Method and system for dynamically predicting risk of harm of feed mycotoxin to livestock and poultry

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a method and a system for dynamically predicting the risk of harm of feed mycotoxin to livestock and poultry. The method comprises the following steps: constructing a metabolic digital twin model based on a multi-organ coupling dynamics mechanism, connecting a gastrointestinal tract, a liver and a kidney through blood flow parameters to form a closed loop topology network, establishing a toxin migration rate equation, and verifying bioavailability; collecting feed toxin concentration and livestock and poultry physiological data in real time; after distributing the initial toxin load, iteratively generating target organ dynamic concentration distribution by adopting a Runge-Kutta method; an organ exposure index is calculated based on an organ specific hazard threshold, key organs are positioned, and low / high risk levels and toxin accumulation paths are output. The method solves the problems that a traditional static model is low in efficiency and poor in organ difference adaptability when being used for dynamically predicting the risk of harm to livestock and poultry caused by feed mycotoxin.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and system for dynamically predicting the risk of livestock and poultry harm caused by feed mycotoxins. Background Art

[0002] Mycotoxin contamination in feed is a key threat to livestock and poultry health. Traditional detection methods rely primarily on static toxicity threshold models, which have significant limitations. Existing technologies typically detect toxin concentrations based on single sampling, failing to dynamically reflect the continuous metabolic processes of toxins in livestock and poultry. Furthermore, fixed threshold models ignore the impact of individual physiological differences, organ specificity (such as differences in liver metabolism and kidney excretion), and growth stage, resulting in delayed risk assessment and insufficient universality. Furthermore, the lack of dynamic simulation of toxin migration pathways across multiple organs makes it difficult to accurately locate high-risk target organs, resulting in poor early warning timeliness and insufficiently targeted prevention and control measures.

[0003] Current prediction systems generally fail to integrate the impact of real-time physiological parameters (such as body temperature, heart rate, and digestion rate) on toxin metabolic dynamics. Furthermore, organ damage thresholds often use a unified standard without linking them to specific biomarkers such as intestinal villus height, ALT activity, and glomerular filtration rate. This static assessment model is unable to adapt to the variability of livestock and poultry metabolic states, and errors are particularly exacerbated during early life or under stress. Therefore, a prediction method that integrates multi-organ coupling mechanisms and adaptive thresholds is urgently needed to address the low efficiency of dynamic prediction of livestock and poultry risk from feed mycotoxins. Summary of the Invention

[0004] The present invention provides a method and system for dynamically predicting the risk of harm to livestock and poultry caused by feed mycotoxins. Its main purpose is to solve the problem of low prediction efficiency caused by the inability of traditional static models to reflect the dynamic migration process of toxins in multiple organs of livestock and poultry in real time and the poor adaptability of organ-specific damage thresholds.

[0005] To achieve the above objectives, the present invention provides a method for dynamically predicting the risk of livestock and poultry damage caused by mycotoxins in feed, comprising:

[0006] Construct a digital twin model of the metabolism of feed mycotoxins in target organs of livestock and poultry. The digital twin model is based on the kinetic mechanisms of toxin absorption, distribution, metabolism, and excretion coupled with multiple organs.

[0007] Real-time collection of mycotoxin concentration data in feed and physiological status data of livestock and poultry;

[0008] Inputting the mycotoxin concentration data and the physiological status data into the metabolic digital twin model to generate a dynamic concentration distribution of mycotoxins in the target organs of the livestock and poultry;

[0009] Based on the dynamic concentration distribution and the preset organ-specific hazard threshold, a livestock and poultry health risk level prediction result is output.

[0010] Optionally, the construction of a digital twin model of the metabolism of feed mycotoxins in target organs of livestock and poultry includes:

[0011] The gastrointestinal tract, liver and kidney of livestock and poultry are connected through hemodynamic parameters to obtain a closed-loop topological network corresponding to the livestock and poultry target organs;

[0012] The migration rate equation of the mycotoxin in the target organs of livestock and poultry is established based on the law of conservation of mass, and the bioavailability of the dynamic concentration distribution is verified by virtual mycotoxin injection.

[0013] Optionally, verifying the biological effectiveness of the dynamic concentration distribution by virtual mycotoxin injection includes:

[0014] Setting a pulsed mycotoxin input at the inlet of the closed-loop topology network;

[0015] Calculate the mycotoxin deviation between the total amount of mycotoxins at the outlet of all livestock and poultry target organs and the pulse mycotoxin input amount;

[0016] If the mycotoxin deviation exceeds a preset tolerance value, reversely optimizing the hemodynamic parameters;

[0017] When the mycotoxin deviation of three consecutive virtual injections is lower than the preset tolerance value, a biovalidity certification of the dynamic concentration distribution is issued.

[0018] Optionally, the migration rate equation is:

[0019]

[0020] in, It is The amount of mycotoxins in the target organs of livestock and poultry, From the livestock and poultry target organs To the livestock target organs The migration rate constant, From the livestock and poultry target organs To the livestock target organs The migration rate constant, It is The amount of mycotoxins in the target organs of livestock and poultry, is an external input to the livestock target organ The mycotoxin rate, The livestock and poultry target organs The excretion rate, is a time stamp, Indicates the The target organs of livestock and poultry, Indicates the The target organs of livestock and poultry.

[0021] Optionally, the real-time collection of mycotoxin concentration data in feed and physiological status data of livestock and poultry includes:

[0022] The mycotoxin concentration data in the feed is obtained through spectral sensors, and the body temperature, heart rate and digestion rate of livestock and poultry are simultaneously collected as the physiological status data of livestock and poultry.

[0023] Optionally, the dynamic concentration distribution of the generated mycotoxins in the target organs of the livestock and poultry includes:

[0024] allocating the initial toxin load of each of the livestock and poultry target organs according to the physiological status data and the mycotoxin concentration data;

[0025] Based on the initial toxin load, the dynamic concentration distribution of mycotoxins in the closed-loop topological network is iteratively solved in combination with the Runge-Kutta method.

