Traditional Chinese medicinal material production place identification method and system and storage medium

Abstract

The invention relates to a traditional Chinese medicinal material producing area identification method and system and a storage medium. Chromatograms, measured through an internal standard method, of traditional Chinese medicinal materials of different producing areas are obtained, the chromatograms are normalized to obtain a chromatogram set, all specific peaks in the chromatograms of the traditional Chinese medicinal materials of different producing areas are obtained, and specific peak areas serve as target areas; calculating the numerical range of each feature of the chromatographic peak of the internal standard substance in the chromatogram set, and constructing a plurality of feature sets, the features in the feature sets being from the features of the chromatographic peak of the internal standard substance, and each feature in the feature sets corresponding to one numerical range; enhancing a part of target areas of a to-be-enhanced chromatogram based on the feature set, and obtaining a training set by utilizing the part of target areas and the chromatogram set; and training the neural network model by using the training set, and obtaining the production place of the traditional Chinese medicinal material by using the chromatogram of the to-be-identified traditional Chinese medicinal material and the trained neural network model. The method can improve the identification accuracy of the production places of the traditional Chinese medicinal materials under the condition of few samples.

Description

Technical Field

[0001] The present invention relates to the field of traditional Chinese medicine, and in particular to a method, system and storage medium for identifying the origin of traditional Chinese medicine. Background Art

[0002] Due to the differences in geographical environment, climate conditions, soil composition and growth mode, Chinese medicinal materials from different origins often have significant differences in chemical composition. In recent years, with the development of analytical technology, it has become an effective means of identification to detect the chemical composition of Chinese medicinal materials and establish a correlation model with the characteristics of the origin. Among them, chromatographic analysis is a widely used method. The chromatogram can reflect the specific information of various compounds in Chinese medicinal materials, and has become an important way to analyze the differences in the components of Chinese medicinal materials and study their origin characteristics. Establishing a chromatographic database of the origin of Chinese medicinal materials can summarize the characteristic differences in the chemical composition of Chinese medicinal materials from different origins, and serve as an important basis for origin identification. The neural network model has strong nonlinear fitting ability and pattern recognition ability, and can extract characteristic information related to the origin from complex chromatographic data. After model training, the origin of unknown samples can be accurately determined; compared with traditional multidimensional data analysis methods, the application of neural network models can significantly improve the accuracy and efficiency of origin identification. However, in order to achieve a high accuracy of the neural network model, a large number of samples are required, and the process of obtaining the chromatogram of Chinese medicinal materials is cumbersome, including a series of preliminary preparations. This leads to fewer training samples and poor accuracy of the trained neural network model. How to ensure high accuracy with fewer samples is an urgent problem to be solved in this field. Summary of the invention

[0003] The embodiments of the present invention provide a method, system and computer-readable storage medium for identifying the origin of Chinese medicinal materials, so as to at least solve the technical problem of small number of samples and low accuracy in identifying the origin of Chinese medicinal materials.

[0004] According to one aspect of an embodiment of the present invention, a method for identifying the origin of a Chinese medicinal material is provided, comprising: Acquire chromatograms of Chinese medicinal materials from different origins measured by an internal standard method, normalize the chromatograms to obtain a chromatogram set, acquire all unique peaks in the chromatograms of Chinese medicinal materials from different origins, and use the unique peak region as a target region; Calculating the numerical range of each feature of the chromatographic peak of the internal standard in the chromatogram set, and constructing multiple feature sets, wherein the features in the feature sets are derived from the features of the chromatographic peak of the internal standard, and each feature in the feature set corresponds to one of the numerical ranges; Based on the feature set, a partial target region of the chromatogram to be enhanced is enhanced, and a training set is obtained using the partial target region and the chromatogram set; The training set is used to train the neural network model, and the chromatogram of the Chinese medicinal materials to be identified and the trained neural network model are used to obtain the origin of the Chinese medicinal materials.

[0005] According to another aspect of this embodiment, the characteristics of the chromatographic peak include retention time, peak height, peak area, peak width, peak skewness, peak kurtosis, and peak standard deviation.

