A platform for identifying the authenticity of traditional Chinese medicine

Abstract

The present invention discloses a Chinese medicinal material authenticity identification platform, which relates to the technical field of Chinese medicinal material authenticity identification, including a data acquisition module, wherein the data acquisition module is connected to a data analysis module, wherein the data analysis module is connected to a result display module; the data acquisition module is used to collect appearance images, spectral data and background information of Chinese medicinal materials. The Chinese medicinal material authenticity identification platform comprehensively analyzes Chinese medicinal materials from multiple angles, avoids the limitations of a single method, and improves the accuracy of identification; it also deeply analyzes the characteristics of Chinese medicinal materials, accurately judges the authenticity and quality; by establishing a quantitative analysis model, it realizes an accurate assessment of the quality of Chinese medicinal materials; the model is verified by a cross-validation method to ensure accuracy and stability, improve the accuracy of identification, and solve the problem that traditional identification methods are difficult to meet the needs of quickly and accurately identifying the authenticity of Chinese medicinal materials.

Description

Technical Field

[0001] The present invention relates to the technical field of authenticity identification of traditional Chinese medicines, and in particular to a traditional Chinese medicine authenticity identification platform. Background Art

[0002] Chinese medicinal materials occupy an important position in traditional Chinese medicine. Their quality is directly related to the clinical treatment effect and the health and safety of patients. At present, there is a problem of mixed authenticity in the Chinese medicinal materials market. At present, the authenticity identification of traditional Chinese medicinal materials mainly relies on the experience of professionals and limited traditional identification methods. For example, trait identification judges the authenticity by observing the appearance characteristics of the medicinal materials such as shape, size, color, and texture, but it is highly subjective and requires the identifier to have rich experience and professional knowledge, and different people may come to different conclusions; microscopic identification uses a microscope to observe the characteristics of the medicinal materials' tissue structure, cell morphology, etc., which also requires a high level of professionalism from the identifier; there are also physical and chemical identification and other methods. The traditional identification method is complex to operate and time-consuming, and is not suitable for large-scale rapid identification. With the expansion of the Chinese medicinal materials market and the increase in variety, traditional identification methods are difficult to meet the needs of rapid and accurate identification of the authenticity of Chinese medicinal materials. On the one hand, the number of professional identification personnel is limited, and they cannot cope with the huge market demand for Chinese medicinal materials. On the other hand, traditional identification methods are inefficient and easily affected by human and environmental factors, and the accuracy and reliability of the identification results are difficult to guarantee. Therefore, there is an urgent need for an efficient, accurate and objective Chinese medicinal materials authenticity identification platform to solve the authenticity identification problem existing in the current Chinese medicinal materials market and ensure the quality of Chinese medicinal materials and clinical drug safety. Summary of the invention

[0003] In order to solve the above technical problems, the present invention is implemented through the following technical solutions: a Chinese herbal medicine authenticity identification platform, comprising a data acquisition module, the data acquisition module is connected to a data analysis module, the data analysis module is connected to a result display module;

[0004] The data acquisition module is used to collect appearance images, spectral data and background information of Chinese medicinal materials;

[0005] The data analysis module uses artificial intelligence algorithms to analyze and process the collected data to determine the authenticity of the Chinese medicinal materials;

[0006] The result display module is used to display the identification results to the user, including the authenticity judgment results and the similarity score.

[0007] Preferably, the data acquisition module includes an appearance acquisition unit, a spectrum acquisition unit and an information collection unit;

[0008] The appearance acquisition unit selects a high-resolution image acquisition device and calibrates it, selects a representative Chinese medicinal material sample and places it on a preset background, adjusts the device angle, distance and lighting conditions for shooting, controls the lighting intensity and direction to avoid shadow reflections, stores the image in JPEG format, obtains the appearance image and transmits it to the data analysis module;

[0009] Select high-resolution image acquisition equipment to ensure that it can capture clear details and accurate colors; calibrate the image acquisition equipment, including color calibration and focal length adjustment, to ensure stable image quality; select representative Chinese medicinal materials samples to ensure the integrity and cleanliness of the samples and avoid the impact of impurities and damage on image acquisition; and place the Chinese medicinal materials samples on a preset background, where the preset background color has a clear contrast with the Chinese medicinal materials to highlight the characteristics of the Chinese medicinal materials; adjust the angle, distance and lighting conditions of the image acquisition equipment to ensure that all parts of the Chinese medicinal materials, including shape, color, and texture characteristics, can be fully and clearly photographed. ; Use multi-angle shooting to obtain different perspective images of Chinese medicinal materials for subsequent analysis; use a fixed shooting angle sequence to ensure the consistency of image acquisition for different samples; control the intensity and direction of light to avoid the impact of shadows and reflections on image quality; use soft natural light or professional photography lights to ensure uniform lighting; store the collected high-resolution images in JPEG format to obtain appearance images to ensure that image quality is not lost; and transmit the images to the data analysis module via wired or wireless means for subsequent processing; use data transmission software or network storage services to ensure the stability and security of image transmission;

[0010] The spectrum acquisition unit uses spectrum analysis technology to collect spectrum data of Chinese medicinal materials to reflect their chemical composition information;

[0011] The information collection unit is used to collect background information of Chinese medicinal materials, and the background information includes the origin and growth environment of the Chinese medicinal materials.

[0012] Preferably, during the period when the spectral acquisition unit collects the spectral data of the Chinese medicinal materials, a near-infrared spectrometer is selected and calibrated and debugged, representative Chinese medicinal material samples are selected and processed and placed in the measurement area, the wavelength range is selected as the measurement parameter, the equipment is started to measure and record the spectral response data and the average value is obtained by multiple measurements, and after pre-processing the collected data, the chemical composition information of the Chinese medicinal materials is analyzed and extracted using chemometric methods, and a quantitative analysis model is established to predict the content of specific components in the Chinese medicinal materials through spectral data, which is compared with standard spectral data to judge the authenticity and quality, and the results are stored to generate a spectral analysis report, including a spectrum diagram, chemical composition information and authenticity judgment results.

