Field traditional Chinese medicinal material database construction method based on simulation system and artificial intelligence
By deeply integrating simulation systems with artificial intelligence, a dynamic database of Chinese medicinal materials is constructed, which solves the problems of static data and insufficient predictive capabilities. It enables dynamic simulation and prediction of the growth and effective components of Chinese medicinal materials, thereby improving the accuracy and timeliness of resource assessment and decision support.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-23
- Publication Date
- 2026-03-27
AI Technical Summary
Existing databases of Chinese medicinal herbs suffer from problems such as static data, lack of predictive capabilities, inability to reflect the impact of environmental factors, and untimely information updates, resulting in limited value for resource exploration and conservation applications.
By employing a method that deeply couples simulation systems with artificial intelligence, a dynamic database of Chinese medicinal materials is constructed. Through multi-source information acquisition, growth mechanism simulation models, medicinal active ingredient prediction models, and intelligent update mechanisms, dynamic simulation and prediction of the growth process and active ingredients of Chinese medicinal materials are achieved.
It enables dynamic simulation and prediction of the growth process and effective component content of Chinese medicinal materials, improves resource assessment and quality prediction capabilities, provides real-time decision support, breaks down the barriers between mechanistic models and data models, and improves the timeliness and accuracy of information.
Smart Images
Figure CN121747764A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of artificial intelligence, in particular to a field Chinese medicinal material database construction method based on simulation system and artificial intelligence. BACKGROUND
[0002] At present, there are some Chinese medicinal material databases at home and abroad, but they generally have the following limitations:
[0003] (1) Data static and one-sided. The existing databases are mostly static information lists, such as plant names, morphological descriptions, medicinal effects, etc. Once these data are entered, they are difficult to update, lack of dynamic market price information, and cannot reflect the influence of environmental factors (such as climate, soil) on the growth and effective component content of medicinal materials. The data dimension is single, and the internal relationship between "environment-growth-component-value" cannot be established.
[0004] (2) Lack of prediction and simulation capabilities. Traditional databases cannot predict the resources of Chinese medicinal materials in unknown areas, nor can they simulate the possible changes in the distribution, yield and quality of medicinal materials under future climate change scenarios. This limits its application value in resource exploration, protection and planning.
[0005] (3) Difficulty in verifying the authenticity of information. The identification of medicinal materials relies on text description and standard pictures, and lacks intelligent identification means. At the same time, the effective component content data of medicinal materials usually come from scattered literature or detection of specific samples, which is difficult to universalize and cannot realize rapid and low-cost estimation on a large scale.
[0006] (4) Data update depends on manual work. Dynamic information such as market price needs to be collected and entered manually, which is inefficient and prone to errors and lag, and cannot provide real-time and accurate decision support for market participants.
[0007] In summary, there is an urgent need in the field for a new method of constructing a field Chinese medicinal material database that can dynamically update, deeply integrate growth mechanisms, have prediction and simulation capabilities, and intelligently associate multi-source information. SUMMARY
[0008] The present application aims to overcome the shortcomings of the prior art and provide a field Chinese medicinal material database construction method based on simulation system and artificial intelligence, to solve the technical problems of existing Chinese medicinal material databases, such as data static isolation, lack of prediction capability, inability to reflect the influence of environment on quality, and delayed dynamic information update.
[0009] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:
[0010] The method for constructing a field Chinese herbal medicine database based on a simulation system and artificial intelligence comprises the following steps:
[0011] S1: Construct a multi-source information collection system to collect field Chinese herbal medicine information, including at least plant morphology information, geographical distribution information, ecological environment information, medicinal value information, and real-time market price information;
[0012] S2: Based on the ecological environment information and plant morphology information, a Chinese herbal medicine growth mechanism simulation model is constructed, which includes a photosynthetic production sub-model based on photosynthetic biochemical theory, a phenological development sub-model based on accumulated temperature theory, and an effective component accumulation sub-model related to secondary metabolism and environmental stress factors, for simulating the growth process of Chinese herbal medicine and the synthesis and accumulation process of medicinal effective components under the action of different environmental factors;
[0013] S3: Use machine learning algorithms to train the multi-source information collected in step S1 to construct a medicinal effective component prediction artificial intelligence model, which is used to predict the content of medicinal effective components according to the input of environmental factors and plant growth stage information;
[0014] S4: Coupling and collaborative optimization of the Chinese herbal medicine growth mechanism simulation model and the medicinal effective component prediction artificial intelligence model to form an intelligent simulation system;
[0015] The coupling process is: taking the minimization of the error between the output value of the Chinese herbal medicine growth mechanism simulation model and the predicted value of the medicinal effective component prediction artificial intelligence model as the optimization objective, and automatically optimizing the key physiological and ecological parameters in the Chinese herbal medicine growth mechanism simulation model through an optimization algorithm;
[0016] S5: Based on the output of the multi-source information collection system and the intelligent simulation system, a field Chinese herbal medicine database is constructed and dynamically updated. The records in the field Chinese herbal medicine database at least include plant name, Latin name, family, genus, geographical location, growth area, elevation range, media files of roots, stems, leaves, flowers and fruits, medicinal range, medicinal parts, medicinal functions, main medicinal effective component content, and real-time market price;
[0017] S6: Connect external authoritative price information sources through a data interface, and use web crawler technology to automatically obtain and update the market price information in the field Chinese herbal medicine database regularly.