[0026] Optionally, the determination of the preset organ-specific hazard threshold includes:

[0027] Screening the benchmark damage thresholds of the gastrointestinal tract, the liver, and the kidney from the species toxicology database;

[0028] The gastrointestinal threshold was correlated with intestinal villus height, the liver threshold was correlated with ALT activity, and the kidney threshold was correlated with glomerular filtration rate;

[0029] Determining the specific safety factors of the livestock and poultry target organs one by one according to the correlation relationship of the livestock and poultry target organs and the growth stage of the livestock and poultry;

[0030] The baseline damage threshold is adjusted based on the specific safety factor to obtain the organ-specific hazard threshold of the livestock and poultry.

[0031] Optionally, the output of the livestock and poultry health risk level prediction result includes:

[0032] The organ exposure index is generated by dividing the dynamic concentration distribution by the threshold value of the corresponding livestock and poultry target organ, wherein the calculation formula of the organ exposure index is:

[0033]

[0034] in, For the The first time point Organ exposure index of target organs of livestock and poultry, For the The first time point Toxin concentration in target organs of livestock and poultry, For the Preset thresholds for target organs of livestock and poultry;

[0035] Screening the livestock and poultry target organs where the maximum value of the organ exposure index is located as key organs;

[0036] If all of the organ exposure indexes are less than a first preset threshold, outputting a low risk;

[0037] If the organ exposure index of any organ exceeds a second preset threshold, a high risk is output, and the key organs and toxin accumulation pathways causing the high risk are marked.

[0038] Optionally, the solution formula for the dynamic concentration distribution is:

[0039]

[0040] in, is the mycotoxin concentration at the current time step, is the mycotoxin concentration at the next time step, is the initial migration rate calculated based on the current state and the migration rate equation, is the migration rate correction value at half the time step, is the quadratic correction value at half the time step, is the final migration rate at the complete time step, is the time stamp, is the time step.

[0041] In order to solve the above problems, the present invention also provides a system for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins, the system comprising:

[0042] A twin model construction module is used to construct a digital twin model of the metabolism of feed mycotoxins in the target organs of livestock and poultry. The metabolic digital twin model is based on the kinetic mechanism of toxin absorption, distribution, metabolism and excretion coupled with multiple organs;

[0043] Data acquisition module, used to collect real-time data on mycotoxin concentration in feed and physiological status of livestock and poultry;

[0044] a concentration distribution generation module, configured to input the mycotoxin concentration data and the physiological status data into the metabolic digital twin model to generate a dynamic concentration distribution of mycotoxins in the target organs of the livestock and poultry;

[0045] The risk level prediction module is used to output the livestock and poultry health risk level prediction result based on the dynamic concentration distribution and the preset organ-specific hazard threshold.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. By constructing a multi-organ closed-loop topological network based on hemodynamic parameters and establishing a toxin migration rate equation based on the law of conservation of mass, a digital twin model is used to map the absorption, distribution, metabolism, and excretion of toxins in the gastrointestinal tract, liver, and kidneys in real time. Virtual mycotoxin injection is used to verify bioavailability, optimize kinetic parameters, and significantly improve the accuracy of dynamic concentration distribution calculations. Combined with the Runge-Kutta method for iterative solution, this approach overcomes the limitation of traditional static models that cannot track the continuous migration path of toxins.

[0048] 2. Dynamically generate organ-specific hazard thresholds based on species toxicology databases and organ physiological indicators (such as intestinal villus height, ALT activity, and glomerular filtration rate), and introduce growth stage safety factor correction to address the problem of poor adaptability of unified thresholds; locate key damaged organs and toxin accumulation pathways through organ exposure indexes, and adjust the initial toxin load in combination with real-time physiological data (body temperature, heart rate, digestion rate) to achieve automated discrimination of low-risk / high-risk levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic flow chart of a method for dynamically predicting the risk of livestock and poultry damage caused by mycotoxins in feed provided in one embodiment of the present invention;

[0050] Figure 2 This is a functional module diagram of a system for dynamically predicting the risk of livestock and poultry damage caused by mycotoxins in feed provided by one embodiment of the present invention;

[0051] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0052] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] The embodiment of the present application provides a method for dynamically predicting the risk of harm to livestock and poultry caused by mycotoxins in feed. The execution subject of the method for dynamically predicting the risk of harm to livestock and poultry caused by mycotoxins in feed includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for dynamically predicting the risk of harm to livestock and poultry caused by mycotoxins in feed can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0054] Reference Figure 1 FIG. 1 is a flow chart of a method for dynamically predicting the risk of livestock and poultry hazard caused by mycotoxins in feed provided by an embodiment of the present invention. In this embodiment, the method for dynamically predicting the risk of livestock and poultry hazard caused by mycotoxins in feed includes:

[0055] S1. Construct a metabolic digital twin model of feed mycotoxins in the target organs of livestock and poultry. The metabolic digital twin model is based on the kinetic mechanism of toxin absorption, distribution, metabolism and excretion coupled with multiple organs.

[0056] In an embodiment of the present invention, the construction of a digital twin model of the metabolism of feed mycotoxins in target organs of livestock and poultry includes:

[0057] The gastrointestinal tract, liver and kidney of livestock and poultry are connected through hemodynamic parameters to obtain a closed-loop topological network corresponding to the livestock and poultry target organs;

[0058] The migration rate equation of the mycotoxin in the target organs of livestock and poultry is established based on the law of conservation of mass, and the bioavailability of the dynamic concentration distribution is verified by virtual mycotoxin injection.