[0006] According to another aspect of the embodiment of the present invention, the enhancing of the partial target area of ​​the chromatogram to be enhanced based on the feature set is specifically: For each partial target area of ​​the chromatogram to be enhanced, randomly select a feature set from the multiple feature sets, and randomly generate an instance numerical sequence of the selected feature set, wherein the range of numerical values ​​in the instance numerical sequence is within the numerical range corresponding to the feature; A numerical sequence corresponding to the selected feature set in the chromatographic peak of the internal standard in the chromatogram to be enhanced is obtained, a change in the example numerical sequence relative to the numerical sequence is calculated, and a partial target area of ​​the chromatogram to be enhanced is enhanced based on the change to obtain an enhanced target area of ​​the chromatogram.

[0007] According to another aspect of this embodiment, the calculation example numerical sequence changes relative to the numerical sequence, and based on the changes, a part of the target area of ​​the chromatogram to be enhanced is enhanced to obtain the enhanced target area of ​​the chromatogram, specifically: For each feature in the selected feature set, calculate the relative change rate of the value corresponding to the feature in the instance numerical sequence relative to the value corresponding to the feature in the numerical sequence; Obtain the characteristic value of the characteristic set of each target area in the partial target area in the chromatogram to be enhanced, and obtain the relative change rate corresponding to the characteristic value; obtain a new characteristic value based on the characteristic value and the relative change rate corresponding to the characteristic value; adjust the target area using the new characteristic value so that the characteristic value corresponding to the characteristic of the target area is the new characteristic value.

[0008] According to another aspect of this embodiment, the use of the partial target region and the chromatogram set to obtain a training set is specifically: Obtain the target area of ​​the chromatogram in the chromatogram set, sample the target area to obtain the feature vector of the target area, splice the feature vectors of all the target areas in the chromatogram into a feature matrix by row, use the feature matrix as the sample in the training set, and annotate the sample as the origin of the Chinese medicinal material corresponding to the chromatogram; The enhanced partial target area is sampled to obtain a feature vector of the enhanced target area, and the feature vector corresponding to the partial target area in the feature matrix of the chromatogram where the partial target area is located is replaced with the feature vector of the enhanced target area to obtain a feature matrix of the enhanced chromatogram, and the feature matrix of the enhanced chromatogram is used as a sample in a training set, and the sample is labeled as the origin of the Chinese medicinal material to which the chromatogram where the partial target area is located belongs.

[0009] According to another aspect of an embodiment of the present invention, a system for identifying the origin of Chinese medicinal materials is provided, comprising: A target region acquisition module is used to obtain chromatograms of Chinese medicinal materials from different origins measured by an internal standard method, normalize the chromatograms to obtain a chromatogram set, obtain all unique peaks in the chromatograms of Chinese medicinal materials from different origins, and use the unique peak region as a target region; A feature set construction module, used to calculate the numerical range of each feature of the chromatographic peak of the internal standard in the chromatogram set, and construct multiple feature sets, wherein the features in the feature sets are derived from the features of the chromatographic peak of the internal standard, and each feature in the feature set corresponds to one of the numerical ranges; A sample enhancement module, used for enhancing a partial target area of ​​the chromatogram to be enhanced based on the feature set, and obtaining a training set using the partial target area and the chromatogram set; The origin identification module is used to train the neural network model using the training set, and obtain the origin of the Chinese medicinal materials using the chromatogram of the Chinese medicinal materials to be identified and the trained neural network model.

[0010] According to another aspect of this embodiment, the characteristics of the chromatographic peak include retention time, peak height, peak area, peak width, peak skewness, peak kurtosis, and peak standard deviation.

[0011] According to another aspect of this embodiment, the enhancing of a partial target area of ​​the chromatogram to be enhanced based on the feature set is specifically: For each partial target area of ​​the chromatogram to be enhanced, randomly select a feature set from the multiple feature sets, and randomly generate an instance numerical sequence of the selected feature set, wherein the range of numerical values ​​in the instance numerical sequence is within the numerical range corresponding to the feature; A numerical sequence corresponding to the selected feature set in the chromatographic peak of the internal standard in the chromatogram to be enhanced is obtained, a change in the example numerical sequence relative to the numerical sequence is calculated, and a partial target area of ​​the chromatogram to be enhanced is enhanced based on the change to obtain an enhanced target area of ​​the chromatogram.