[0018] Further, in step S1, the plant morphology information is acquired by an image acquisition device, and an image recognition effective ingredient prediction artificial intelligence model based on a convolutional neural network (CNN) is used to automatically identify and classify the media file. When the recognition confidence is higher than a preset threshold, it is automatically associated with the corresponding traditional Chinese medicinal material item; when the confidence is lower than the preset threshold, the media file is marked for manual review.
[0019] Further, in step S2, the input of the traditional Chinese medicinal material growth mechanism simulation model is ecological environment information, including climate data, soil composition data and terrain data; the effective ingredient accumulation sub-model takes environmental stress factors as key input variables to stimulate the synthesis of medicinal effective ingredients, and outputs include plant biomass, phenological period and the accumulation amount of one or more medicinal effective ingredients.
[0020] Further, in step S3, the medicinal effective ingredient prediction artificial intelligence model adopts a gradient boosting decision tree model or a deep neural network (DNN) model; wherein the DNN model contains at least three hidden layers and uses the Dropout technology for regularization to prevent overfitting.
[0021] Further, in step S3, the medicinal effective ingredient prediction artificial intelligence model is a multi-modal fusion network, which includes a first deep neural network (DNN) branch for processing ecological environment information, a second convolutional neural network (CNN) branch for processing plant morphology image information, and a joint decision layer for fusing the feature vectors of the two branches and outputting the final prediction result.
[0022] Further, in step S4, the collaborative optimization is an iterative process of constructing a digital twin model of traditional Chinese medicinal materials: simulated data is generated using the parameter-optimized traditional Chinese medicinal material growth mechanism simulation model, and the simulated data is added to the training set of the medicinal effective ingredient prediction artificial intelligence model to train a next-generation medicinal effective ingredient prediction artificial intelligence model with stronger prediction ability, to train a next-generation medicinal effective ingredient prediction artificial intelligence model with stronger ability, so as to realize the co-evolution of the traditional Chinese medicinal material growth mechanism simulation model and the medicinal effective ingredient prediction artificial intelligence model in the closed-loop feedback.
[0023] Further, in step S5, the field traditional Chinese medicinal material database also stores the traditional Chinese medicinal material distribution and quality prediction data generated by the intelligent simulation system for different future climate scenarios.
[0024] An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method when executing the program.
[0025] A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method.
[0026] Compared with the prior art, the present application has the following remarkable beneficial effects:
[0027] From static list to dynamic prediction and simulation: the present application realizes dynamic simulation and prediction of the growth process and effective component content of traditional Chinese medicinal materials through an intelligent simulation system, can especially perform resource and quality evaluation on unsampled areas, and simulates the influence of future climate change on medicinal materials, thereby providing an unprecedented decision support tool for resource protection and strategic planning.
[0028] From information isolation to deep correlation: the present application first deeply couples a traditional Chinese medicinal material growth mechanism simulation model and a medicinal effective component prediction artificial intelligence model, breaks the barrier between the mechanism model and the data model. AI improves the accuracy of the traditional Chinese medicinal material growth mechanism simulation model, and the traditional Chinese medicinal material growth mechanism simulation model provides generated data and mechanism constraints for AI, forms a "1+1>2" synergistic effect, and deeply reveals the internal law between "environment-growth-component".
[0029] From passive recording to active perception and updating: by introducing an image recognition medicinal effective component prediction artificial intelligence model, intelligent auxiliary identification of traditional Chinese medicinal materials is realized; through network crawler technology, automatic and real-time updating of market price information is realized, and the timeliness and practicality of the database are greatly improved.
[0030] Improve the breadth and depth of data: through a multi-modal fusion medicinal effective component prediction artificial intelligence model, structured environmental data and unstructured image data can be comprehensively utilized for more accurate comprehensive prediction, greatly enriching the information dimension and value of the database. BRIEF DESCRIPTION OF DRAWINGS
[0031] Figure 1 is the technical process of an embodiment of the present application. DETAILED DESCRIPTION
[0032] The present application will be further described in detail below in combination with a preferred embodiment.
