, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module; the pre-processing module is used to perform physical and chemical treatment on the received Chinese patent medicine oral liquid samples to ensure that they meet the requirements of subsequent testing; the processed samples are distributed to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module for different types of spectral analysis;

[0007] The infrared spectrum detection module is equipped with an infrared spectrometer for emitting infrared light to penetrate the sample and collect the infrared absorption spectrum information of the sample;

[0008] The fingerprint detection module integrates a high-performance liquid chromatograph to separate and detect the oral liquid of Chinese patent medicine to obtain its multi-component characteristic fingerprint.

[0009] The UV-visible spectrophotometer is used to measure the absorbance of the sample in the UV and visible regions.

[0010] The infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module are connected to the result analysis module; the result analysis module receives data from the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module, and uses spectrum analysis algorithms and statistical methods to comprehensively analyze the collected infrared spectrum, fingerprint spectrum and ultraviolet-visible spectrum data to determine the quality status of the sample;

[0011] The analysis module connects the cloud server and database; the cloud server is built on the cloud platform and stores the standard database, historical detection records, model parameters and user access rights management required by the detection system; it supports remote data upload, storage, backup and retrieval;

[0012] The database stores standard spectra of various Chinese patent medicine oral liquids;

[0013] The result feedback module generates a detailed test report based on the processing results of the analysis module and pushes it to the user through the network or mobile terminal.

[0014] The pretreatment module includes a liquid reaction chamber, a precipitation chamber and a supernatant extraction device;

[0015] The liquid reaction chamber is equipped with at least two input pipelines, the first one is used to introduce the Chinese patent medicine oral liquid sample to be processed, and the second one is used to add the reaction reagent; the liquid reaction chamber is equipped with a temperature control unit and a stirrer to perform temperature control treatment and sufficient mixing of the Chinese patent medicine oral liquid sample; the inner wall of the reaction chamber is made of corrosion-resistant material;

[0016] The liquid reaction chamber is connected to the precipitation chamber through an output pipe with an automatic valve;

[0017] The sedimentation chamber is connected to the liquid reaction chamber, and a filtering system is built in the top to remove suspended particles and insoluble matter in the sample after the reaction; a discharge port is provided at the bottom of the sedimentation chamber to discharge the settled impurities; the upper part of the sedimentation chamber is connected to the supernatant extraction device;

[0018] The supernatant extraction device is provided with a pumping system for extracting supernatant from the mixed liquid in the sedimentation chamber and supplying the supernatant to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet visible spectroscopic spectrum detection module respectively through an output pipeline.

[0019] The infrared spectrum detection module is equipped with a Fourier transform infrared spectrometer, including a light source, a Michelson interferometer, a sample chamber, a detector and an electronic control system; the light source generates continuous infrared light, which is divided into two beams through the interferometer to interfere with each other to form an interference spectrum, and the sample is placed in the sample chamber. After the light beam passes through the sample, the detector captures the interference signal; the sample chamber is designed with a temperature control and an automatic sampling device; the infrared spectrum detection module is equipped with a spectrum analysis software, which can perform qualitative and quantitative analysis on the absorption peak;

[0020] The fingerprint detection module includes a high-pressure infusion pump, an automatic sampler, a chromatographic column temperature control box, a detector and a data processing workstation; the high-pressure infusion pump ensures stable delivery of the mobile phase, the automatic sampler realizes unattended continuous analysis of samples, and the chromatographic column temperature control box is used to adjust the separation conditions and improve the separation effect; the detector can monitor multiple wavelengths at the same time, collect spectral information, and construct a fingerprint spectrum;

[0021] The UV-visible spectroscopic detection module consists of a light source, a monochromator, a sample chamber, a detector and a data processing system; the light source emits continuous UV-visible light, the monochromator decomposes the composite light into monochromatic light, the sample is placed in the sample chamber, absorbs or transmits light of a specific wavelength, and is converted into an electrical signal by the detector.

[0022] The result analysis module adopts a multi-step analysis strategy, combining the measured data of UV-visible spectrometry, infrared spectrum and fingerprint spectrum to analyze the quality of Chinese patent medicine oral liquid, as follows:

[0023] Using the absorption peak information collected by the UV-visible spectroscopic detection module, the absorption peak position of the sample is compared with the preset standard absorption peak database through a matching algorithm to preliminarily determine the subdivision type of the Chinese patent medicine oral liquid;

[0024] According to the initially identified subdivision type, the system automatically selects the standardized infrared spectrum and fingerprint of the corresponding type of Chinese patent medicine oral liquid;

[0025] The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, the peak shift analysis is used to quantify the differences, and these difference data are standardized;

[0026] The processed spectral difference data are input into a pre-trained quality evaluation model, and the model predicts and outputs the quality grade of the sample based on the difference data.