[0059] In detail, the verification of the biological effectiveness of the dynamic concentration distribution by virtual mycotoxin injection includes:

[0060] Setting a pulsed mycotoxin input at the inlet of the closed-loop topology network;

[0061] Calculate the mycotoxin deviation between the total amount of mycotoxins at the outlet of all livestock and poultry target organs and the pulse mycotoxin input amount;

[0062] If the mycotoxin deviation exceeds a preset tolerance value, reversely optimizing the hemodynamic parameters;

[0063] When the mycotoxin deviation of three consecutive virtual injections is lower than the preset tolerance value, a biovalidity certification of the dynamic concentration distribution is issued.

[0064] In detail, the migration rate equation is:

[0065]

[0066] in, It is The amount of mycotoxins in the target organs of livestock and poultry, From the livestock and poultry target organs To the livestock target organs The migration rate constant, From the livestock and poultry target organs To the livestock target organs The migration rate constant, It is The amount of mycotoxins in the target organs of livestock and poultry, is an external input to the livestock target organ The mycotoxin rate, The livestock and poultry target organs The excretion rate, is a time stamp, Indicates the The target organs of livestock and poultry, Indicates the The target organs of livestock and poultry.

[0067] In detail, Corresponding to the gastrointestinal tract, the liver and the kidneys.

[0068] In detail, the metabolic digital twin model is a virtual model constructed through digital technology. It can map and simulate the metabolic process of feed mycotoxins in the target organs of livestock and poultry in real time based on the kinetic mechanism of toxin absorption, distribution, metabolism and excretion of multiple organs; target organs refer to the main organs in the body of livestock and poultry that are easily affected by feed mycotoxins, specifically the gastrointestinal tract, liver and kidneys in this invention.

[0069] In detail, hemodynamic parameters are used to describe the parameters of blood flow in livestock and poultry, including blood flow rate, blood flow velocity, etc. These parameters can reflect the flow state of blood between organs, and thus affect the migration of mycotoxins between organs.

[0070] Specifically, a closed-loop topological network refers to a network structure formed by connecting the gastrointestinal tract, liver, and kidneys through hemodynamic parameters. The migration path of mycotoxins in this network forms a closed loop, which can more accurately simulate the circulation and metabolic process of toxins between organs; the migration rate equation refers to an equation based on the law of conservation of mass that is used to describe the migration rate of mycotoxins in the target organs of livestock and poultry. This equation can be used to calculate the changes in the amount of mycotoxins in each target organ at different time points.

[0071] In detail, pulsed mycotoxin input refers to an instantaneous, concentrated mycotoxin input mode set at the entrance of the closed-loop topology network, which is similar to a short pulse signal and is used to verify the biological validity of the model.

[0072] Specifically, mycotoxin deviation refers to the ratio of the difference between the total amount of mycotoxins at the outlet of all target organs of livestock and poultry and the pulse mycotoxin input to the input amount, which is used to measure the degree of deviation between the model calculation results and the actual situation.

[0073] In detail, the preset tolerance value refers to a pre-set threshold used to determine whether the mycotoxin deviation is acceptable. When the deviation is lower than this value, the calculation result of the model is considered to meet the requirements.

[0074] Furthermore, the gastrointestinal tract, liver, and kidneys of livestock and poultry are specifically identified as target organs for research. These organs are the primary sites for mycotoxin absorption, metabolism, and excretion after entering the body. Through research and experimental measurement of livestock and poultry physiological characteristics, hemodynamic parameters of each target organ, such as gastrointestinal blood flow and liver blood flow velocity, can be obtained. These parameters can be obtained through medical testing methods and statistical analysis of animal experimental data.

[0075] Furthermore, the connectivity and strength of each target organ are determined based on hemodynamic parameters. For example, blood flows from the gastrointestinal tract to the liver, and then from the liver to the kidneys. These return paths connect the three organs into a closed-loop topological network. When constructing the network, the direction and volume of blood flow must be considered to ensure that the network accurately reflects the migration paths of toxins between organs.

[0076] Furthermore, the law of conservation of mass states that in a closed system, the total amount of a substance does not change over time. For mycotoxins in target organs of livestock and poultry, the change in the amount of toxin in a particular organ is equal to the amount of toxin entering that organ minus the amount of toxin leaving that organ, plus the amount of toxin input minus the amount of toxin excreted.

[0077] In detail, the meanings of the parameters in the migration rate equation are as follows: Indicates the The amount of mycotoxins in target organs, Corresponding to the gastrointestinal tract, liver and kidneys; From the target organ To target organs The migration rate constant, which reflects the toxin's Migration to organs The speed of migration can be determined experimentally, for example, by simulating the toxin migration process between organs in vitro, measuring the changes in toxin concentration at different time points, and thus calculating the migration rate constant. From the target organ To target organs The migration rate constant of is determined by similar experimental methods. It is The amount of mycotoxins in each target organ; External input to target organs The rate of mycotoxins in feed entering the gastrointestinal tract through livestock and poultry ingestion; Target organs The excretion rate, that is, the rate at which toxins are eliminated from that organ, such as the rate at which the kidneys excrete toxins through urine; It is a time marker used to indicate different time points.

[0078] Furthermore, according to the law of conservation of mass and the defined variables, a migration rate equation is established, which represents the The rate of change of the amount of mycotoxins in a target organ over time is equal to the sum of the toxins that migrate into this organ from all other target organs minus the sum of the toxins that migrate from this organ to all other target organs, plus the rate of toxins externally input into this organ minus the excretion rate of this organ.

[0079] Furthermore, a pulsed mycotoxin input is set at the entrance of the closed-loop topology network, such as the entrance to the gastrointestinal tract. Specifically, a certain amount of mycotoxin is instantaneously injected at a certain moment to simulate the situation where livestock and poultry suddenly ingest a certain amount of feed containing mycotoxins. The input amount can be set based on actual conditions and experimental requirements, for example, setting the input amount to 0 micrograms.