[0013] Preferably, the process of establishing the quantitative analysis model is as follows:

[0014] Establish a quantitative analysis model: collect a large number of spectral data of Chinese herbal medicine samples with known component contents. The Chinese herbal medicine samples cover Chinese herbal medicines from different origins, batches, and quality grades to ensure data diversity and representativeness; at the same time, determine the content of specific components in the Chinese herbal medicine samples through chemical analysis methods as the training target of the model; pre-process the collected spectral data to remove noise and perform baseline correction to improve data quality; use feature extraction algorithms to extract characteristic variables related to the content of specific components from the spectral data, including absorbance values ​​at specific wavelengths and shape characteristics of spectral curves; select the partial least squares PLS quantitative analysis model;

[0015] The model is trained and optimized using the extracted characteristic variables as input and the known content of specific components as output; the model parameters are adjusted so that the model can accurately predict the content of specific components in unknown samples; the established quantitative analysis model is verified using the cross-validation method; a portion of known samples are used as validation sets, predicted using the trained model, and compared with the actual component content to evaluate the accuracy and reliability of the model.

[0016] Preferably, the process of comparing with known Chinese herbal medicine standard spectral data to determine authenticity and quality is as follows:

[0017] Collect a large amount of standard spectrum data of Chinese herbal medicines with known authenticity and quality grades, and establish a standard spectrum database; the standard spectrum data comes from authoritative institutions, reliable experimental research or strictly verified samples; collect spectra of the Chinese herbal medicine samples to be tested to obtain their spectrum data; compare the spectrum data of the sample to be tested with the standard spectrum data in the standard spectrum database, use cosine similarity as the spectrum similarity calculation method, and calculate the similarity value between the sample to be tested and the standard spectrum; judge the authenticity of the Chinese herbal medicine to be tested based on the similarity value;

[0018] A similarity threshold is set. When the similarity value between the sample to be tested and the standard authentic medicinal material spectrum is higher than the threshold, it can be preliminarily judged as an authentic medicinal material; when the similarity value is lower than the threshold, it is judged as a fake medicinal material or a suspicious medicinal material; for samples judged to be authentic medicinal materials, their quality grades are further evaluated based on their similarity values ​​with standard spectra of different quality grades; samples with high similarity to high-quality standard spectra are considered to be high-quality medicinal materials, and samples with high similarity to low-quality standard spectra are considered to be low-quality medicinal materials; the judgment results are presented in the form of a report, including the authenticity judgment of the Chinese medicinal materials, the quality grade assessment and the information on the similarity value with the standard spectrum, providing a basis for the identification and quality control of Chinese medicinal materials.

[0019] Preferably, the data analysis module includes a comparison unit and an identification unit:

[0020] The comparison unit obtains image data based on the acquired appearance image after standard processing, and then uses an artificial intelligence algorithm to analyze the collected image data, automatically identifies the appearance characteristics of the Chinese medicinal materials, and introduces a similarity judgment mechanism to compare the appearance characteristics with known authentic sample characteristics to obtain a comparison result; wherein the similarity judgment mechanism is composed of DNA barcode technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology NIR common components;

[0021] The identification unit analyzes the collected spectral data, determines the chemical composition of the Chinese medicinal materials, obtains the analysis results, and conducts a comprehensive analysis of the comparison results and the analysis results in combination with background information to determine whether they meet the characteristics of the authentic product and obtain the identification results; by conducting a comprehensive analysis in combination with background information, the accuracy of identification is improved.

[0022] Preferably, the comparison unit obtains image data based on the acquired appearance image through standard processing in the following process:

[0023] The median filtering method is used to denoise the acquired Chinese medicinal materials appearance images, remove noise points in the image, and improve the image quality; the histogram equalization method is used to enhance the image, enhance the contrast and clarity of the image, and highlight the characteristics of the Chinese medicinal materials; the image is uniformly adjusted to the set size and resolution, and the size and resolution of the image are adjusted to meet the requirements of subsequent analysis; the image is converted into the HSV color space, and the color value is normalized, the color of the image is standardized, and the color consistency between different images is ensured to obtain image data.

[0024] Preferably, during the analysis of image data using an artificial intelligence algorithm, a convolutional neural network (CNN) is used to extract features from the standardized image data; CNN can automatically learn the features of Chinese medicinal materials images, including shape, color, and texture; the extracted features are a combination of local and global features of the image to comprehensively describe the appearance characteristics of the Chinese medicinal materials; the principal component analysis (PCA) method is used to reduce the dimension of the features, remove redundant information, analyze and process the extracted features, and use a clustering algorithm to classify the features, group Chinese medicinal materials with similar appearance features into one category, and obtain the appearance characteristics of the Chinese medicinal materials.

[0025] Preferably, based on the similarity judgment mechanism being composed of DNA barcode technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology NIR common components, the process of introducing the similarity judgment mechanism to obtain the comparison result is as follows:

[0026] S1: Evaluation process through DNA barcoding technology: Collect samples of Chinese herbal medicines to be tested and samples of known authenticity, and extract DNA from the samples;

[0027] A DNA barcode region is selected as a target DNA region, and a short single-stranded nucleic acid sequence is designed according to the target DNA region for PCR amplification to obtain a DNA barcode sequence;

[0028] Sequence the amplified DNA barcode sequence, compare the sequence of the sample to be tested with the database of known authentic samples, calculate the sequence similarity using the basic local alignment search tool, and obtain the similarity value of the DNA barcode sequence;

[0029] S2: Chemical fingerprint similarity evaluation process:

[0030] Extract and purify the Chinese medicinal materials to be tested and the samples with known authenticity, and use chromatography technology to obtain chemical fingerprints, including high performance liquid chromatography fingerprints and gas chromatography-mass spectrometry fingerprints;

[0031] Extract features from chemical fingerprints, including peak area, retention time, and relative content;

[0032] The cosine similarity algorithm is used to calculate the similarity between the chemical fingerprint of the sample to be tested and the known true and false samples, and the similarity value of the chemical fingerprint is obtained;

[0033] S3: Near infrared spectroscopy (NIR) evaluation process:

[0034] Collect near-infrared spectra of the Chinese medicinal materials to be tested and the known authenticity samples, perform denoising and baseline correction on the collected near-infrared spectrum data, and extract the absorbance characteristics at specific wavelengths of the near-infrared spectrum;

[0035] A quantitative analysis model is established based on the partial least squares method. The near-infrared spectrum characteristics of the sample to be tested are input into the model and compared with known true and false samples to obtain the similarity judgment results of the near-infrared spectrum.