[0012] According to another aspect of this embodiment, the calculation example numerical sequence changes relative to the numerical sequence, and based on the changes, a part of the target area of ​​the chromatogram to be enhanced is enhanced to obtain the enhanced target area of ​​the chromatogram, specifically: For each feature in the selected feature set, calculate the relative change rate of the value corresponding to the feature in the instance numerical sequence relative to the value corresponding to the feature in the numerical sequence; Obtain the characteristic value of the characteristic set of each target area in the partial target area in the chromatogram to be enhanced, and obtain the relative change rate corresponding to the characteristic value; obtain a new characteristic value based on the characteristic value and the relative change rate corresponding to the characteristic value; adjust the target area using the new characteristic value so that the characteristic value corresponding to the characteristic of the target area is the new characteristic value.

[0013] According to another aspect of this embodiment, the use of the partial target region and the chromatogram set to obtain a training set is specifically: Obtain the target area of ​​the chromatogram in the chromatogram set, sample the target area to obtain the feature vector of the target area, splice the feature vectors of all the target areas in the chromatogram into a feature matrix by row, use the feature matrix as the sample in the training set, and annotate the sample as the origin of the Chinese medicinal material corresponding to the chromatogram; The enhanced partial target area is sampled to obtain a feature vector of the enhanced target area, and the feature vector corresponding to the partial target area in the feature matrix of the chromatogram where the partial target area is located is replaced with the feature vector of the enhanced target area to obtain a feature matrix of the enhanced chromatogram, and the feature matrix of the enhanced chromatogram is used as a sample in a training set, and the sample is labeled as the origin of the Chinese medicinal material to which the chromatogram where the partial target area is located belongs.

[0014] According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0015] The present invention uses the change of the chromatographic peak of the internal standard to enhance the region of the unique peak in the chromatogram of the Chinese herbal medicine, so that the generated enhanced sample can maintain the basic structure and characteristics of the original data and avoid excessive deviation from the actual situation. In addition, the present invention constructs a feature matrix by extracting the information of the unique peak, avoids the influence of large calculation amount and matrix effect on the result, and improves the accuracy of origin identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of Embodiment 1; Figure 2 It is a schematic diagram of the unique peaks of Chinese medicinal materials from a certain place of origin; Figure 3A schematic diagram of a target area in a chromatogram of a Chinese herbal medicine from a certain origin; Figure 4 Schematic diagram for constructing the enhanced sample. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0018] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] Example 1 This embodiment provides a method for identifying the origin of Chinese medicinal materials. Figure 1 As shown, including: S1, obtaining chromatograms of Chinese medicinal materials from different origins measured by an internal standard method, normalizing the chromatograms to obtain a chromatogram set, obtaining all unique peaks in the chromatograms of Chinese medicinal materials from different origins, and taking the unique peak region as the target region; The samples of Chinese medicinal materials from different origins are processed to obtain sample solutions, internal standards are added to the sample solutions, and then the samples are analyzed using high performance liquid chromatography. The peak height of each chromatogram is normalized to the range of 0 to 1, so that the peaks of all samples are consistent in magnitude. After normalization, the chromatograms of ginseng from different origins constitute a chromatogram set; the chromatograms in the chromatogram set include chromatograms of Chinese medicinal materials from multiple origins, and there are multiple chromatograms of Chinese medicinal materials from each origin. In the chromatogram, some peaks only appear in the chromatograms of Chinese medicinal materials from one or several origins, or have unique intensities and positions in the chromatograms of Chinese medicinal materials from one or several origins. These unique peaks usually reflect the differences in the chemical composition of Chinese medicinal materials from different origins. Figure 2The characteristic peaks of Chinese medicinal materials from one origin are shown. The area of ​​the characteristic peaks of all origins constitutes the target area. These characteristic peaks correspond to the unique components of Chinese medicinal materials from different origins, and are important marks for identifying different origins. After determining the unique peaks in Chinese medicinal materials from different origins, the area where these peaks are located can be defined as the target area. In one embodiment, the target area is obtained by determining the starting point and end point of the characteristic peak, and taking the area between the starting point and end point of the characteristic peak as the target area. Figure 3 A schematic diagram showing a target region in a chromatogram of a Chinese medicinal material from one origin.