[0027] The result feedback module is arranged in the user's mobile terminal; the result feedback module integrates data transmission and processing functions, automatic report generation functions, notification functions and user interaction functions.

[0028] A method for detecting the quality of oral liquid of Chinese patent medicine, using the oral liquid of Chinese patent medicine quality detection device, comprises the following steps:

[0029] a) Pretreatment: Introduce the Chinese patent medicine oral liquid sample into the liquid reaction chamber through the input pipeline, and add the necessary reaction reagents through other pipelines; under the control of the temperature control unit, the sample is precisely temperature controlled and fully mixed with a stirrer; after the reaction is completed, the sample is sent to the precipitation chamber through the output pipeline controlled by the automatic valve;

[0030] b) Extraction of supernatant: In the sedimentation chamber, the suspended particles and insoluble matter in the sample are removed by the built-in filtration system, and the sample is allowed to settle. After sedimentation, the impurities at the bottom are discharged through the discharge port, and the supernatant is extracted by the pumping system of the supernatant extraction device and distributed to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module;

[0031] c) Spectral detection: In the infrared spectrum detection module, a Fourier transform infrared spectrometer is used to obtain an infrared absorption spectrum, a high performance liquid chromatograph is used to obtain a multi-component characteristic fingerprint spectrum in the fingerprint spectrum detection module, and the ultraviolet and visible light absorbance of the sample is measured in the ultraviolet and visible light spectrum detection module;

[0032] d) Data analysis: using the absorption peak information collected by the UV-visible spectroscopic detection module, the absorption peak position of the sample is compared with the preset standard absorption peak database through a matching algorithm to preliminarily determine the subdivision type of the Chinese patent medicine oral liquid; according to the preliminarily identified subdivision type, the system automatically selects the standard infrared spectrum and standard fingerprint spectrum of the Chinese patent medicine oral liquid of the corresponding subdivision type; the measured infrared spectrum data and fingerprint spectrum data are respectively compared with the selected standard infrared spectrum and standard fingerprint spectrum, and the peak shift analysis is used to quantify the difference, and these difference data are standardized;

[0033] e) Quality assessment: The processed spectral difference data is input into a pre-trained quality assessment model, and the model predicts and outputs the quality grade of the sample based on the difference data;

[0034] f) Report generation and feedback: The result analysis module generates a detailed test report, which contains the spectral data and quality grade evaluation results of the sample; the report is pushed to the user's mobile terminal through the result feedback module. The user can view the test results in real time and interact with the system as needed to query historical records or make further analysis requests.

[0035] The specific types of Chinese patent medicine oral liquid include:

[0036] The identified absorption peak position is matched with the preset "standard absorption peak database", which has previously included standard ultraviolet-visible absorption peak data of known ingredients of Chinese patent medicine oral liquids; when matching, the corresponding name of the Chinese patent medicine oral liquid is first selected from the database, and each name of the Chinese patent medicine oral liquid includes multiple standard absorbance spectra of different subdivided types; wherein the subdivided types are classified according to the absorption peak position of the Chinese patent medicine oral liquid, and are divided into at least 5 categories, and each subdivided type has its own standard infrared spectrum and standard fingerprint spectrum.

[0037] The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, and the peak shift analysis is used to quantify the differences, and the difference data are standardized, including:

[0038] Infrared spectrum data comparison and peak shift analysis:

[0039] Import the measured infrared spectrum data into the analysis software and accurately align the time or wave number axis with the selected standard infrared spectrum to ensure a consistent basis for comparison;

[0040] The main absorption peaks in the measured spectrum and standard spectrum are automatically marked using the peak recognition algorithm;

[0041] For each matched peak pair, the offset between the measured peak position and the standard peak position and the relative change of the peak area are calculated;

[0042] Fingerprint data comparison and difference analysis:

[0043] Time-calibrate the measured fingerprint data with the standard fingerprint to ensure the correct alignment of the chromatographic peaks;

[0044] Matching the chromatographic peaks in the measured fingerprint with the corresponding peaks in the standard fingerprint by retention time and peak shape;

[0045] Calculate the retention time shift between the measured chromatographic peak and the standard fingerprint peak, as well as the relative change of the peak area;

[0046] Standardization of difference data:

[0047] All quantified peak position shifts and peak area changes were normalized to obtain normalized peak position shift Δν and peak area change percentage ΔA.