[0036] S4: The comparison results are obtained by integrating similarity evaluation mechanism:

[0037] According to the accuracy, reliability and applicability of the evaluation methods in steps S1, S2 and S3, corresponding weights are assigned to DNA barcoding technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology;

[0038] The similarity values ​​obtained by the three evaluation methods are multiplied by the corresponding weights, and the weighted values ​​are summed to obtain a comprehensive similarity score. A threshold is set according to the comprehensive similarity score to determine the authenticity of the Chinese medicinal materials to be tested;

[0039] If the comprehensive similarity score is higher than the threshold of the true sample, it is judged to be true; if it is lower than the threshold of the false sample, it is judged to be false; if it is in the middle range, it is suspicious and awaits further analysis; finally, a comparison result based on the similarity judgment mechanism is obtained, which provides an accurate basis for the authenticity identification of Chinese medicinal materials.

[0040] Preferably, the process of obtaining the identification result is as follows:

[0041] The median filtering algorithm is used to denoise the collected spectral data, remove the noise caused by instrument errors and environmental interference, and improve the quality of the data; the polynomial fitting method is used to perform baseline correction to eliminate the baseline drift phenomenon in the spectral data and make the spectral curve more stable; the features that can reflect the chemical composition of Chinese medicinal materials are extracted from the pre-processed spectral data, including the extraction of absorbance values ​​at specific wavelengths, peak values ​​and valley values ​​of the spectral curve; the principal component analysis PCA in the chemometric method is used to reduce the dimension of the spectral data, extract the main characteristic variables, and reduce the complexity of the data; based on the extracted features, combined with the known spectral database and chemical knowledge, the extracted features are compared with the standard spectrum to identify the spectral features corresponding to the specific chemical components and determine the chemical composition of the Chinese medicinal materials;

[0042] The results obtained from the appearance image comparison, the chemical composition results obtained from the spectral data analysis, and the background information of the Chinese medicinal materials are integrated, among which the appearance image comparison results include similarity scores and authenticity judgments, and the background information includes the origin and growth environment; according to the importance and reliability of different information, corresponding weights are assigned to the appearance image comparison results, spectral data analysis results, and background information; among them, the appearance image comparison results have a higher weight in judging the variety and authenticity of Chinese medicinal materials, while the spectral data analysis results are more important in determining the chemical composition and quality, and the background information can be used as a basis for auxiliary judgment; for each judgment indicator, a weighted calculation is performed according to its weight and score to obtain a comprehensive score; the comprehensive score is evaluated to determine whether the Chinese medicinal materials meet the characteristics of genuine products; an evaluation threshold is set, and when the comprehensive score is higher than the evaluation threshold, it is judged as a genuine product; when it is lower than the evaluation threshold, it is judged as a suspected counterfeit or substandard product;

[0043] Based on the results of comprehensive analysis, the identification results of Chinese medicinal materials are determined; if they are judged to be authentic, a quality grade assessment is given; if they are judged to be suspected counterfeit or substandard products, verification and analysis are carried out; the identification results are presented in the form of a report, including the authenticity judgment of the Chinese medicinal materials, chemical composition, quality grade assessment, and the process and basis of comprehensive analysis; the report can provide a scientific basis for quality control, market supervision and clinical application of Chinese medicinal materials; through the above process, a variety of information can be comprehensively utilized to conduct comprehensive and accurate identification of Chinese medicinal materials to obtain reliable identification results.

[0044] The present invention provides a platform for identifying the authenticity of traditional Chinese medicines, which has the following beneficial effects:

[0045] This Chinese medicinal material authenticity identification platform has a multi-dimensional data collection function. Through the appearance collection unit, spectrum collection unit and information collection unit, it respectively obtains the appearance image, spectrum data and background information of the Chinese medicinal materials, comprehensively analyzes the Chinese medicinal materials from multiple angles, avoids the limitations of a single method, and improves the accuracy of identification; uses artificial intelligence algorithms, DNA barcode technology, chemical fingerprint spectrum similarity evaluation and near-infrared spectroscopy technology to deeply analyze the characteristics of Chinese medicinal materials and accurately judge the authenticity and quality; by establishing a quantitative analysis model, the content of specific components of Chinese medicinal materials is predicted through spectral data, and compared with standard spectral data, an accurate assessment of the quality of Chinese medicinal materials is achieved; the model is verified by cross-validation method to ensure accuracy and stability, improve identification accuracy, and solve the problem that traditional identification methods are difficult to meet the needs of quickly and accurately identifying the authenticity of Chinese medicinal materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 The present invention is a flowchart of a platform for identifying the authenticity of traditional Chinese medicine. DETAILED DESCRIPTION

[0047] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0048] like Figure 1 As shown, the present invention provides a technical solution: a Chinese medicinal material authenticity identification platform, comprising a data acquisition module, the data acquisition module is connected to a data analysis module, and the data analysis module is connected to a result display module;

[0049] The data acquisition module is used to collect appearance images, spectral data and background information of Chinese medicinal materials;

[0050] The data analysis module uses artificial intelligence algorithms to analyze and process the collected data to determine the authenticity of Chinese medicinal materials;

[0051] The result display module is used to display the identification results to the user, including the authenticity judgment results and similarity scores;

[0052] It needs to be further explained that, in the specific implementation process, a comprehensive Chinese medicinal material authenticity identification platform is provided, integrating data collection, analysis and result display functions, providing a one-stop solution for Chinese medicinal materials authenticity identification; using artificial intelligence algorithms to improve identification efficiency and accuracy and reduce human interference; and displaying detailed identification results to users, facilitating an intuitive understanding of the quality of Chinese medicinal materials.