[0020] Among them, obtaining all the unique peaks in the chromatograms of Chinese medicinal materials from different origins means obtaining the unique peaks of Chinese medicinal materials from all origins. For example, the unique peaks of origin 1 are the chromatographic peaks corresponding to retention times 1, 9, and 10, and the unique peaks of origin 2 are the chromatographic peaks corresponding to retention times 4 and 8. Then all the unique peaks include the chromatographic peaks corresponding to retention times 1, 4, 8, 9, and 10.

[0021] S2, calculating the numerical range of each feature of the chromatographic peak of the internal standard in the chromatogram set, and constructing multiple feature sets, wherein the features in the feature sets are derived from the features of the chromatographic peak of the internal standard, and each feature in the feature set corresponds to one of the numerical ranges; Internal standards of the same concentration should be the same on different chromatograms, but are affected by noise, matrix, injection speed, pH value, temperature, etc., even internal standards of the same concentration are different on different chromatograms. The chromatograms of multiple samples of Chinese medicinal materials from different origins are included in the chromatogram collection, and each chromatogram contains the chromatographic peak of the internal standard, and the chromatographic peak of the internal standard is also different. In one embodiment, the characteristics of the chromatographic peak include retention time, peak height, peak area, peak width, skewness of the peak, peak kurtosis, and standard deviation of the peak. Of course, only a part of them can be used as the characteristics of the chromatographic peak. Among them, the retention time is the time point when the peak value of the chromatographic peak appears, the peak height is the maximum value of the chromatographic peak, the peak area is the area below the chromatographic peak, the peak width is the peak width of the chromatographic peak, that is, the width of the peak starting point and the peak end point, the skewness of the peak is the symmetry of the chromatographic peak, and the skewness is 0 when it means symmetry, and negative or positive values ​​respectively indicate left or right deviation, the kurtosis of the peak is the sharpness of the peak, and the standard deviation of the peak is the standard deviation of the data in the chromatographic peak.

[0022] Extract the chromatographic peaks of the internal standard from all chromatograms in the sample set, calculate the minimum and maximum values ​​of each chromatographic peak feature in different samples, and obtain the numerical range of the feature. The features in each feature set, such as peak height, retention time, etc., are extracted from the chromatographic peak features of the internal standard, and each feature has a corresponding numerical range, that is, the minimum and maximum values ​​of the internal standard features from different samples. For example, the retention time may be between 10.1 and 10.3 minutes, so this range becomes the numerical range of the retention time feature. Construct multiple feature sets, the number of features included in each feature set is not exactly the same, and the specific features are not exactly the same. For example, <retention time, peak height> constitutes a feature set, and <peak height, peak area, peak width, peak skewness> constitutes a feature set.

[0023] S3, enhancing a partial target region of the chromatogram to be enhanced based on the feature set, and obtaining a training set using the partial target region and the chromatogram set; A pre-built feature set is used to perform data enhancement on a partial target area in a chromatogram to be enhanced to expand a data set. A chromatogram to be enhanced may include multiple target areas, and a portion of them is enhanced. In an optional embodiment, specifically, for each partial target area of ​​the chromatogram to be enhanced, a feature set is randomly selected from the multiple feature sets, and an instance numerical sequence of the selected feature set is randomly generated, and the range of the values ​​in the instance numerical sequence is within the numerical range corresponding to the feature; each feature in the feature set corresponds to a value in the instance numerical sequence, and the number of elements in the feature set and the instance numerical sequence are the same.

[0024] A numerical sequence corresponding to the selected feature set in the chromatographic peak of the internal standard in the chromatogram to be enhanced is obtained, a change in the example numerical sequence relative to the numerical sequence is calculated, and a partial target area of ​​the chromatogram to be enhanced is enhanced based on the change to obtain an enhanced target area of ​​the chromatogram.

[0025] In each chromatogram to be enhanced, the features of the target area are selectively and randomly enhanced. Specifically, a feature set is randomly selected from the multiple feature sets, and a random instance value sequence is generated based on the value range of the set. This sequence contains the value range that each feature (such as retention time, peak height, etc.) needs to reach when enhanced.