[0048] The quality evaluation model of Chinese patent medicine oral liquid uses random forest as the core algorithm, with normalized peak position shift Δν and peak area change percentage ΔA as input, and the output is the quality grade of the sample.

[0049] The beneficial effects of the present invention are:

[0050] Through the comprehensive application of infrared spectroscopy, fingerprint mapping and ultraviolet-visible spectroscopic detection technology, comprehensive coverage of various ingredients in Chinese patent medicine oral liquid is achieved, the accuracy and reliability of detection are improved, and the chemical composition and structural information of the samples can be revealed more finely.

[0051] By using the absorbance information of the UV-visible spectrophotometer to make a preliminary judgment on the subdivision type of Chinese patent medicine oral liquid in the early stage of data analysis, the efficiency of the entire detection process is further enhanced and the targeted analysis results are more accurate.

[0052] The whole process from pre-processing to testing, data analysis, quality assessment to result feedback is automated, which reduces human operation errors and improves testing efficiency. Combining cloud servers and result analysis modules, advanced algorithms are used to automatically match and analyze data to achieve rapid quality judgment.

[0053] With the support of cloud servers, remote storage, backup and access of test data are realized, which facilitates the traceability and analysis of historical data. At the same time, the result feedback module directly pushes the test report to the user's mobile terminal, which improves the immediacy of information transmission and facilitates supervision and decision-making.

[0054] The quality evaluation model established using machine learning algorithms such as random forests can process normalized peak position shift and peak area change data, achieve accurate grading of the quality of Chinese patent medicine oral liquids, effectively distinguish products of different qualities, and help improve the overall quality control level of the industry.

[0055] Users can easily obtain test reports, conduct historical record queries and request in-depth analysis through mobile terminals, which enhances user participation and satisfaction and provides users with a one-stop, convenient quality testing solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0057] Attached Figure 1 It is a schematic diagram of the overall structure of the device of the present invention;

[0058] Attached Figure 2 It is a structural schematic diagram of the preprocessing module of the present invention.

[0059] Among them: 1 liquid reaction chamber, 2 precipitation chamber, 3 supernatant extraction device, 4 infrared spectrometer cuvette, 5 liquid chromatography injection pool, 6 UV-visible spectrophotometer cuvette, 7 sample container, 8 reagent container, 9 liquid reaction chamber valve, 10 precipitation chamber valve. DETAILED DESCRIPTION

[0060] Embodiment 1:

[0061] See also Figure 1 and Figure 2 The present invention provides a quality detection device for oral liquid of Chinese patent medicine, comprising a pre-processing module, an infrared spectrum detection module, a fingerprint spectrum detection module and an ultraviolet-visible spectroscopic spectrum detection module; a result analysis module and a cloud server, a database and a result feedback module are also provided;

[0062] The pre-processing module is connected to the infrared spectrum detection modules, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module; the pre-processing module is used to perform physical and chemical treatment on the received Chinese patent medicine oral liquid samples to ensure that they meet the requirements of subsequent testing; the processed samples are distributed to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module for different types of spectral analysis;

[0063] The infrared spectrum detection module is equipped with an infrared spectrometer for emitting infrared light to penetrate the sample and collect the infrared absorption spectrum information of the sample;

[0064] The fingerprint detection module integrates a high-performance liquid chromatograph to separate and detect the oral liquid of Chinese patent medicine to obtain its multi-component characteristic fingerprint.

[0065] The UV-visible spectrophotometer is used to measure the absorbance of the sample in the UV and visible regions.

[0066] The infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module are connected to the result analysis module; the result analysis module receives data from the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module, and uses spectrum analysis algorithms and statistical methods to comprehensively analyze the collected infrared spectrum, fingerprint spectrum and ultraviolet-visible spectrum data to determine the quality status of the sample;

[0067] The analysis module connects the cloud server and database; the cloud server is built on the cloud platform and stores the standard database, historical detection records, model parameters and user access rights management required by the detection system; it supports remote data upload, storage, backup and retrieval;

[0068] The database stores standard spectra of various Chinese patent medicine oral liquids;

[0069] The result feedback module generates a detailed test report based on the processing results of the analysis module and pushes it to the user through the network or mobile terminal.