[0080] Furthermore, after a period of time after the pulse mycotoxin injection, the total amount of mycotoxins at the outlet of all target organs was calculated. The mycotoxin deviation is then calculated , the formula is This deviation reflects the degree of discrepancy between the total amount of toxins exported as calculated by the model and the actual amount of toxins input. Optimizing hemodynamic parameters: If the calculated mycotoxin deviation exceeds the preset tolerance, it indicates that the model's results deviate significantly from the actual situation and requires reverse optimization of hemodynamic parameters. For example, parameters such as liver blood flow and blood velocity can be adjusted, and then pulse input and deviation calculations can be repeated until the deviation meets the requirements.

[0081] Furthermore, when virtual injection experiments were conducted three times in a row and the mycotoxin deviation was lower than the preset tolerance value each time, it indicated that the model could accurately simulate the metabolic process of toxins under different input conditions. At this time, the bioavailability certification of the dynamic concentration distribution was issued, proving that the constructed metabolic digital twin model had high accuracy and reliability.

[0082] In summary, by constructing a metabolic digital twin model, a dynamic simulation of the metabolic processes of feed mycotoxins in the target organs of livestock and poultry was achieved. Based on the kinetic mechanism of multi-organ coupling, this model can accurately reflect the migration, transformation, and excretion of toxins between organs. Virtual injection verification ensures the bioavailability of the model, allowing the model to more realistically simulate actual conditions. This process addresses the problem in the background art that traditional prediction methods are unable to dynamically and accurately predict the risk of feed mycotoxins to livestock and poultry, improves the accuracy and reliability of predictions, and provides strong technical support for real-time monitoring and early warning of livestock and poultry health risks.

[0083] In general, constructing a closed-loop topological network is the basis for establishing the migration rate equation. Only by first determining the connection relationship and hemodynamic parameters between the target organs can the equation describing toxin migration be accurately established. Establishing the migration rate equation is the prerequisite for virtual injection verification. Through this equation, the distribution and migration of toxins between organs under pulsed input can be calculated. Virtual injection verification verifies the accuracy of the constructed model and the established equation, and ensures that the model can accurately simulate the actual metabolic process by optimizing the hemodynamic parameters. These three sub-steps are closely linked and interlocking, and together constitute the complete process of building a metabolic digital twin model, laying a solid foundation for subsequent toxin concentration data collection, model input and risk level prediction.

[0084] S2. Real-time collection of mycotoxin concentration data in feed and physiological status data of livestock and poultry.

[0085] In an embodiment of the present invention, the real-time collection of mycotoxin concentration data in feed and physiological status data of livestock and poultry includes:

[0086] The mycotoxin concentration data in the feed is obtained through spectral sensors, and the body temperature, heart rate and digestion rate of livestock and poultry are simultaneously collected as the physiological status data of livestock and poultry.

[0087] In detail, a spectral sensor is an instrument that can obtain information about the composition of a substance by detecting the absorption, reflection or transmission characteristics of the substance to light of different wavelengths. In the present invention, it is used to detect the concentration of mycotoxins in feed.

[0088] Specifically, mycotoxin concentration data refers to the numerical value of the mycotoxin content contained in the feed, which is obtained after the feed is tested and analyzed by a spectral sensor and is used to reflect the degree of mycotoxin contamination in the feed.

[0089] In detail, the physiological status data of livestock and poultry is a series of data that characterizes the body functions and physiological activities of livestock and poultry. In the present invention, it specifically includes body temperature, heart rate and digestion rate. These data can reflect the health status and metabolic level of livestock and poultry. Among them, body temperature refers to the temperature inside the body of livestock and poultry. It is one of the important indicators of the physiological activities of livestock and poultry. It is measured by a specific temperature sensor and can be used to determine whether the livestock and poultry are in a normal physiological state; heart rate refers to the number of times the heart of livestock and poultry beats per minute. It is obtained through heart rate monitoring equipment and can reflect the heart function and physical stress state of livestock and poultry; digestion rate refers to the speed at which livestock and poultry digest the ingested food. In the present invention, it is evaluated through relevant physiological indicators or signals to reflect the functional status of the digestive system of livestock and poultry.

[0090] Furthermore, a spectral sensor suitable for feed testing, such as a near-infrared spectral sensor, should be selected. Its operating wavelength range typically covers the visible to near-infrared region, enabling specific optical interactions with mycotoxins. This sensor should be installed at a suitable location along the feed transport path, such as above a conveyor belt in a feed processing line or at the sampling port of a feed storage container, ensuring that the sensor can directly detect feed samples.

[0091] Furthermore, as feed passes through the sensor's detection area, the sensor emits light of a specific wavelength onto the feed sample and simultaneously receives light signals reflected or transmitted from the feed sample. Photoelectric conversion elements within the sensor convert the light signals into electrical signals, which in turn generate spectral data of the feed. This process occurs in real time, with the sensor continuously collecting spectral data as the feed continues to flow.

[0092] The collected spectral data is then analyzed and processed using a pre-established mathematical model linking mycotoxin concentration and spectral data. Using machine learning algorithms or chemometric methods, such as partial least squares (PLS), spectral features are correlated with mycotoxin standards of known concentration to create a calibration curve. Once the spectral data of an unknown feed sample is acquired, it is substituted into the calibration curve to calculate the mycotoxin concentration in the feed. To ensure detection accuracy, the sensor is regularly calibrated using standard samples, and model parameters are adjusted to ensure the reliability of concentration calculations.

[0093] Furthermore, wireless intelligent body temperature sensors are used, such as implantable or wearable sensors. Implantable sensors are implanted in a suitable location in the livestock's body, such as subcutaneously or intraperitoneally, through minimally invasive surgery. The sensors monitor the livestock's internal temperature in real time and transmit the data to a data acquisition terminal via wireless communication. Wearable sensors are attached or worn on the surface of the livestock's body, such as the ears or back, and measure body temperature using infrared thermometry or thermistor principles, also transmitting data wirelessly. The data acquisition terminal receives and records body temperature data at a set sampling frequency, such as once per minute.