[0026] Among them, for each partial target area of ​​the chromatogram to be enhanced, a feature set is randomly selected from the multiple feature sets. One way is to randomly select a feature set from the multiple feature sets for each partial target area of ​​each chromatogram to be enhanced, that is, one target area corresponds to one feature set; another way is to randomly select a feature set from the multiple feature sets for all the partial target areas of each chromatogram to be enhanced, that is, all the partial target areas in a chromatogram to be enhanced use the same randomly selected feature set.

[0027] A feature set is randomly selected from the multiple feature sets, and a random instance value is generated based on the value range of each feature in the set. For example, the randomly selected feature set is <retention time, peak height>, and the instance value sequence is <3.5, 0.8>. The feature value sequence of the internal standard peak is extracted from the chromatogram to be enhanced, and it is compared with the randomly generated instance value sequence to calculate the relative change of the internal standard feature value and the instance value. For example, the value sequence of the feature set <retention time, peak height> of the chromatographic peak of the internal standard extracted from the chromatogram to be enhanced is <3.3, 0.9>. Further assume that the chromatogram to be enhanced Figure 1 The partial target regions are target regions 1 and 4, and the target regions 1 and 4 are enhanced so that the relative changes of the target regions 1 and 4 after enhancement relative to the target regions 1 and 4 before enhancement are the same as the relative changes of the calculated internal standard characteristic values ​​and the instance values.

[0028] In an optional embodiment, a relative change rate is used as the relative change situation, and the change situation of the numerical sequence of the example is calculated relative to the numerical sequence. Based on the change situation, a part of the target area of ​​the chromatogram to be enhanced is enhanced to obtain the enhanced target area of ​​the chromatogram, specifically: For each feature in the selected feature set, calculate the relative change rate of the value corresponding to the feature in the instance numerical sequence relative to the value corresponding to the feature in the numerical sequence; Obtain the characteristic value of the characteristic set of each target area in the partial target area in the chromatogram to be enhanced, and obtain the relative change rate corresponding to the characteristic value; obtain a new characteristic value based on the characteristic value and the relative change rate corresponding to the characteristic value; adjust the target area using the new characteristic value so that the characteristic value corresponding to the characteristic of the target area is the new characteristic value.

[0029] Assume that the characteristic retention time of the chromatographic peak of the internal standard of the chromatogram to be enhanced is 10.0 minutes and the peak height is 0.6. The retention time in the example numerical sequence is 10.5 minutes and the peak height is 0.65. The relative change rate of the retention time is 5%, and the relative change rate of the peak height is 7.7%. The target area 1 is adjusted so that the relative change rate of the retention time of the target area 1 after adjustment is 5% and the relative change rate of the peak height is 7.7% relative to the target area 1 before adjustment. For example, the retention time of the target area 1 before adjustment is 15, and the retention time after adjustment is 10.75.

[0030] After enhancing the target region, the training set is obtained using the partial target region and the chromatogram set. In an optional embodiment, the target region of the chromatogram in the chromatogram set is obtained, the target region is sampled to obtain a feature vector of the target region, the feature vectors of all the target regions in the chromatogram are spliced ​​row by row into a feature matrix, the feature matrix is ​​used as a sample in the training set, and the sample is labeled as the origin of the Chinese medicinal material corresponding to the chromatogram; The enhanced partial target area is sampled to obtain a feature vector of the enhanced target area, and the feature vector corresponding to the partial target area in the feature matrix of the chromatogram where the partial target area is located is replaced with the feature vector of the enhanced target area to obtain a feature matrix of the enhanced chromatogram, and the feature matrix of the enhanced chromatogram is used as a sample in a training set, and the sample is labeled as the origin of the Chinese medicinal material to which the chromatogram where the partial target area is located belongs.

[0031] A unique peak region is extracted from each chromatogram, and characteristic data is extracted from the region. Specifically, multiple data are collected at equal intervals in each target region. For example, a target region has 100 data, and 50 data are extracted at equal intervals as the characteristic vector of the target region. If there are 10 unique peaks determined in step S1, 10 characteristic vectors will be obtained for each chromatogram. These 10 characteristic vectors are spliced ​​into a characteristic matrix by row as samples in the training set, that is, the samples in the training set are the characteristic matrices obtained from the chromatograms, and the labels of the samples are the origins of the Chinese medicinal materials corresponding to the chromatograms.