[0033] Figure 1 is the technical roadmap of an embodiment of the present application, which shows the overall process and system architecture of the method. As shown in Figure 1 , the technical roadmap includes four main stages:
[0034] First stage: Data cornerstone and preprocessing. This stage includes multi-source data collection (corresponding to S1), data cleaning and labeling, and structured data warehouse construction. Multi-source data collection covers plant morphology information, geographic distribution information, ecological environment information, medicinal value information, and real-time market price information.
[0035] Second stage: Core model construction and coupling. This stage includes the construction of a growth mechanism simulation model (corresponding to S2), an image recognition model, a medicinal effective component prediction artificial intelligence model (corresponding to S3), and model coupling and optimization (corresponding to S4). Model coupling and optimization is an iterative optimization loop, in which the medicinal effective component prediction artificial intelligence model outputs prediction data while generating simulated data feedback training, thereby optimizing the traditional Chinese medicinal material growth mechanism simulation model.
[0036] Third stage: System integration and database generation. This stage includes the formation of an intelligent simulation system, the connection of external data interfaces (corresponding to S6), and dynamic database construction (corresponding to S5). The intelligent simulation system integrates the optimized model, the external data interface is used to update market prices, and the dynamic database construction stores all information.
[0037] Fourth stage: Application and service. This stage provides various application services through a Web platform / API interface, including resource prediction and zoning, market insight and early warning, quality assessment and traceability, etc., supporting various practical applications.
[0038] This example takes the construction of a database for a rare highland traditional Chinese medicinal material as an example, and the specific steps are as follows:
[0039] (1) Multi-source information collection
[0040] Obtain the ecological environment information of the target area from satellite remote sensing, weather stations, and soil survey reports, including annual mean temperature, annual precipitation, sunshine hours, soil type, and pH value.
[0041] Through field investigation and cooperation with research institutions, obtain the plant morphology information (high-definition photos of roots, stems, leaves, flowers, and fruits) and medicinal value information (medicinal parts, functions, and laboratory-determined effective component content) of the medicinal material.
[0042] Obtain the historical and current market prices of the medicinal material from authoritative price information sources such as the national traditional Chinese medicinal material price information platform.
[0043] (2) Model construction and coupling optimization
[0044] Construction of the simulation model of the growth mechanism of the Chinese medicinal material: In the simulation platform, the growth mechanism model of the medicinal material is established. The phenology sub-model uses the accumulated temperature rule to determine its flowering period; the photosynthesis sub-model calculates its biomass accumulation; and the effective component accumulation sub-model sets the synthesis peak period at the flowering period, and introduces a "water stress factor" as a key parameter for increasing the content of the component.
[0045] Construction of the artificial intelligence model for predicting the medicinal effective component: about 500 groups of samples containing environmental data and corresponding measured component contents are collected. Using 80% of the data, an effective component prediction model is trained by using the XGBoost library.
[0046] Model coupling: the five parameters of "maximum photosynthetic rate", "component synthesis basic rate" and the like in the simulation model of the growth mechanism of the Chinese medicinal material are set as adjustable variables. Taking the prediction results of the artificial intelligence model for predicting the medicinal effective component on all samples as the benchmark, the Bayesian optimization algorithm is adopted to automatically adjust the five parameters, so that the mean square error between the output of the simulation model of the growth mechanism of the Chinese medicinal material and the AI prediction value at all sample points is minimized. After 1000 iterations, the optimal solution of the parameters is obtained, and at this time, the prediction accuracy of the simulation model of the growth mechanism of the Chinese medicinal material is improved by 35% compared with that before coupling.
[0047] (3) Database construction and intelligent application
[0048] All the above information and the optimized intelligent simulation system are integrated into a database management system.
[0049] Application example 1 (resource prediction): inputting the environmental data of an uninvestigated area, the intelligent simulation system can predict that the area is suitable for the growth of the medicinal material, and estimate that the content of the effective component can reach 1.2 times of the standard value, thereby providing accurate targeting for field investigation.
[0050] Application example 2 (image recognition): a user uploads a leaf photo of the medicinal material, and the image recognition model based on the ResNet50 transfer learning can quickly identify and return the complete field Chinese medicinal material database entry of the medicinal material.
[0051] Application example 3 (market monitoring): a background web crawler program is automatically run every day to capture the latest market price and update the field Chinese medicinal material database, and users can query the price trend in real time.
[0052] The above merely illustrates the preferred embodiments of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can make equivalent replacements or changes to the technical solutions and inventive concepts of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.