[0094] Furthermore, bioelectric sensors or Doppler ultrasound sensors are used. Bioelectric sensors use electrodes attached to the animal's body to collect bioelectric signals generated by cardiac electrical activity, known as electrocardiograms (ECGs). After signal amplification and filtering, data acquisition equipment calculates the heart rate. Doppler ultrasound sensors transmit ultrasound waves and receive reflected waves from the movement of the heart valves and myocardium, calculating the heart rate based on the Doppler effect. Both sensors maintain good contact with the animal's body to ensure stable signals, and the collected data is transmitted in real time to a terminal for storage and processing.

[0095] Furthermore, the digestion rate can be comprehensively evaluated by installing pressure sensors or electrogastric sensors in the gastrointestinal tract of livestock and poultry, and combining them with feed intake monitoring. The pressure sensor can detect pressure changes during gastrointestinal peristalsis, and the electrogastric sensor monitors the rhythm and intensity of gastric electrical activity. The changes in these signals are closely related to the digestion process. At the same time, the feed intake of livestock and poultry is recorded by an automatic feeding device, and the amount of feces discharged within a certain period of time is measured at the feces collection point. Using a dedicated signal processing algorithm, the gastrointestinal sensor signal is correlated with the feed intake and feces discharge data to calculate the amount of feed digested per unit time, that is, the digestion rate. For example, by analyzing the frequency and amplitude changes of the electrogastric signal, combined with the feed intake time and feces discharge time, the speed at which food passes through the gastrointestinal tract is determined, thereby obtaining the digestion rate.

[0096] In general, the use of spectral sensors to obtain real-time data on mycotoxin concentrations in feed, as well as synchronous collection of physiological status data such as body temperature, heart rate, and digestion rate of livestock and poultry, solves the problem in the background technology that traditional methods cannot obtain relevant data in real time and comprehensively. Traditional methods may require manual sampling and then testing in the laboratory, which is time-consuming and cannot reflect the real-time situation, resulting in delayed risk prediction. This step realizes real-time data collection and dynamic monitoring, and provides accurate and timely input data for the subsequent metabolic digital twin model, enabling the model to more accurately simulate the metabolic process of mycotoxins in livestock and poultry, thereby improving the efficiency and accuracy of dynamic prediction of livestock and poultry hazard risks caused by feed mycotoxins, and realizing real-time early warning of livestock and poultry health risks.

[0097] In general, real-time data collection on mycotoxin concentrations in feed and livestock physiological status is the foundation for subsequent data input into the metabolic digital twin model. The collected mycotoxin concentration data directly serves as the external input parameter for toxin quantity in the model. Physiological data on livestock, such as body temperature, heart rate, and digestion rate, influence the metabolic rate and hemodynamic parameters of the animal body in the model. For example, the digestion rate affects the absorption rate of mycotoxins, while changes in body temperature affect the activity of toxin-metabolizing enzymes, thereby altering the metabolism and excretion rates of toxins. These data serve as input variables for the model. After computational processing, the model generates a dynamic concentration distribution of mycotoxins within the target organs of the animal. Ultimately, based on this distribution and preset hazard thresholds, a risk level prediction is output.

[0098] S3. Input the mycotoxin concentration data and the physiological status data into the metabolic digital twin model to generate a dynamic concentration distribution of mycotoxins in the target organs of the livestock and poultry.

[0099] In detail, the dynamic concentration distribution of the generated mycotoxins in the target organs of the livestock and poultry includes:

[0100] allocating the initial toxin load of each of the livestock and poultry target organs according to the physiological status data and the mycotoxin concentration data;

[0101] Based on the initial toxin load, the dynamic concentration distribution of mycotoxins in the closed-loop topological network is iteratively solved in combination with the Runge-Kutta method.

[0102] In detail, the solution formula for the dynamic concentration distribution is:

[0103]

[0104] in, is the mycotoxin concentration at the current time step, is the mycotoxin concentration at the next time step, is the initial migration rate calculated based on the current state and the migration rate equation, is the migration rate correction value at half the time step, is the quadratic correction value at half the time step, is the final migration rate at the complete time step, is the time stamp, is the time step.

[0105] In detail, the initial toxin load refers to the initial amount of mycotoxins distributed to each target organ (gastrointestinal tract, liver, and kidney) based on the current physiological state of livestock and poultry and the toxin concentration of feed when the mycotoxin concentration data and physiological status data are input into the metabolic digital twin model. This amount serves as the starting condition for the iterative calculation of the model.

[0106] In detail, the Runge-Kutta method is a numerical method for solving ordinary differential equations. By calculating the slope at multiple time points and taking the weighted average, the numerical solution of the equation is iteratively solved with high precision. In the present invention, it is used to iteratively solve the dynamic concentration distribution of mycotoxins in a closed-loop topological network.

[0107] In detail, the closed-loop topological network is a network structure formed by connecting the gastrointestinal tract, liver and kidney of livestock and poultry through hemodynamic parameters. The migration path of mycotoxins in this network constitutes a closed loop, which can reflect the circulation and metabolic process of toxins among organs.

[0108] Specifically, dynamic concentration distribution refers to the concentration distribution of mycotoxins in target organs of livestock and poultry over time, which is calculated through a model and used to reflect the dynamic process of toxin migration and metabolism in the body.

[0109] In detail, the time step refers to the time interval between two adjacent time points in the iterative solution process, which is represented by the symbol Representation, which is used to discretize continuous time processes for numerical calculations.

[0110] Furthermore, the animal's body temperature, heart rate, and digestion rate are derived from real-time physiological data. Body temperature affects the activity of metabolic enzymes, which in turn alters the rate of toxin metabolism; heart rate reflects blood flow, affecting the migration of toxins between organs; and digestion rate is directly related to toxin absorption efficiency.

[0111] Furthermore, the mycotoxin concentration data in the feed is obtained. This data is detected in real time by a spectral sensor and represents the initial content of toxins in the feed.