[0032] Repeat the feature extraction process for the enhanced target area to generate the enhanced target area feature vector, which represents the feature information of the enhanced area. Replace the corresponding target area feature vector in the original chromatogram feature matrix with the enhanced feature vector to generate the enhanced feature matrix, such as Figure 4 As shown in Figure 2, the enhanced feature matrix is ​​added to the training set as a new training sample and annotated with the origin information of the original chromatogram. The training set includes two parts, one from the original collected chromatogram and the other from the enhanced target area chromatogram.

[0033] S4, using the training set to train the neural network model, and using the chromatogram of the Chinese medicinal materials to be identified and the trained neural network model to obtain the origin of the Chinese medicinal materials.

[0034] The neural network model is trained using the previously constructed training set (including the original and enhanced chromatogram feature matrices). In one embodiment, the neural network model preferably uses the Bert model or the BeiT model or the ViT model. In the neural network model, each row of the feature matrix is ​​used as a token. The feature matrix of each sample in the training set contains the target area features of the chromatogram and marks the corresponding origin of the Chinese herbal medicine. During the training process, the neural network model gradually optimizes the model parameters by learning the relationship between the features and origins of these samples, so that the model can recognize the characteristic patterns of chromatograms from different origins. After the model training is completed, the chromatogram of the Chinese herbal medicine to be identified is normalized and feature extracted to generate a feature matrix of the chromatogram. This feature matrix has the same format as the training sample and contains the target area feature information of the chromatogram of the Chinese herbal medicine to be identified. The feature matrix of the Chinese herbal medicine to be identified is input into the trained neural network model, and the model predicts the origin of the input sample.

[0035] Example 2 This embodiment provides a Chinese herbal medicine origin identification system, including: A target region acquisition module is used to obtain chromatograms of Chinese medicinal materials from different origins measured by an internal standard method, normalize the chromatograms to obtain a chromatogram set, obtain all unique peaks in the chromatograms of Chinese medicinal materials from different origins, and use the unique peak region as a target region; A feature set construction module, used to calculate the numerical range of each feature of the chromatographic peak of the internal standard in the chromatogram set, and construct multiple feature sets, wherein the features in the feature sets are derived from the features of the chromatographic peak of the internal standard, and each feature in the feature set corresponds to one of the numerical ranges; A sample enhancement module, used for enhancing a partial target area of ​​the chromatogram to be enhanced based on the feature set, and obtaining a training set using the partial target area and the chromatogram set; The origin identification module is used to train the neural network model using the training set, and obtain the origin of the Chinese medicinal materials using the chromatogram of the Chinese medicinal materials to be identified and the trained neural network model.

[0036] Optionally, the characteristics of the chromatographic peak include retention time, peak height, peak area, peak width, peak skewness, peak kurtosis, and peak standard deviation.

[0037] Optionally, the enhancing a partial target area of ​​the chromatogram to be enhanced based on the feature set is specifically: For each partial target area of ​​the chromatogram to be enhanced, randomly select a feature set from the multiple feature sets, and randomly generate an instance numerical sequence of the selected feature set, wherein the range of numerical values ​​in the instance numerical sequence is within the numerical range corresponding to the feature; A numerical sequence corresponding to the selected feature set in the chromatographic peak of the internal standard in the chromatogram to be enhanced is obtained, a change in the example numerical sequence relative to the numerical sequence is calculated, and a partial target area of ​​the chromatogram to be enhanced is enhanced based on the change to obtain an enhanced target area of ​​the chromatogram.

[0038] Optionally, the calculation example numerical sequence changes relative to the numerical sequence, and based on the changes, a partial target area of ​​the to-be-enhanced chromatogram is enhanced to obtain the enhanced target area of ​​the chromatogram, specifically: For each feature in the selected feature set, calculate the relative change rate of the value corresponding to the feature in the instance numerical sequence relative to the value corresponding to the feature in the numerical sequence; Obtain the characteristic value of the characteristic set of each target area in the partial target area in the chromatogram to be enhanced, and obtain the relative change rate corresponding to the characteristic value; obtain a new characteristic value based on the characteristic value and the relative change rate corresponding to the characteristic value; adjust the target area using the new characteristic value so that the characteristic value corresponding to the characteristic of the target area is the new characteristic value.