[0070] The pretreatment module includes a liquid reaction chamber, a precipitation chamber and a supernatant extraction device;

[0071] The liquid reaction chamber is equipped with at least two input pipelines, the first one is used to introduce the Chinese patent medicine oral liquid sample to be processed, and the second one is used to add the reaction reagent; the liquid reaction chamber is equipped with a temperature control unit and a stirrer to perform temperature control treatment and sufficient mixing of the Chinese patent medicine oral liquid sample; the inner wall of the reaction chamber is made of corrosion-resistant material;

[0072] The liquid reaction chamber is connected to the precipitation chamber through an output pipe with an automatic valve;

[0073] The sedimentation chamber is connected to the liquid reaction chamber, and a filtering system is built in the top to remove suspended particles and insoluble matter in the sample after the reaction; a discharge port is provided at the bottom of the sedimentation chamber to discharge the settled impurities; the upper part of the sedimentation chamber is connected to the supernatant extraction device;

[0074] The supernatant extraction device is provided with a pumping system for extracting supernatant from the mixed liquid in the sedimentation chamber and supplying the supernatant to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet visible spectroscopic spectrum detection module respectively through an output pipeline.

[0075] The infrared spectrum detection module is equipped with a Fourier transform infrared spectrometer, including a light source, a Michelson interferometer, a sample chamber, a detector and an electronic control system; the light source generates continuous infrared light, which is divided into two beams through the interferometer to interfere with each other to form an interference spectrum, and the sample is placed in the sample chamber. After the light beam passes through the sample, the detector captures the interference signal; the sample chamber is designed with a temperature control and an automatic sampling device; the infrared spectrum detection module is equipped with a spectrum analysis software, which can perform qualitative and quantitative analysis on the absorption peak;

[0076] The fingerprint detection module includes a high-pressure infusion pump, an automatic sampler, a chromatographic column temperature control box, a detector and a data processing workstation; the high-pressure infusion pump ensures stable delivery of the mobile phase, the automatic sampler realizes unattended continuous analysis of samples, and the chromatographic column temperature control box is used to adjust the separation conditions and improve the separation effect; the detector can monitor multiple wavelengths at the same time, collect spectral information, and construct a fingerprint spectrum;

[0077] The UV-visible spectroscopic detection module consists of a light source, a monochromator, a sample chamber, a detector and a data processing system; the light source emits continuous UV-visible light, the monochromator decomposes the composite light into monochromatic light, the sample is placed in the sample chamber, absorbs or transmits light of a specific wavelength, and is converted into an electrical signal by the detector.

[0078] The result analysis module adopts a multi-step analysis strategy, combining the measured data of UV-visible spectrometry, infrared spectrum and fingerprint spectrum to analyze the quality of Chinese patent medicine oral liquid, as follows:

[0079] Using the absorption peak information collected by the UV-visible spectroscopic detection module, the absorption peak position of the sample is compared with the preset standard absorption peak database through a matching algorithm to preliminarily determine the subdivision type of the Chinese patent medicine oral liquid;

[0080] According to the initially identified subdivision type, the system automatically selects the standardized infrared spectrum and fingerprint of the corresponding type of Chinese patent medicine oral liquid;

[0081] The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, the peak shift analysis is used to quantify the differences, and these difference data are standardized;

[0082] The processed spectral difference data are input into a pre-trained quality evaluation model, and the model predicts and outputs the quality grade of the sample based on the difference data.

[0083] The result feedback module is arranged in the user's mobile terminal; the result feedback module integrates data transmission and processing functions, automatic report generation functions, notification functions and user interaction functions.

[0084] Embodiment 2:

[0085] A method for detecting the quality of oral liquid of Chinese patent medicine, using the oral liquid of Chinese patent medicine quality detection device, comprises the following steps:

[0086] a) Pretreatment: Introduce the Chinese patent medicine oral liquid sample into the liquid reaction chamber through the input pipeline, and add the necessary reaction reagents through other pipelines; under the control of the temperature control unit, the sample is precisely temperature controlled and fully mixed with a stirrer; after the reaction is completed, the sample is sent to the precipitation chamber through the output pipeline controlled by the automatic valve;

[0087] b) Extraction of supernatant: In the sedimentation chamber, the suspended particles and insoluble matter in the sample are removed by the built-in filtration system, and the sample is allowed to settle. After sedimentation, the impurities at the bottom are discharged through the discharge port, and the supernatant is extracted by the pumping system of the supernatant extraction device and distributed to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module;

[0088] c) Spectral detection: In the infrared spectrum detection module, a Fourier transform infrared spectrometer is used to obtain an infrared absorption spectrum, a high performance liquid chromatograph is used to obtain a multi-component characteristic fingerprint spectrum in the fingerprint spectrum detection module, and the ultraviolet and visible light absorbance of the sample is measured in the ultraviolet and visible light spectrum detection module;