[0112] Furthermore, different basic distribution coefficients are set for the gastrointestinal tract, liver, and kidneys based on the physiological functions of the target organs. For example, the gastrointestinal tract, as the primary site of toxin absorption, has a higher basic distribution coefficient; the liver, a key organ for metabolism, has a lower distribution coefficient; and the kidneys, responsible for excretion, have a relatively lower distribution coefficient.

[0113] Furthermore, correction functions related to body temperature, heart rate, and digestion rate were constructed. For example, a higher digestion rate indicates a higher gastrointestinal absorption efficiency, and the corresponding distribution coefficient correction factor is larger. When body temperature deviates from the normal range, the enzyme activity influence coefficient is used to adjust the distribution ratio of each organ.

[0114] Further, the The formula for calculating the initial toxin load in each target organ is as follows:

[0115]

[0116] in: For the Initial toxin load in target organs, is the concentration of mycotoxins in feed, The volume of the target organ can be obtained through livestock and poultry anatomical data or imaging measurements. For example, the volume of an adult pig liver is about 1L. is the basic distribution coefficient for target organs, such as the gastrointestinal tract ,liver ,kidney , is the physiological state correction factor, Correction factors corresponding to body temperature, heart rate, and digestion rate are obtained by fitting experimental data, such as the body temperature correction factor , is body temperature, in units of .

[0117] For example, if the feed toxin concentration , gastrointestinal volume , basic distribution coefficient ,body temperature When the correction factor , correction factors for heart rate and digestion rate , then the initial gastrointestinal load .

[0118] Furthermore, according to the toxin metabolism rate and calculation accuracy requirements, an appropriate time step (such as minutes) to ensure that the endotoxin concentration changes relatively slowly at this time step, which is convenient for numerical calculation.

[0119] Furthermore, the initial toxin load of each target organ Convert to initial concentration , as the initial value of the iterative calculation; obtain the migration rate equation defined in the previous step , where the migration rate constant The migration rate constant from the gastrointestinal tract to the liver has been determined by previous experiments (such as ), external input rate and excretion rate Updated based on real-time data.

[0120] In detail, the iterative calculation process of the Runge-Kutta method is as follows:

[0121] calculate (Initial migration rate): At the current time step , based on the current concentration in each target organ and the migration rate equation to calculate the initial migration rate For example, for the liver ( ), ,in, .

[0122] calculate (half time step correction value): at time At, assuming the concentration is The rate changes half time step, and the temporary concentration is obtained , and then calculate the migration rate correction value of the half time step based on the temporary concentration , calculated in the same way as Similar, but substitute the temporary concentration.

[0123] calculate (Second correction value): Also at time Use The calculated temporary concentration is corrected again to obtain a new temporary concentration , based on which the secondary correction value is calculated , further improving the accuracy of the intermediate point concentration.

[0124] calculate (Full time step rate): at time At, assuming the concentration is After half a time step of rate change, the concentration of the full time step is obtained , the final migration rate was calculated based on this concentration , reflecting the migration rate within a complete time step.

[0125] Furthermore, according to the fourth-order Runge-Kutta formula , take the weighted average of the four slope values and calculate the concentration at the next time step For example, the current liver concentration , calculated , , , , , then the concentration in the next time step is .

[0126] Furthermore, according to the above steps, starting from the initial time step, the concentration value of each time step is calculated in sequence to form a dynamic concentration distribution sequence; the termination condition can be set to reach the preset prediction time range (such as the next 24 hours), or when the concentration change amplitude is less than the set threshold (such as 0.1% / time step), the iteration is stopped, and the concentration distribution is considered to be stable.

[0127] In general, by inputting mycotoxin concentration data and physiological status data into the metabolic digital twin model, and combining it with the Runge-Kutta method to iteratively solve the dynamic concentration distribution, the problem in the background technology that traditional prediction methods cannot take into account the changes in the physiological status of livestock and poultry and the dynamic metabolic process of toxins in real time is solved. Traditional methods mostly use static models or simple kinetic equations, which cannot accurately reflect the impact of physiological indicators such as body temperature and heart rate on toxin metabolism, resulting in a large deviation between the prediction results and the actual situation. This step dynamically adjusts the initial toxin load through real-time input data, and uses high-precision Runge-Kutta method iterative calculations to simulate the migration, metabolism and excretion process of toxins in livestock and poultry in real time, making the prediction results closer to the actual situation, improving the accuracy and timeliness of the dynamic prediction of the risk of livestock and poultry harm caused by feed mycotoxins, and providing a reliable basis for timely prevention and control measures.

[0128] In general, inputting mycotoxin concentration and physiological status data into the metabolic digital twin model is a prerequisite for generating dynamic concentration distributions. The distribution of initial toxin loads is a key link between data input and model calculations. Physiological status data and mycotoxin concentration data determine the initial toxin loads in each target organ. These initial loads, in turn, serve as initial conditions for the iterative solution using the Runge-Kutta method, directly influencing the concentration calculation results at each subsequent time step.

[0129] S4. Based on the dynamic concentration distribution and the preset organ-specific hazard threshold, output the livestock and poultry health risk level prediction result.

[0130] In an embodiment of the present invention, the determination of the preset organ-specific hazard threshold includes:

[0131] Screening the benchmark damage thresholds of the gastrointestinal tract, the liver, and the kidney from the species toxicology database;

[0132] The gastrointestinal threshold was correlated with intestinal villus height, the liver threshold was correlated with ALT activity, and the kidney threshold was correlated with glomerular filtration rate;

[0133] Determining the specific safety factors of the livestock and poultry target organs one by one according to the correlation relationship of the livestock and poultry target organs and the growth stage of the livestock and poultry;

[0134] The baseline damage threshold is adjusted based on the specific safety factor to obtain the organ-specific hazard threshold of the livestock and poultry.