Claims
1. A method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence, characterized in that, Includes the following steps: S1: Construct a multi-source information collection system. The collected information on wild Chinese medicinal materials should include at least: plant morphology information, geographical distribution information, ecological environment information, medicinal value information, and real-time market price information. S2: Based on the ecological environment information and plant morphology information, a simulation model of the growth mechanism of Chinese medicinal materials is constructed. The simulation model of the growth mechanism of Chinese medicinal materials includes a photosynthetic production sub-model based on photosynthetic biochemistry theory, a phenological development sub-model based on accumulated temperature theory, and an effective component accumulation sub-model related to secondary metabolism and environmental stress factors. It is used to simulate the growth process of Chinese medicinal materials and the synthesis and accumulation process of medicinal effective components under the action of different environmental factors. S3: Using machine learning algorithms, train the multi-source information collected in step S1 to construct an artificial intelligence model for predicting the content of medicinal active ingredients. The artificial intelligence model for predicting the content of medicinal active ingredients is used to predict the content of medicinal active ingredients based on the input environmental factors and plant growth stage information. S4: Couple and synergistically optimize the simulation model of the growth mechanism of Chinese medicinal materials with the artificial intelligence model for predicting the effective components of medicinal materials to form an intelligent simulation system; The coupling process is as follows: with the optimization objective of minimizing the error between the output value of the simulation model of the growth mechanism of Chinese medicinal materials and the prediction value of the artificial intelligence model for predicting the effective medicinal components, the key physiological and ecological parameters in the simulation model of the growth mechanism of Chinese medicinal materials are automatically optimized through an optimization algorithm. S5: Based on the output of the multi-source information acquisition system and the intelligent simulation system, construct and dynamically update the wild Chinese medicinal materials database; the records in the wild Chinese medicinal materials database shall include at least: plant name, Latin name, family, genus, geographical location, growing area, altitude range, media files of roots, stems, leaves, flowers and fruits, medicinal scope, medicinal parts, medicinal functions, content of main medicinal active ingredients and real-time market price. S6: Connect to an external authoritative price information source through a data interface, and use web crawler technology to regularly and automatically obtain and update market price information in the wild Chinese medicinal materials database.
2. The method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence according to claim 1, characterized in that, In step S1, the plant morphology information is acquired through an image acquisition device, and the media file is automatically identified and classified using an artificial intelligence model for predicting medicinal active ingredients based on a convolutional neural network (CNN). When the identification confidence level is higher than a preset threshold, it is automatically associated with the corresponding Chinese medicinal material item; when the confidence level is lower than the preset threshold, the media file is marked as requiring manual review.
3. The method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence according to claim 1, characterized in that, In step S2, the input to the simulation model of the growth mechanism of Chinese medicinal materials is ecological environment information, including climate data, soil composition data and topographic data; the effective component accumulation sub-model takes environmental stress factors as key input variables to stimulate the synthesis of medicinal effective components, and outputs plant biomass, phenological period and accumulation of one or more medicinal effective components.
4. The method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence according to claim 1, characterized in that, In step S3, the artificial intelligence model for predicting the active pharmaceutical ingredient adopts a gradient boosting decision tree model or a deep neural network (DNN) model; wherein the DNN model contains at least three hidden layers and uses Dropout technology for regularization to prevent overfitting.
5. The method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence according to claim 1 or 4, characterized in that, In step S3, the artificial intelligence model for predicting the active medicinal ingredient is a multimodal fusion network, which includes: a first deep neural network (DNN) branch for processing ecological environment information, a second convolutional neural network (CNN) branch for processing plant morphology image information, and a joint decision layer for fusing the feature vectors of the two branches and outputting the final prediction result.
6. The method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence according to claim 1, characterized in that, In step S4, the collaborative optimization is an iterative process of constructing a digital twin model of Chinese medicinal materials: using the simulated model of the growth mechanism of Chinese medicinal materials with optimized parameters to generate simulated data, and adding the simulated data to the training set of the artificial intelligence model for predicting the effective medicinal components, so as to train a next-generation artificial intelligence model for predicting the effective medicinal components with stronger predictive ability, and so on, thereby achieving the co-evolution of the simulated model of the growth mechanism of Chinese medicinal materials and the artificial intelligence model for predicting the effective medicinal components in the closed-loop feedback.
7. The method for constructing a field database of Chinese medicinal herbs based on simulation systems and artificial intelligence according to claim 1, characterized in that, In step S5, the field Chinese medicinal herb database also stores the distribution and quality prediction data of Chinese medicinal herbs under different future climate scenarios generated by the intelligent simulation system.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1-7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1-7.