[0053] The data acquisition module includes an appearance acquisition unit, a spectrum acquisition unit and an information collection unit;

[0054] The appearance acquisition unit selects and calibrates a high-resolution image acquisition device, selects representative Chinese medicinal material samples and places them on a preset background, adjusts the device angle, distance and lighting conditions for shooting, controls the light intensity and direction to avoid shadow reflections, stores the image in JPEG format, obtains the appearance image and transmits it to the data analysis module;

[0055] It should be further explained that, in the specific implementation process, high-resolution image acquisition equipment is selected to ensure that it can capture clear details and accurate colors; the image acquisition equipment is calibrated, including color calibration and focal length adjustment, to ensure the stability of the acquired image quality; representative Chinese medicinal material samples are selected to ensure the integrity and cleanliness of the samples to avoid the influence of impurities and damage on image acquisition; and the Chinese medicinal material samples are placed on a preset background, wherein the preset background color has a clear contrast with the Chinese medicinal materials so as to highlight the characteristics of the Chinese medicinal materials; the angle, distance and lighting conditions of the image acquisition equipment are adjusted to ensure that all parts of the Chinese medicinal materials can be fully and clearly photographed, including Shape, color, and texture features; use multi-angle shooting to obtain different perspective images of Chinese medicinal materials for subsequent analysis; use a fixed shooting angle sequence to ensure the consistency of image acquisition for different samples; control the intensity and direction of light to avoid the impact of shadows and reflections on image quality; use soft natural light or professional photography lights to ensure uniform lighting; store the collected high-resolution images in JPEG format to obtain appearance images and ensure that the image quality is not lost; and transmit the images to the data analysis module via wired or wireless means for subsequent processing; use data transmission software or network storage services to ensure the stability and security of image transmission;

[0056] It should be further explained that, in the specific implementation process, the specific scheme of using a fixed shooting angle sequence is as follows: according to the shape, size and complexity of the characteristics of the Chinese medicinal materials, the number of appropriate shooting angles is determined to be 8; for Chinese medicinal materials with complex shapes or multiple faces, the number of shooting angles is increased; representative angles and directions are selected to display the main characteristics of the Chinese medicinal materials, including the front, back, side, top and bottom surfaces; for Chinese medicinal materials with special textures, shapes or marks, specific angles and directions can be determined according to the characteristics, including tilt angles and close-up angles; thus, the number and directions of shooting angles are determined;

[0057] Determine a fixed shooting sequence to maintain consistency among different samples, shoot in a clockwise or counterclockwise order, or in a specific logical order, from the whole to the part, from the main features to the secondary features; and give priority to shooting key angles to improve identification efficiency;

[0058] Keep the shooting interval stable to ensure that the position and distance are relatively consistent. You can use a fixed shooting bracket or marker to help determine the position. Determine the appropriate interval based on the size of the Chinese medicinal materials and the performance of the shooting equipment, which is between 12cm. Thus, the shooting order and interval are determined.

[0059] Use image acquisition equipment with high resolution and good color reproduction capabilities, including professional digital cameras and industrial cameras; select the corresponding focal length and aperture according to the size of the Chinese medicinal materials and the shooting distance. For smaller Chinese medicinal materials, you can choose a macro lens or a high magnification lens; set fixed shooting parameters, including exposure time, sensitivity, and white balance, and maintain consistency to ensure consistent image quality and color performance; perform appropriate focus and composition to ensure that the Chinese medicinal materials are clearly visible and occupy an appropriate proportion. You can use automatic or manual focus adjustment; complete the setting of the shooting equipment and parameter adjustment;

[0060] Choose a uniform and stable lighting environment to avoid shadows and reflections affecting image quality. Use professional photography lights or natural light, and adjust the position and intensity of the lights to control the lighting effect. Keep the lighting direction and intensity consistent to avoid changes in lighting conditions at different angles. Use a light box or a soft light cover to achieve uniform lighting. Choose a simple, single background color to highlight the appearance characteristics of Chinese medicinal materials. White or black backgrounds are commonly used and should be selected according to the color and characteristics of the Chinese medicinal materials. Ensure the flatness and cleanliness of the background to avoid impurities and defects affecting image quality. Use a background cloth or background board to achieve a uniform background effect. This allows for shooting environment control.

[0061] Shoot in sequence according to the determined shooting angle sequence and order, keep the handheld steady or use a tripod to avoid shaking and blurring; take multiple photos at each angle and select the best image for subsequent analysis, and use continuous shooting mode or timed shooting function to improve efficiency; after shooting, conduct a quality check on the image at each angle, check the clarity, color reproduction, and exposure indicators to ensure that they meet the requirements; images that do not meet the requirements can be re-shot or post-processed, including adjusting brightness, contrast, color, etc. to improve quality; complete the shooting operation and quality inspection;

[0062] The spectrum acquisition unit uses spectrum analysis technology to collect spectrum data of Chinese medicinal materials to reflect their chemical composition information;

[0063] The information collection unit is used to collect background information of Chinese medicinal materials, including the origin and growth environment of the Chinese medicinal materials;

[0064] It needs to be further explained that, in the specific implementation process, the information collection unit collects the background information of Chinese medicinal materials as follows: consult the Chinese medicinal materials resource database and the quality standards and specification documents of Chinese medicinal materials in various places, search the academic literature database, and pay attention to academic conferences and seminars related to Chinese medicinal materials; communicate with growers, farmers' cooperatives, and enterprises in the origin of Chinese medicinal materials, invite experts and scholars to conduct on-site inspections, and obtain information collection channels; determine the specific location of the origin, understand the geographical characteristics and historical planting traditions, and obtain origin information; investigate the natural environment and human intervention, record the key nodes of the growth cycle and environmental changes, and obtain growth environment information; classify and organize the information collection channels, origin information, and growth environment information obtained, summarize them into the background information of Chinese medicinal materials, and establish a Chinese medicinal materials background information database, use tools to store, manage and encode and annotate; verify and validate the background information to ensure accuracy, and investigate and study controversial information; pay attention to industry trends and policy changes, update regularly to ensure timeliness, establish a maintenance mechanism to clarify responsibilities and processes, and back up and securely manage the database.