[0089] d) Data analysis: using the absorption peak information collected by the UV-visible spectroscopic detection module, the absorption peak position of the sample is compared with the preset standard absorption peak database through a matching algorithm to preliminarily determine the subdivision type of the Chinese patent medicine oral liquid; according to the preliminarily identified subdivision type, the system automatically selects the standard infrared spectrum and standard fingerprint spectrum of the Chinese patent medicine oral liquid of the corresponding subdivision type; the measured infrared spectrum data and fingerprint spectrum data are respectively compared with the selected standard infrared spectrum and standard fingerprint spectrum, and the peak shift analysis is used to quantify the difference, and these difference data are standardized;

[0090] e) Quality assessment: The processed spectral difference data is input into a pre-trained quality assessment model, and the model predicts and outputs the quality grade of the sample based on the difference data;

[0091] f) Report generation and feedback: The result analysis module generates a detailed test report, which contains the spectral data and quality grade evaluation results of the sample; the report is pushed to the user's mobile terminal through the result feedback module. The user can view the test results in real time and interact with the system as needed to query historical records or make further analysis requests.

[0092] The specific types of Chinese patent medicine oral liquid include:

[0093] The identified absorption peak position is matched with the preset "standard absorption peak database", which has previously included standard ultraviolet-visible absorption peak data of known ingredients of Chinese patent medicine oral liquids; when matching, the corresponding name of the Chinese patent medicine oral liquid is first selected from the database, and each name of the Chinese patent medicine oral liquid includes multiple standard absorbance spectra of different subdivided types; wherein the subdivided types are classified according to the absorption peak position of the Chinese patent medicine oral liquid, and are divided into at least 5 categories, and each subdivided type has its own standard infrared spectrum and standard fingerprint spectrum.

[0094] The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, and the peak shift analysis is used to quantify the differences, and the difference data are standardized, including:

[0095] Infrared spectrum data comparison and peak shift analysis:

[0096] Import the measured infrared spectrum data into the analysis software and accurately align the time or wave number axis with the selected standard infrared spectrum to ensure a consistent basis for comparison;

[0097] The main absorption peaks in the measured spectrum and standard spectrum are automatically marked using the peak recognition algorithm;

[0098] For each matched peak pair, the offset between the measured peak position and the standard peak position and the relative change of the peak area are calculated;

[0099] Fingerprint data comparison and difference analysis:

[0100] Time-calibrate the measured fingerprint data with the standard fingerprint to ensure the correct alignment of the chromatographic peaks;

[0101] Matching the chromatographic peaks in the measured fingerprint with the corresponding peaks in the standard fingerprint by retention time and peak shape;

[0102] Calculate the retention time shift between the measured chromatographic peak and the standard fingerprint peak, as well as the relative change of the peak area;

[0103] Standardization of difference data:

[0104] All quantified peak position shifts and peak area changes were normalized to obtain normalized peak position shift Δν and peak area change percentage ΔA.

[0105] The quality evaluation model of Chinese patent medicine oral liquid uses random forest as the core algorithm, with normalized peak position shift Δν and peak area change percentage ΔA as input, and the output is the quality grade of the sample.

[0106] Embodiment 3:

[0107] The model is as follows:

[0108] When using Random Forest (RF) as the core algorithm to build a quality evaluation model for Chinese patent medicine oral liquid, its structure is implemented according to the following key steps:

[0109] 1. Data preparation and feature engineering

[0110] Input features: The input data of the model includes two key features, namely the normalized peak position shift (Δν) and the percentage change of peak area (ΔA). These two features have been pre-processed and normalized to ensure their comparability and consistency in the model.

[0111] 2. Random Forest Model Architecture

[0112] Forest construction: Random forests are composed of multiple decision trees, each of which is trained independently. When constructing each tree, RF uses the bootstrap sampling method to randomly extract a certain proportion (66%-70%) of samples from the original training set to train the tree, thereby increasing the diversity of the model.

[0113] Random feature selection: At each node of the tree, instead of considering all features, only a part of all features is randomly selected (for example, sqrt(n), where n is the total number of features) to determine the best split features and split points. This step further enhances the randomness and anti-overfitting ability of the model.

[0114] Depth and complexity control: Each decision tree grows until a certain stopping condition is met, such as the number of samples in the node is lower than a certain threshold, the predetermined maximum depth is reached, or the information gain (or other splitting criteria) is lower than a certain threshold.

[0115] 3. Model learning and prediction

[0116] Parallel learning: Since each tree is constructed independently, the process can be executed in parallel, greatly improving the speed of model training.