[0135] Specifically, the output of the livestock and poultry health risk level prediction results includes:

[0136] The organ exposure index is generated by dividing the dynamic concentration distribution by the threshold value of the corresponding livestock and poultry target organ, wherein the calculation formula of the organ exposure index is:

[0137]

[0138] in, For the The first time point Organ exposure index of target organs of livestock and poultry, For the The first time point Toxin concentration in target organs of livestock and poultry, For the Preset thresholds for target organs of livestock and poultry;

[0139] Screening the livestock and poultry target organs where the maximum value of the organ exposure index is located as key organs;

[0140] If all of the organ exposure indexes are less than a first preset threshold, outputting a low risk;

[0141] If the organ exposure index of any organ exceeds a second preset threshold, a high risk is output, and the key organs and toxin accumulation pathways causing the high risk are marked.

[0142] In detail, the dynamic concentration distribution refers to the concentration distribution of mycotoxins in target organs such as the gastrointestinal tract, liver, and kidneys of livestock and poultry, which is calculated by the metabolic digital twin model and updated over time. It can reflect the migration and metabolic dynamics of toxins in the body.

[0143] In detail, the preset organ-specific hazard threshold is set for each target organ of livestock and poultry, and is used to determine the critical concentration value of whether the organ poses a health risk due to mycotoxin exposure. The threshold is based on toxicological data and adjusted in combination with the physiological characteristics of livestock and poultry.

[0144] In detail, the species toxicology database is a professional database that stores data on the toxic responses of different species (such as pigs, cattle, poultry, etc.) to various toxins. It contains information such as the damage thresholds of toxins to organs and dose-effect relationships, providing data support for the screening of benchmark damage thresholds in the present invention.

[0145] In detail, the benchmark damage threshold is screened from the species toxicology database, and represents the concentration reference value of the toxin causing initial damage to the target organs of livestock and poultry. It is the basic data for determining the organ-specific hazard threshold.

[0146] In detail, intestinal villus height is an important indicator for measuring the structural integrity of the gastrointestinal mucosa and is closely related to the absorption function of the gastrointestinal tract. In the present invention, it is used to establish a correlation with the baseline damage threshold of the gastrointestinal tract to reflect the degree of damage to the gastrointestinal tract caused by mycotoxins; ALT activity, namely alanine aminotransferase activity, is an enzyme present in liver cells. When the liver is damaged, ALT will be released into the blood. Its activity level can be used as an indicator for evaluating liver function and the degree of damage, and is used to correlate with the baseline damage threshold of the liver; glomerular filtration rate is an important indicator for measuring the excretion function of the kidneys, which indicates the amount of blood filtered by the kidneys per unit time. It is related to the kidneys' ability to excrete toxins and is used to correlate with the baseline damage threshold of the kidneys.

[0147] In detail, the specific safety factor is a correction coefficient determined based on the correlation between the target organs of livestock and poultry and the growth stage of livestock and poultry (such as young, growing, and adult). It is used to adjust the baseline damage threshold to obtain an organ-specific hazard threshold that is more in line with actual conditions.

[0148] In detail, the organ exposure index is the ratio obtained by dividing the mycotoxin concentration in the target organ at a certain point in time by the preset threshold value of the organ, which is used to quantitatively assess the degree of mycotoxin exposure of the target organ; the key organ refers to the organ corresponding to the maximum exposure index in the organ exposure index of each target organ, and this organ is the part with the highest current risk of mycotoxin hazard.

[0149] In detail, the first preset threshold is one of the critical values used to judge the health risk level of livestock and poultry. When the exposure index of all organs is less than the threshold, it is judged to be a low-risk state; the second preset threshold is another critical value used to judge the health risk level of livestock and poultry. When the exposure index of any organ exceeds the threshold, it is judged to be a high-risk state.

[0150] In detail, the toxin accumulation pathway refers to the migration and accumulation trajectory of mycotoxins among target organs, starting from entering the body of livestock and poultry, being absorbed through the gastrointestinal tract, transported through the blood circulation to the liver for metabolism, and then excreted through the kidneys.

[0151] Furthermore, the species toxicology database is accessed to search for benchmark damage thresholds for mycotoxins in the gastrointestinal tract, liver, and kidneys, based on the livestock species (e.g., pigs). For example, the database reveals that the benchmark damage threshold for a certain type of mycotoxin in pig liver is X ppm. This threshold, determined through animal toxicity studies, indicates that significant tissue damage or functional abnormalities begin to occur when the toxin concentration in the liver reaches X ppm.

[0152] Furthermore, multiple sets of similar data in the database are integrated and verified to eliminate outliers and ensure the reliability of the baseline injury threshold. For example, if there are multiple studies reporting baseline injury thresholds for liver disease, their average is calculated and combined with the confidence interval to determine the final baseline value used in subsequent steps.

[0153] Furthermore, the gastrointestinal threshold is associated with the height of intestinal villi: the data on the change of intestinal villi height in the gastrointestinal tract of livestock and poultry under different mycotoxin concentrations are collected, and a mathematical model is established through regression analysis, such as establishing a functional relationship ,in, is the gastrointestinal hazard threshold, is the height of intestinal villi. When the height of intestinal villi drops to a certain critical value, the corresponding toxin concentration becomes the baseline damage threshold of the gastrointestinal tract. This model realizes the dynamic correlation between the threshold and the height of intestinal villi.

[0154] Furthermore, animal experiments were conducted to measure changes in ALT activity in the blood at different liver toxin concentrations, and a correlation curve was constructed between ALT activity and the liver toxin threshold. For example, when ALT activity exceeds Y% of the normal reference value, the corresponding liver toxin concentration is the baseline liver damage threshold. The liver damage threshold can then be adjusted by monitoring ALT activity.