[0065] It needs to be further explained that, in the specific implementation process, the data acquisition module is subdivided into appearance acquisition, spectrum acquisition and information collection units to comprehensively obtain the appearance, chemical composition and background information of Chinese medicinal materials, providing rich data support for accurate identification; the appearance acquisition unit standardizes operation to ensure the acquisition of high-quality appearance images, which helps to accurately identify the appearance characteristics; the spectrum acquisition and information collection units respectively collect chemical composition and background information to provide multi-dimensional data for comprehensive analysis.

[0066] During the process of collecting spectral data of Chinese medicinal materials, the spectral acquisition unit selects a near-infrared spectrometer and calibrates and debugs it, selects representative Chinese medicinal material samples and processes them before placing them in the measurement area, selects a wavelength range as a measurement parameter, starts the equipment to measure and record spectral response data, and takes an average value after multiple measurements. After preprocessing the collected data, the chemical composition information of the Chinese medicinal materials is analyzed and extracted using chemometric methods. A quantitative analysis model is established to predict the content of specific components in the Chinese medicinal materials through spectral data, which is then compared with standard spectral data to determine the authenticity and quality, and the results are stored to generate a spectral analysis report, including a spectral graph, chemical composition information, and authenticity determination results.

[0067] It should be further explained that, in the specific implementation process, spectral analysis equipment suitable for Chinese medicinal materials analysis is selected, including near-infrared spectrometers or Raman spectrometers; the spectral equipment is calibrated and debugged to ensure its stable performance and accurate measurement; representative Chinese medicinal materials samples are selected and crushed, ground or sliced ​​to ensure the uniformity of spectral measurement; the processed Chinese medicinal materials samples are placed in the measurement area of ​​the spectral equipment to ensure good contact between the sample and the measurement probe or window; according to the characteristics of the Chinese medicinal materials and analysis requirements, appropriate spectral measurement parameters are selected, including wavelength range, resolution, and integration time; the spectral equipment is started for measurement, and the spectral response data of the Chinese medicinal materials samples at different wavelengths are recorded. The system can collect spectral data from the traditional Chinese medicine (TCM) and analyze the spectral data to extract the chemical composition information of TCM. The system can also use spectral data to predict the content of specific components in TCM by establishing a quantitative analysis model. The system can compare the spectral data with the known standard spectral data of TCM to determine the authenticity and quality of TCM. The spectral measurement results and analysis data can be stored for subsequent query and comparison. The system can generate a spectral analysis report, including the spectrum of TCM, chemical composition information, and authenticity judgment results, to provide a scientific basis for the identification of TCM.

[0068] The process of establishing a quantitative analysis model is as follows:

[0069] Establish a quantitative analysis model: collect a large number of spectral data of Chinese herbal medicine samples with known component contents. The Chinese herbal medicine samples cover Chinese herbal medicines from different origins, batches, and quality grades to ensure data diversity and representativeness; at the same time, determine the content of specific components in the Chinese herbal medicine samples through chemical analysis methods as the training target of the model; pre-process the collected spectral data to remove noise and perform baseline correction to improve data quality; use feature extraction algorithms to extract characteristic variables related to the content of specific components from the spectral data, including absorbance values ​​at specific wavelengths and shape characteristics of spectral curves; select the partial least squares PLS quantitative analysis model;

[0070] The model is trained and optimized using the extracted characteristic variables as input and the known content of specific components as output; the model parameters are adjusted so that the model can accurately predict the content of specific components in unknown samples; the established quantitative analysis model is verified using the cross-validation method; a portion of known samples are used as validation sets, predicted using the trained model, and compared with the actual component content to evaluate the accuracy and reliability of the model.

[0071] The process of comparing the spectral data of known Chinese herbal medicine standards to determine authenticity and quality is as follows:

[0072] Collect a large amount of standard spectrum data of Chinese herbal medicines with known authenticity and quality grades, and establish a standard spectrum database; the standard spectrum data comes from authoritative institutions, reliable experimental research or strictly verified samples; collect spectra of the Chinese herbal medicine samples to be tested to obtain their spectrum data; compare the spectrum data of the sample to be tested with the standard spectrum data in the standard spectrum database, use cosine similarity as the spectrum similarity calculation method, and calculate the similarity value between the sample to be tested and the standard spectrum; judge the authenticity of the Chinese herbal medicine to be tested based on the similarity value;

[0073] A similarity threshold is set. When the similarity value between the sample to be tested and the standard authentic medicinal material spectrum is higher than the threshold, it can be preliminarily judged as an authentic medicinal material; when the similarity value is lower than the threshold, it is judged as a fake medicinal material or a suspicious medicinal material; for samples judged to be authentic medicinal materials, their quality grades are further evaluated based on their similarity values ​​with standard spectra of different quality grades; samples with high similarity to high-quality standard spectra are considered to be high-quality medicinal materials, and samples with high similarity to low-quality standard spectra are considered to be low-quality medicinal materials; the judgment results are presented in the form of a report, including the authenticity judgment of the Chinese medicinal materials, the quality grade assessment and the information on the similarity value with the standard spectrum, providing a basis for the identification and quality control of Chinese medicinal materials.

[0074] It needs to be further explained that, in the specific implementation process, the spectral acquisition unit adopts advanced spectral analysis technology to accurately reflect the chemical composition information, providing a scientific basis for judging authenticity and quality; establishes a quantitative analysis model and compares it with standard spectral data to predict the content of specific components and judge the authenticity and quality, thereby improving the identification accuracy; generates a spectral analysis report to provide detailed information for quality control and research; in the process of establishing a quantitative analysis model, collects a large amount of sample data, preprocesses, extracts features and optimizes model training to improve model accuracy and reliability; uses a cross-validation method to verify the model to ensure generalization ability and stability, thereby providing a guarantee for accurate prediction of component content.

[0075] The data analysis module includes a comparison unit and an identification unit:

[0076] The comparison unit obtains image data based on the acquired appearance image after standard processing, and then uses artificial intelligence algorithm to analyze the collected image data, automatically identifies the appearance characteristics of Chinese medicinal materials, and introduces a similarity judgment mechanism to compare the appearance characteristics with the known characteristics of authentic samples to obtain the comparison results; wherein, the similarity judgment mechanism is composed of DNA barcode technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology NIR.