[0117] Majority voting: For classification tasks, when predicting a new Chinese patent medicine oral solution sample, each tree will give a predicted category. The final prediction result is obtained by majority voting of all tree prediction results, that is, whichever category receives the most votes is considered the final prediction category, thus determining the quality level of the sample.

[0118] 4. Hyperparameter Tuning

[0119] Number of trees (n_estimators): determines the total number of trees in the forest. Generally, more trees can reduce the variance of the model, but also increase the computational cost.

[0120] Max_features: determines the proportion of features considered at each split. Adjusting this parameter can affect the bias-variance balance of the model.

[0121] The depth of the tree (max_depth) or the minimum number of samples of leaf nodes (min_samples_leaf): control the complexity of the tree to avoid overfitting.

[0122] 5. Model evaluation and validation

[0123] Cross-validation: Use methods such as k-fold cross-validation to evaluate the generalization ability of the model by splitting the dataset into k subsets, taking turns using one of the subsets as the validation set and the rest as the training set, repeating this k times and averaging the results.

[0124] Performance indicators: Select appropriate evaluation indicators according to actual needs, such as accuracy, recall, F1 score, precision, etc., to measure model performance.

[0125] Thus far, the description of the above-described embodiments is provided for the purpose of illustration and description. It is not intended to be exhaustive or to limit the present disclosure. The individual elements or features of a specific embodiment are generally not subject to the limitations of the specific embodiment, but when applicable, they can be interchanged and used for selected embodiments even if not specifically shown or described. In many respects, the same element or feature also can be changed. This variation is not considered to be a departure from the present disclosure, and all such modifications are intended to be included in the scope of the present disclosure.

[0126] Example embodiments are provided so that the present disclosure will be thorough and the scope will be fully conveyed to those skilled in the art. In order to thoroughly understand the embodiments of the present disclosure, numerous details are set forth, such as examples of specific parts, devices, and methods. Obviously, for those skilled in the art, it is not necessary to use specific details, the example embodiments can be implemented in many different forms, and neither should be construed as limiting the scope of the present disclosure. In certain example embodiments, well-known processes, well-known device structures, and well-known technologies are not described in detail.

[0127] Here, specialized vocabulary is used only for the purpose of describing specific example embodiments, and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "a" and "the" used herein may be intended to include plural forms as well. The terms "including" and "having" are inclusive, and therefore specify the presence of the claimed features, wholes, steps, operations, elements and / or components, but do not exclude the presence or additional presence of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof. Unless the order of execution is clearly indicated, the method steps, processing and operations described herein are not interpreted as necessarily needing to be performed in the specific order discussed and shown. It should also be understood that additional or optional steps may be adopted.

Claims

1. A quality detection device for oral liquid of Chinese patent medicine, comprising a pretreatment module, an infrared spectrum detection module, a fingerprint spectrum detection module and an ultraviolet-visible spectroscopic spectrum detection module; characterized in that: A result analysis module, a cloud server, a database and a result feedback module are also provided; The pre-processing module is connected to each infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet visible spectroscopic spectrum detection module; The pre-processing module is used to perform physical and chemical treatment on the received Chinese patent medicine oral liquid samples to ensure that they meet the requirements of subsequent testing; the processed samples are assigned to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module for different types of spectral analysis; The infrared spectrum detection module is equipped with an infrared spectrometer for emitting infrared light to penetrate the sample and collect the infrared absorption spectrum information of the sample; The fingerprint detection module integrates a high-performance liquid chromatograph to separate and detect the oral liquid of Chinese patent medicine to obtain its multi-component characteristic fingerprint. The UV-visible spectrophotometer is used to measure the absorbance of the sample in the UV and visible regions. The infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module are connected to the result analysis module; the result analysis module receives data from the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module, and uses spectrum analysis algorithms and statistical methods to comprehensively analyze the collected infrared spectrum, fingerprint spectrum and ultraviolet-visible spectrum data to determine the quality status of the sample; The result analysis module connects the cloud server and database; the cloud server is built on the cloud platform and stores the standard database, historical test records, model parameters and user access rights management required by the detection system; it supports remote data upload, storage, backup and retrieval; The database stores standard spectra of various Chinese patent medicine oral liquids; The result feedback module generates a detailed test report based on the processing results of the result analysis module and pushes it to the user through the network or mobile terminal; The result analysis module adopts a multi-step analysis strategy, combining the measured data of UV-visible spectrometry, infrared spectrum and fingerprint spectrum to analyze the quality of Chinese patent medicine oral liquid, as follows: Using the absorption peak information collected by the UV-visible spectroscopic detection module, the absorption peak position of the sample is compared with the preset standard absorption peak database through a matching algorithm to preliminarily determine the subdivision type of the Chinese patent medicine oral liquid; According to the initially identified subdivision type, the system automatically selects the standardized infrared spectrum and fingerprint of the corresponding type of Chinese patent medicine oral liquid; The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, the peak shift analysis is used to quantify the differences, and these difference data are standardized; The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, and the peak shift analysis is used to quantify the differences, and the difference data are standardized, including: Infrared spectrum data comparison and peak shift analysis: Import the measured infrared spectrum data into the analysis software and accurately align the time or wave number axis with the selected standard infrared spectrum to ensure a consistent basis for comparison; The main absorption peaks in the measured spectrum and standard spectrum are automatically marked using the peak recognition algorithm; For each matched peak pair, the offset between the measured peak position and the standard peak position and the relative change of the peak area are calculated; Fingerprint data comparison and difference analysis: Time-calibrate the measured fingerprint data with the standard fingerprint to ensure the correct alignment of the chromatographic peaks; Matching the chromatographic peaks in the measured fingerprint with the corresponding peaks in the standard fingerprint by retention time and peak shape; Calculate the retention time shift between the measured chromatographic peak and the standard fingerprint peak, as well as the relative change of the peak area; Standardization of difference data: All quantified peak shifts and peak area changes were normalized to obtain normalized peak shift Δν and peak area change percentage ΔA; The processed spectral difference data is input into a pre-trained quality evaluation model, and the model predicts and outputs the quality grade of the sample based on the difference data; The quality evaluation model of Chinese patent medicine oral liquid uses random forest as the core algorithm, with normalized peak position shift Δν and peak area change percentage ΔA as input, and the output is the quality grade of the sample.