[0039] Optionally, the using the partial target region and the chromatogram set to obtain a training set is specifically: Obtain the target area of ​​the chromatogram in the chromatogram set, sample the target area to obtain the feature vector of the target area, splice the feature vectors of all the target areas in the chromatogram into a feature matrix by row, use the feature matrix as the sample in the training set, and annotate the sample as the origin of the Chinese medicinal material corresponding to the chromatogram; The enhanced partial target area is sampled to obtain a feature vector of the enhanced target area, and the feature vector corresponding to the partial target area in the feature matrix of the chromatogram where the partial target area is located is replaced with the feature vector of the enhanced target area to obtain a feature matrix of the enhanced chromatogram, and the feature matrix of the enhanced chromatogram is used as a sample in a training set, and the sample is labeled as the origin of the Chinese medicinal material to which the chromatogram where the partial target area is located belongs.

[0040] Example 3 This embodiment provides a computer-readable storage medium. Optionally, in this embodiment, the storage medium stores a computer program, and when the computer program is executed by a processor, the method for identifying the origin of Chinese medicinal materials provided in the above embodiment 1 is implemented.

[0041] Example 4 This embodiment provides a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, the method for identifying the origin of Chinese medicinal materials provided in the above embodiment 1 is implemented.

[0042] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0043] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0044] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0045] The above is only a preferred embodiment of the present application. It should be noted that ordinary technicians in this technical field can make several improvements and modifications without departing from the principle of the present application, and these improvements and modifications should also be regarded as the protection scope of the present application.

Claims

1. A method for identifying the origin of Chinese medicinal materials, characterized in that: The method comprises the following steps: Acquire chromatograms of Chinese medicinal materials from different origins measured by an internal standard method, normalize the chromatograms to obtain a chromatogram set, acquire all unique peaks in the chromatograms of Chinese medicinal materials from different origins, and use the unique peak region as a target region; Calculating the numerical range of each feature of the chromatographic peak of the internal standard in the chromatogram set, and constructing multiple feature sets, wherein the features in the feature sets are derived from the features of the chromatographic peak of the internal standard, and each feature in the feature set corresponds to one of the numerical ranges; Based on the feature set, a partial target region of the chromatogram to be enhanced is enhanced, and a training set is obtained using the partial target region and the chromatogram set; The training set is used to train the neural network model, and the chromatogram of the Chinese medicinal materials to be identified and the trained neural network model are used to obtain the origin of the Chinese medicinal materials.

2. The method according to claim 1, characterized in that The characteristics of the chromatographic peak include retention time, peak height, peak area, peak width, peak skewness, peak kurtosis, and peak standard deviation.

3. The method according to claim 1, characterized in that The enhancing of the partial target area of ​​the chromatogram to be enhanced based on the feature set is specifically: For each partial target area of ​​the chromatogram to be enhanced, randomly select a feature set from the multiple feature sets, and randomly generate an instance numerical sequence of the selected feature set, wherein the range of numerical values ​​in the instance numerical sequence is within the numerical range corresponding to the feature; A numerical sequence corresponding to the selected feature set in the chromatographic peak of the internal standard in the chromatogram to be enhanced is obtained, a change in the example numerical sequence relative to the numerical sequence is calculated, and a partial target area of ​​the chromatogram to be enhanced is enhanced based on the change to obtain an enhanced target area of ​​the chromatogram.

4. The method according to claim 3, characterized in that The calculation example numerical sequence changes relative to the numerical sequence, and based on the changes, a part of the target area of ​​the to-be-enhanced chromatogram is enhanced to obtain the enhanced target area of ​​the chromatogram, specifically: For each feature in the selected feature set, calculate the relative change rate of the value corresponding to the feature in the instance numerical sequence relative to the value corresponding to the feature in the numerical sequence; Obtain the characteristic value of the characteristic set of each target area in the partial target area in the chromatogram to be enhanced, and obtain the relative change rate corresponding to the characteristic value; obtain a new characteristic value based on the characteristic value and the relative change rate corresponding to the characteristic value; adjust the target area using the new characteristic value so that the characteristic value corresponding to the characteristic of the target area is the new characteristic value.