[0077] The identification unit analyzes the collected spectral data, determines the chemical composition of the Chinese medicinal materials, obtains the analysis results, and conducts a comprehensive analysis of the comparison results and the analysis results in combination with the background information to determine whether they meet the characteristics of the authentic products and obtain the identification results; by conducting a comprehensive analysis in combination with the background information, the accuracy of identification is improved.

[0078] The comparison unit obtains image data based on the acquired appearance image after standard processing as follows:

[0079] The median filtering method is used to denoise the acquired Chinese medicinal materials appearance images, remove noise points in the image, and improve the image quality; the histogram equalization method is used to enhance the image, enhance the contrast and clarity of the image, and highlight the characteristics of the Chinese medicinal materials; the image is uniformly adjusted to the set size and resolution, and the size and resolution of the image are adjusted to meet the requirements of subsequent analysis; the image is converted into the HSV color space, and the color value is normalized, the color of the image is standardized, and the color consistency between different images is ensured to obtain image data.

[0080] During the analysis of image data using artificial intelligence algorithms, convolutional neural network (CNN) is used to extract features from standardized image data; CNN can automatically learn the features of Chinese medicinal materials images, including shape, color, and texture; the extracted features are a combination of local and global features of the image to comprehensively describe the appearance characteristics of Chinese medicinal materials; the principal component analysis (PCA) method is used to reduce the dimension of the features, remove redundant information, analyze and process the extracted features, and use a clustering algorithm to classify the features, grouping Chinese medicinal materials with similar appearance features into one category to obtain the appearance characteristics of the Chinese medicinal materials.

[0081] Based on the similarity judgment mechanism, which is composed of DNA barcode technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology NIR, the process of introducing the similarity judgment mechanism to obtain the comparison results is as follows:

[0082] S1: Evaluation process through DNA barcoding technology: Collect samples of Chinese herbal medicines to be tested and samples of known authenticity, and extract DNA from the samples;

[0083] A DNA barcode region is selected as a target DNA region, and a short single-stranded nucleic acid sequence is designed according to the target DNA region for PCR amplification to obtain a DNA barcode sequence;

[0084] Sequence the amplified DNA barcode sequence, compare the sequence of the sample to be tested with the database of known authentic samples, calculate the sequence similarity using the basic local alignment search tool, and obtain the similarity value of the DNA barcode sequence;

[0085] S2: Chemical fingerprint similarity evaluation process:

[0086] Extract and purify the Chinese medicinal materials to be tested and the samples with known authenticity, and use chromatography technology to obtain chemical fingerprints, including high performance liquid chromatography fingerprints and gas chromatography-mass spectrometry fingerprints;

[0087] Extract features from chemical fingerprints, including peak area, retention time, and relative content;

[0088] The cosine similarity algorithm is used to calculate the similarity between the chemical fingerprint of the sample to be tested and the known true and false samples, and the similarity value of the chemical fingerprint is obtained;

[0089] S3: Near infrared spectroscopy (NIR) evaluation process:

[0090] Collect near-infrared spectra of the Chinese medicinal materials to be tested and the known authenticity samples, perform denoising and baseline correction on the collected near-infrared spectrum data, and extract the absorbance characteristics at specific wavelengths of the near-infrared spectrum;

[0091] A quantitative analysis model is established based on the partial least squares method. The near-infrared spectrum characteristics of the sample to be tested are input into the model and compared with known true and false samples to obtain the similarity judgment results of the near-infrared spectrum.

[0092] S4: The comparison results are obtained by integrating similarity evaluation mechanism:

[0093] According to the accuracy, reliability and applicability of the evaluation methods in steps S1, S2 and S3, corresponding weights are assigned to DNA barcoding technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology;

[0094] The similarity values ​​obtained by the three evaluation methods are multiplied by the corresponding weights, and the weighted values ​​are summed to obtain a comprehensive similarity score. A threshold is set according to the comprehensive similarity score to determine the authenticity of the Chinese medicinal materials to be tested;

[0095] If the comprehensive similarity score is higher than the threshold of the true sample, it is judged to be true; if it is lower than the threshold of the false sample, it is judged to be false; if it is in the middle range, it is suspicious and awaits further analysis; finally, a comparison result based on the similarity judgment mechanism is obtained, which provides an accurate basis for the authenticity identification of Chinese medicinal materials.

[0096] The process of obtaining the identification result is as follows:

[0097] The median filtering algorithm is used to denoise the collected spectral data, remove the noise caused by instrument errors and environmental interference, and improve the quality of the data; the polynomial fitting method is used to perform baseline correction to eliminate the baseline drift phenomenon in the spectral data and make the spectral curve more stable; the features that can reflect the chemical composition of Chinese medicinal materials are extracted from the pre-processed spectral data, including the extraction of absorbance values ​​at specific wavelengths, peak values ​​and valley values ​​of the spectral curve; the principal component analysis PCA in the chemometric method is used to reduce the dimension of the spectral data, extract the main characteristic variables, and reduce the complexity of the data; based on the extracted features, combined with the known spectral database and chemical knowledge, the extracted features are compared with the standard spectrum to identify the spectral features corresponding to the specific chemical components and determine the chemical composition of the Chinese medicinal materials;

[0098] The results obtained from the appearance image comparison, the chemical composition results obtained from the spectral data analysis, and the background information of the Chinese medicinal materials are integrated, among which the appearance image comparison results include similarity scores and authenticity judgments, and the background information includes the origin and growth environment; according to the importance and reliability of different information, corresponding weights are assigned to the appearance image comparison results, spectral data analysis results, and background information; among them, the appearance image comparison results have a higher weight in judging the variety and authenticity of Chinese medicinal materials, while the spectral data analysis results are more important in determining the chemical composition and quality, and the background information can be used as a basis for auxiliary judgment; for each judgment indicator, a weighted calculation is performed according to its weight and score to obtain a comprehensive score; the comprehensive score is evaluated to determine whether the Chinese medicinal materials meet the characteristics of genuine products; an evaluation threshold is set, and when the comprehensive score is higher than the evaluation threshold, it is judged as a genuine product; when it is lower than the evaluation threshold, it is judged as a suspected counterfeit or substandard product;

[0099] Based on the results of comprehensive analysis, the identification results of Chinese medicinal materials are determined; if they are judged to be authentic, a quality grade assessment is given; if they are judged to be suspected counterfeit or substandard products, verification and analysis are carried out; the identification results are presented in the form of a report, including the authenticity judgment of the Chinese medicinal materials, chemical composition, quality grade assessment, and the process and basis of comprehensive analysis; the report can provide a scientific basis for quality control, market supervision and clinical application of Chinese medicinal materials; through the above process, a variety of information can be comprehensively utilized to conduct comprehensive and accurate identification of Chinese medicinal materials to obtain reliable identification results.