2. The quality detection device for oral liquid of Chinese patent medicine according to claim 1, characterized in that: The pretreatment module includes a liquid reaction chamber, a precipitation chamber and a supernatant extraction device; The liquid reaction chamber is equipped with at least two input pipelines, the first one is used to introduce the Chinese patent medicine oral liquid sample to be processed, and the second one is used to add the reaction reagent; the liquid reaction chamber is equipped with a temperature control unit and a stirrer to perform temperature control treatment and sufficient mixing of the Chinese patent medicine oral liquid sample; the inner wall of the reaction chamber is made of corrosion-resistant material; The liquid reaction chamber is connected to the precipitation chamber through an output pipe with an automatic valve; The sedimentation chamber is connected to the liquid reaction chamber, and a filtering system is built in the top to remove suspended particles and insoluble matter in the sample after the reaction; a discharge port is provided at the bottom of the sedimentation chamber to discharge the settled impurities; the upper part of the sedimentation chamber is connected to the supernatant extraction device; The supernatant extraction device is provided with a pumping system for extracting supernatant from the mixed liquid in the sedimentation chamber and supplying the supernatant to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet visible spectroscopic spectrum detection module respectively through an output pipeline.

3. The quality detection device for oral liquid of Chinese patent medicine according to claim 2, characterized in that: The infrared spectrum detection module is equipped with a Fourier transform infrared spectrometer, including a light source, a Michelson interferometer, a sample chamber, a detector and an electronic control system; the light source generates continuous infrared light, which is divided into two beams through the interferometer to interfere with each other to form an interference spectrum, and the sample is placed in the sample chamber. After the light beam passes through the sample, the detector captures the interference signal; the sample chamber is designed with a temperature control and an automatic sampling device; the infrared spectrum detection module is equipped with a spectrum analysis software, which can perform qualitative and quantitative analysis on the absorption peak; The fingerprint detection module includes a high-pressure infusion pump, an automatic sampler, a chromatographic column temperature control box, a detector and a data processing workstation; the high-pressure infusion pump ensures stable delivery of the mobile phase, the automatic sampler realizes unattended continuous analysis of samples, and the chromatographic column temperature control box is used to adjust the separation conditions and improve the separation effect; the detector can monitor multiple wavelengths at the same time, collect spectral information, and construct a fingerprint spectrum; The UV-visible spectroscopic detection module consists of a light source, a monochromator, a sample chamber, a detector and a data processing system; the light source emits continuous UV-visible light, the monochromator decomposes the composite light into monochromatic light, the sample is placed in the sample chamber, absorbs or transmits light of a specific wavelength, and is converted into an electrical signal by the detector.

4. The quality detection device for oral liquid of Chinese patent medicine according to claim 3, characterized in that: The result feedback module is arranged in the user's mobile terminal; the result feedback module integrates data transmission and processing functions, automatic report generation functions, notification functions and user interaction functions.