5. The method according to claim 1, characterized in that The method of obtaining a training set by using the partial target region and the chromatogram set is specifically as follows: Obtain the target area of ​​the chromatogram in the chromatogram set, sample the target area to obtain the feature vector of the target area, splice the feature vectors of all the target areas in the chromatogram into a feature matrix by row, use the feature matrix as the sample in the training set, and annotate the sample as the origin of the Chinese medicinal material corresponding to the chromatogram; The enhanced partial target area is sampled to obtain a feature vector of the enhanced target area, and the feature vector corresponding to the partial target area in the feature matrix of the chromatogram where the partial target area is located is replaced with the feature vector of the enhanced target area to obtain a feature matrix of the enhanced chromatogram, and the feature matrix of the enhanced chromatogram is used as a sample in a training set, and the sample is labeled as the origin of the Chinese medicinal material to which the chromatogram where the partial target area is located belongs.

6. A Chinese herbal medicine origin identification system, characterized in that: The system includes the following modules: A target region acquisition module is used to obtain chromatograms of Chinese medicinal materials from different origins measured by an internal standard method, normalize the chromatograms to obtain a chromatogram set, obtain all unique peaks in the chromatograms of Chinese medicinal materials from different origins, and use the unique peak region as a target region; A feature set construction module, used to calculate the numerical range of each feature of the chromatographic peak of the internal standard in the chromatogram set, and construct multiple feature sets, wherein the features in the feature sets are derived from the features of the chromatographic peak of the internal standard, and each feature in the feature set corresponds to one of the numerical ranges; A sample enhancement module, used for enhancing a partial target area of ​​the chromatogram to be enhanced based on the feature set, and obtaining a training set using the partial target area and the chromatogram set; The origin identification module is used to train the neural network model using the training set, and obtain the origin of the Chinese medicinal materials using the chromatogram of the Chinese medicinal materials to be identified and the trained neural network model.

7. The system according to claim 6, characterized in that The enhancing of the partial target area of ​​the chromatogram to be enhanced based on the feature set is specifically: For each partial target area of ​​the chromatogram to be enhanced, randomly select a feature set from the multiple feature sets, and randomly generate an instance numerical sequence of the selected feature set, wherein the range of numerical values ​​in the instance numerical sequence is within the numerical range corresponding to the feature; A numerical sequence corresponding to the selected feature set in the chromatographic peak of the internal standard in the chromatogram to be enhanced is obtained, a change in the example numerical sequence relative to the numerical sequence is calculated, and a partial target area of ​​the chromatogram to be enhanced is enhanced based on the change to obtain an enhanced target area of ​​the chromatogram.

8. The system according to claim 7, characterized in that The calculation example numerical sequence changes relative to the numerical sequence, and based on the changes, a part of the target area of ​​the to-be-enhanced chromatogram is enhanced to obtain the enhanced target area of ​​the chromatogram, specifically: For each feature in the selected feature set, calculate the relative change rate of the value corresponding to the feature in the instance numerical sequence relative to the value corresponding to the feature in the numerical sequence; Obtain the characteristic value of the characteristic set of each target area in the partial target area in the chromatogram to be enhanced, and obtain the relative change rate corresponding to the characteristic value; obtain a new characteristic value based on the characteristic value and the relative change rate corresponding to the characteristic value; adjust the target area using the new characteristic value so that the characteristic value corresponding to the characteristic of the target area is the new characteristic value.

9. The system according to claim 6, characterized in that The method of obtaining a training set by using the partial target region and the chromatogram set is specifically as follows: Obtain the target area of ​​the chromatogram in the chromatogram set, sample the target area to obtain the feature vector of the target area, splice the feature vectors of all the target areas in the chromatogram into a feature matrix by row, use the feature matrix as the sample in the training set, and annotate the sample as the origin of the Chinese medicinal material corresponding to the chromatogram; The enhanced partial target area is sampled to obtain a feature vector of the enhanced target area, and the feature vector corresponding to the partial target area in the feature matrix of the chromatogram where the partial target area is located is replaced with the feature vector of the enhanced target area to obtain a feature matrix of the enhanced chromatogram, and the feature matrix of the enhanced chromatogram is used as a sample in a training set, and the sample is labeled as the origin of the Chinese medicinal material to which the chromatogram where the partial target area is located belongs.

10. A computer storage device, wherein a computer program is stored on the storage device, characterized in that: When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 5.

Images