[0100] It needs to be further explained that, in the specific implementation process, the comparison process with the standard spectral data of known Chinese medicinal materials can quickly and accurately determine the authenticity and quality grade, providing strong support for market supervision and clinical applications; the cosine similarity calculation method is adopted to make the similarity value calculation more objective and accurate, and improve the credibility of the identification results; the judgment results are presented in the form of a report to facilitate understanding of the quality status; and the data analysis module comparison unit and identification unit have clear division of labor, respectively analyzing the appearance characteristics and chemical composition to improve the comprehensiveness and accuracy of the identification results; the introduction of a similarity judgment mechanism combines multiple technical advantages to improve the identification accuracy.

[0101] It should be further explained that, in the specific implementation process, similarity evaluation mechanisms are applied in combination, giving full play to the advantages of multiple technologies, identifying from different angles, and improving the reliability of identification results; weights are allocated and calculated comprehensively to make the identification results more objective and accurate, and reduce the limitations of a single technology;

[0102] The spectral data analysis process is scientific and reasonable, accurately determining the chemical composition and providing an important basis for comprehensive analysis. It integrates the appearance image comparison results, spectral data analysis results and background information for comprehensive analysis, comprehensively considers all aspects of Chinese medicinal materials, and improves the accuracy and reliability of identification results. It presents the identification results in the form of a report, providing detailed information and basis for quality control, market supervision and clinical application.

[0103] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A Chinese herbal medicine authenticity identification platform, characterized in that: It includes a data acquisition module, the data acquisition module is connected to a data analysis module, and the data analysis module is connected to a result display module; The data acquisition module is used to collect appearance images, spectral data and background information of Chinese medicinal materials; The data analysis module uses artificial intelligence algorithms to analyze and process the collected data to determine the authenticity of the Chinese medicinal materials; The result display module is used to display the identification results to the user, including the authenticity judgment results and the similarity score; The data analysis module includes a comparison unit and an identification unit: The comparison unit obtains image data based on the acquired appearance image after standard processing, and then uses an artificial intelligence algorithm to analyze the collected image data, automatically identifies the appearance characteristics of the Chinese medicinal materials, and introduces a similarity judgment mechanism to obtain a comparison result; wherein the similarity judgment mechanism is composed of DNA barcode technology, chemical fingerprint similarity evaluation and near-infrared spectroscopy technology. The identification unit analyzes the collected spectral data to determine the chemical composition of the Chinese medicinal materials, obtains the analysis results, and conducts a comprehensive analysis of the comparison results and the analysis results in combination with background information to determine whether they meet the characteristics of the authentic products and obtain the identification results.

2. A Chinese medicinal material authenticity identification platform according to claim 1, characterized in that: The data acquisition module includes an appearance acquisition unit, a spectrum acquisition unit and an information collection unit; The appearance acquisition unit selects a high-resolution image acquisition device and calibrates it, selects a representative Chinese medicinal material sample and places it on a preset background, adjusts the device angle, distance and lighting conditions for shooting, controls the lighting intensity and direction to avoid shadow reflections, stores the image in JPEG format, obtains the appearance image and transmits it to the data analysis module; The spectrum acquisition unit uses spectrum analysis technology to collect spectrum data of Chinese medicinal materials to reflect their chemical composition information; The information collection unit is used to collect background information of Chinese medicinal materials, and the background information includes the origin and growth environment of the Chinese medicinal materials.

3. A Chinese medicinal material authenticity identification platform according to claim 2, characterized in that: During the process of collecting spectral data of Chinese medicinal materials, the spectral acquisition unit selects a near-infrared spectrometer and calibrates and debugs it, selects representative Chinese medicinal material samples and places them in the measurement area after processing, selects a wavelength range as a measurement parameter, starts the equipment to measure and record spectral response data, and takes an average value after multiple measurements. After preprocessing the collected data, a chemometric method is used to analyze and extract the chemical composition information of the Chinese medicinal materials. By establishing a quantitative analysis model, the content of specific components in the Chinese medicinal materials is predicted through spectral data, which is compared with standard spectral data to judge the authenticity and quality, and the results are stored to generate a spectral analysis report, including a spectral graph, chemical composition information and authenticity judgment results.

4. A Chinese medicinal material authenticity identification platform according to claim 3, characterized in that: The process of establishing the quantitative analysis model is as follows: Establish a quantitative analysis model: collect a large number of spectral data of Chinese herbal medicine samples with known component content, and at the same time, determine the content of specific components in the Chinese herbal medicine samples through chemical analysis methods as the training target of the model; pre-process the collected spectral data, including removing noise and baseline correction, using feature extraction algorithms to extract characteristic variables related to the content of specific components from the spectral data, and select the partial least squares PLS quantitative analysis model; The model is trained and optimized using the extracted characteristic variables as input and the known specific component content as output; the established quantitative analysis model is verified using the cross-validation method; a portion of known samples is used as a validation set, predicted using the trained model, and compared with the actual component content.