5. A method for detecting the quality of oral liquid of Chinese patent medicine, using the device for detecting the quality of oral liquid of Chinese patent medicine according to claim 4, characterized in that The steps include: a) Pretreatment: Introduce the Chinese patent medicine oral liquid sample into the liquid reaction chamber through the input pipeline, and add the necessary reaction reagents through other pipelines; under the control of the temperature control unit, the sample is precisely temperature controlled and fully mixed with a stirrer; After the reaction is completed, the sample is sent to the precipitation chamber through the output pipeline controlled by the automatic valve; b) Extraction of supernatant: In the sedimentation chamber, the suspended particles and insoluble matter in the sample are removed by the built-in filtration system, and the sample is allowed to settle. After sedimentation, the impurities at the bottom are discharged through the discharge port, and the supernatant is extracted by the pumping system of the supernatant extraction device and distributed to the infrared spectrum detection module, the fingerprint spectrum detection module and the ultraviolet-visible spectroscopic spectrum detection module; c) Spectral detection: In the infrared spectrum detection module, a Fourier transform infrared spectrometer is used to obtain an infrared absorption spectrum, a high performance liquid chromatograph is used to obtain a multi-component characteristic fingerprint spectrum in the fingerprint spectrum detection module, and the ultraviolet and visible light absorbance of the sample is measured in the ultraviolet and visible light spectrum detection module; d) Data analysis: using the absorption peak information collected by the UV-visible spectroscopic detection module, the absorption peak position of the sample is compared with the preset standard absorption peak database through a matching algorithm to preliminarily determine the subdivision type of the Chinese patent medicine oral liquid; according to the preliminarily identified subdivision type, the system automatically selects the standard infrared spectrum and standard fingerprint spectrum of the Chinese patent medicine oral liquid of the corresponding subdivision type; the measured infrared spectrum data and fingerprint spectrum data are respectively compared with the selected standard infrared spectrum and standard fingerprint spectrum, and the peak shift analysis is used to quantify the difference, and these difference data are standardized; The measured infrared spectrum data and fingerprint spectrum data are compared with the selected standard infrared spectrum and standard fingerprint spectrum respectively, and the peak shift analysis is used to quantify the differences, and the difference data are standardized, including: Infrared spectrum data comparison and peak shift analysis: Import the measured infrared spectrum data into the analysis software and accurately align the time or wave number axis with the selected standard infrared spectrum to ensure a consistent basis for comparison; The main absorption peaks in the measured spectrum and standard spectrum are automatically marked using the peak recognition algorithm; For each matched peak pair, the offset between the measured peak position and the standard peak position and the relative change of the peak area are calculated; Fingerprint data comparison and difference analysis: Time-calibrate the measured fingerprint data with the standard fingerprint to ensure the correct alignment of the chromatographic peaks; Matching the chromatographic peaks in the measured fingerprint with the corresponding peaks in the standard fingerprint by retention time and peak shape; Calculate the retention time shift between the measured chromatographic peak and the standard fingerprint peak, as well as the relative change of the peak area; Standardization of difference data: All quantified peak shifts and peak area changes were normalized to obtain normalized peak shift Δν and peak area change percentage ΔA; e) Quality assessment: The processed spectral difference data is input into a pre-trained quality assessment model, and the model predicts and outputs the quality grade of the sample based on the difference data; The quality evaluation model of Chinese patent medicine oral liquid uses random forest as the core algorithm, with normalized peak position shift Δν and peak area change percentage ΔA as input, and the output is the quality grade of the sample; f) Report generation and feedback: The result analysis module generates a detailed test report, which contains the spectral data and quality grade evaluation results of the sample; the report is pushed to the user's mobile terminal through the result feedback module. The user can view the test results in real time and interact with the system as needed to query historical records or make further analysis requests.

6. A method for quality testing of oral liquid Chinese patent medicine according to claim 5, characterized in that: The specific types of Chinese patent medicine oral liquid include: Match the identified absorption peak position with the preset "standard absorption peak database", which contains the standard UV-visible absorption peak data of known Chinese patent medicine oral liquid ingredients in advance; when matching, first select the name of the corresponding Chinese patent medicine oral liquid from the database, and each Chinese patent medicine oral liquid name includes multiple standard absorbance spectra of different subdivided types; Among them, the subdivided types are classified according to the absorption peak position of the Chinese patent medicine oral liquid, and are divided into at least 5 categories. Each subdivided type has its own standard infrared spectrum and standard fingerprint spectrum.

Images