5. A Chinese medicinal material authenticity identification platform according to claim 4, characterized in that: The process of comparing the spectral data of known Chinese herbal medicine standards to determine authenticity and quality is as follows: Collect a large amount of standard spectrum data of Chinese herbal medicines with known authenticity and quality grades, and establish a standard spectrum database; collect spectra of the Chinese herbal medicine samples to be tested to obtain their spectrum data; compare the spectrum data of the samples to be tested with the standard spectrum data in the standard spectrum database, and use cosine similarity as a spectrum similarity calculation method to calculate the similarity value between the samples to be tested and the standard spectrum; Determine the authenticity of the Chinese medicinal materials to be tested based on the similarity value; A similarity threshold is set. When the similarity value between the sample to be tested and the standard authentic medicinal material spectrum is higher than the threshold, it can be preliminarily judged as an authentic medicinal material; when the similarity value is lower than the threshold, it is judged as a fake medicinal material or a suspicious medicinal material; for samples judged to be authentic medicinal materials, their quality grades are further evaluated based on their similarity values ​​with standard spectra of different quality grades; samples with high similarity to high-quality standard spectra are considered to be high-quality medicinal materials, and samples with high similarity to low-quality standard spectra are considered to be low-quality medicinal materials; the judgment results are presented in the form of a report, including information on the authenticity judgment of the Chinese medicinal materials, the quality grade evaluation, and the similarity value with the standard spectrum.

6. A Chinese medicinal material authenticity identification platform according to claim 5, characterized in that: The comparison unit obtains image data based on the acquired appearance image through standard processing as follows: The median filtering method is used to denoise the acquired Chinese medicinal materials appearance images and remove noise points in the image; the histogram equalization method is used to enhance the image and enhance the contrast and clarity of the image; the image is uniformly adjusted to the set size and resolution; the image is converted into the HSV color space, and the color value is normalized to obtain image data.

7. A Chinese medicinal material authenticity identification platform according to claim 6, characterized in that: During the analysis of image data using artificial intelligence algorithms, convolutional neural network CNN is used to extract features from standardized image data; the extracted features are a combination of local features and global features of the image; the principal component analysis PCA method is used to reduce the dimension of the features and remove redundant information, the extracted features are analyzed and processed, and a clustering algorithm is used to classify the features, and Chinese medicinal materials with similar appearance features are grouped together to obtain the appearance characteristics of the Chinese medicinal materials.

8. A Chinese medicinal material authenticity identification platform according to claim 7, characterized in that: The process of introducing a similarity judgment mechanism to obtain comparison results is as follows: S1: Evaluation process through DNA barcoding technology: Collect samples of Chinese herbal medicines to be tested and samples of known authenticity, and extract DNA from the samples; A DNA barcode region is selected as a target DNA region, and a short single-stranded nucleic acid sequence is designed according to the target DNA region for PCR amplification to obtain a DNA barcode sequence; Sequence the amplified DNA barcode sequence, compare the sequence of the sample to be tested with the database of known authentic samples, calculate the sequence similarity using the basic local alignment search tool, and obtain the similarity value of the DNA barcode sequence; S2: Chemical fingerprint similarity evaluation process: Extract and purify the Chinese medicinal materials to be tested and the samples with known authenticity, and use chromatography technology to obtain chemical fingerprints, including high performance liquid chromatography fingerprints and gas chromatography-mass spectrometry fingerprints; Extract features from chemical fingerprints, including peak area, retention time, and relative content; The cosine similarity algorithm is used to calculate the similarity between the chemical fingerprint of the sample to be tested and the known true and false samples, and the similarity value of the chemical fingerprint is obtained; S3: Near infrared spectroscopy technology evaluation process: Collect near-infrared spectra of the Chinese medicinal materials to be tested and the known authenticity samples, perform denoising and baseline correction on the collected near-infrared spectrum data, and extract the absorbance characteristics at specific wavelengths of the near-infrared spectrum; A quantitative analysis model is established based on the partial least squares method. The near-infrared spectrum characteristics of the sample to be tested are input into the model and compared with known true and false samples to obtain the similarity judgment results of the near-infrared spectrum. S4: The comparison results are obtained by integrating similarity evaluation mechanism: According to the evaluation methods in steps S1, S2 and S3, corresponding weights are assigned to DNA barcoding technology, chemical fingerprint similarity evaluation and near infrared spectroscopy technology; The similarity values ​​obtained by the three evaluation methods are multiplied by the corresponding weights, and the weighted values ​​are summed to obtain a comprehensive similarity score. A threshold is set according to the comprehensive similarity score to determine the authenticity of the Chinese medicinal materials to be tested; If the comprehensive similarity score is higher than the threshold of the true sample, it is judged to be true; if it is lower than the threshold of the false sample, it is judged to be false; if it is in the middle range, it is suspicious and awaits further analysis; finally, a comparison result based on the similarity judgment mechanism is obtained.

9. A Chinese medicinal material authenticity identification platform according to claim 8, characterized in that: The process of obtaining the identification result is as follows: The median filtering algorithm is used to denoise the collected spectral data; the polynomial fitting method is used for baseline correction; the features that can reflect the chemical composition of Chinese medicinal materials are extracted from the pre-processed spectral data, including the absorbance value at a specific wavelength, the peak and valley of the spectral curve; the principal component analysis PCA in the chemometric method is used to reduce the dimension of the spectral data and extract the main characteristic variables; based on the extracted features, combined with the known spectral database and chemical knowledge, the extracted features are compared with the standard spectrum to identify the spectral features corresponding to the specific chemical components and determine the chemical composition of the Chinese medicinal materials; The results obtained from the appearance image comparison, the chemical composition results obtained from the spectral data analysis, and the background information of the Chinese herbal medicine are integrated, wherein the appearance image comparison results include similarity scores and authenticity judgments, and the background information includes the origin and growth environment; according to the importance and reliability of different information, corresponding weights are assigned to the appearance image comparison results, spectral data analysis results, and background information; for each judgment indicator, a weighted calculation is performed according to its weight and score to obtain a comprehensive score; the comprehensive score is evaluated to determine whether the Chinese herbal medicine meets the characteristics of authentic products; an evaluation threshold is set, and when the comprehensive score is higher than the evaluation threshold, it is judged to be authentic; when it is lower than the evaluation threshold, it is judged to be a suspected counterfeit or substandard product; Based on the results of comprehensive analysis, the identification results of the Chinese medicinal materials are determined; if it is judged to be authentic, a quality grade assessment is given; if it is judged to be a suspected counterfeit or substandard product, verification and analysis are carried out; the identification results are presented in the form of a report, including the authenticity judgment of the Chinese medicinal materials, chemical composition, quality grade assessment, and the process and basis of the comprehensive analysis.

Images