[0155] Furthermore, the relationship between the decrease in glomerular filtration rate and the concentration of renal toxins was determined through renal perfusion experiments and glomerular filtration rate tests, and the relationship between the decrease in glomerular filtration rate and the concentration of renal toxins was established. The correlation formula of is the kidney damage threshold, is the glomerular filtration rate, is an experimentally determined coefficient relating the renal threshold to the glomerular filtration rate.

[0156] Furthermore, considering the physiological connections between the gastrointestinal tract, liver, and kidneys, for example, how the liver's ability to metabolize toxins affects the kidneys' excretion burden, a weighted matrix of inter-organ associations was established. Using expert knowledge and historical data, the weight of the gastrointestinal tract's impact on the liver was determined to be a, the weight of the liver's impact on the kidneys to be b, and so on, forming a network of associations.

[0157] Furthermore, the growth coefficient of each stage is determined according to the growth stage of livestock and poultry (such as young stage, growing stage, and adult stage). The metabolic system of young livestock and poultry is not yet fully developed and has a low tolerance to toxins. The growth coefficient is set as ; Adult livestock and poultry have strong tolerance, and the growth coefficient is set as ( ).

[0158] Furthermore, for each target organ, the specific safety factor ,in, is the weight of the organ in the association network (calculated by analytic hierarchy process, such as the weight of gastrointestinal tract ,liver ,kidney ), is the growth factor corresponding to the growth stage. For example, the liver-specific safety factor for the young stage is =0.35 .

[0159] Furthermore, using the formula The baseline damage threshold is adjusted, where is the adjusted organ-specific hazard threshold, is the baseline damage threshold of the organ, For example, if the baseline damage threshold of the liver is Xppm and the specific safety factor of the young stage is 0.8, then the adjusted liver threshold This threshold is more in line with the actual tolerance of the liver of young livestock and poultry to toxins.

[0160] In detail, the calculation results of the metabolic digital twin model are used to obtain the (like , the time interval is ) Target organs Mycotoxin concentrations (gastrointestinal tract, liver, kidneys) .

[0161] Furthermore, the preset organ-specific hazard thresholds are used to calculate the organ exposure index. For example, if the toxin concentration in the liver is 150 ppb at a certain point in time, and the preset threshold for the liver is 100 ppb, then the liver exposure index at that point in time is 1.5.

[0162] Furthermore, the exposure index of each target organ at the same time point is compared, the maximum value is found, and the target organ corresponding to the maximum value is determined as the key organ at that time point. For example, if the gastrointestinal exposure index at a certain time point is 1.2, the liver is 1.5, and the kidney is 0.8, then the key organ is the liver.

[0163] Specifically, the exposure indexes for all time points and all target organs are examined. If all exposure indices are less than the first preset threshold (e.g., 0.8), the livestock and poultry are judged to be in a low-risk state, and a "low-risk" prediction result is output. If the exposure index of any organ at a certain time point exceeds the second preset threshold (e.g., 1.2), the animal is judged to be in a high-risk state. At this point, the key organs at that time point are marked, and by backtracking the migration path of the toxins in the metabolic digital twin model, the specific path where the toxin concentration accumulates beyond the threshold during absorption from the gastrointestinal tract, metabolism through the liver, and excretion through the kidneys is determined, such as the accumulation path of "gastrointestinal tract → liver", and the result of "high risk, key organ: liver, toxin accumulation path: gastrointestinal tract → liver" is output.

[0164] In general, by outputting risk level prediction results based on dynamic concentration distribution and preset organ-specific hazard thresholds, the problem of inaccurate prediction results caused by the traditional risk prediction methods in the background technology due to the use of fixed thresholds and the failure to consider organ specificity and differences in livestock and poultry growth stages is solved. Traditional methods usually use a unified toxicity threshold, which cannot adapt to the differences in sensitivity of different organs to toxins and the physiological changes of livestock and poultry at different growth stages, and can easily lead to misjudgment. This step determines the specific hazard threshold by combining species toxicology data, organ physiological indicators and growth stage characteristics, and calculates the exposure index based on the dynamic concentration distribution. It can dynamically and accurately assess the hazard risk of mycotoxins to livestock and poultry, improve the accuracy and reliability of the prediction, provide a scientific decision-making basis for toxin prevention and control in the livestock and poultry breeding process, and achieve accurate early warning of livestock and poultry health risks.

[0165] like Figure 2 FIG. 1 is a functional module diagram of a system for dynamically predicting the risk of livestock and poultry damage caused by mycotoxins in feed provided by an embodiment of the present invention.

[0166] The system 100 for dynamically predicting the risk of livestock and poultry hazard posed by feed mycotoxins, as described herein, can be installed in an electronic device. Depending on the functionality implemented, the system 100 can include a twin model construction module 101, a data acquisition module 102, a concentration distribution generation module 103, and a risk level prediction module 104. A module, also referred to as a unit, is a series of computer program segments that can be executed by an electronic device processor and perform a fixed function, and is stored in the electronic device's memory.

[0167] In this embodiment, the functions of each module / unit are as follows:

[0168] The twin model construction module 101 is used to construct a metabolic digital twin model of feed mycotoxins in target organs of livestock and poultry. The metabolic digital twin model is based on the kinetic mechanism of toxin absorption, distribution, metabolism and excretion coupled with multiple organs;

[0169] The data acquisition module 102 is used to collect the mycotoxin concentration data in the feed and the physiological status data of the livestock and poultry in real time;

[0170] The concentration distribution generating module 103 is configured to input the mycotoxin concentration data and the physiological status data into the metabolic digital twin model to generate a dynamic concentration distribution of mycotoxins in the target organs of the livestock and poultry;

[0171] The risk level prediction module 104 is used to output a livestock and poultry health risk level prediction result based on the dynamic concentration distribution and a preset organ-specific hazard threshold.

[0172] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the module division is merely a logical function division, and other division methods may be used in actual implementation.

[0173] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0174] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0175] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0176] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for dynamically predicting the risk of livestock and poultry damage caused by mycotoxins in feed, characterized in that: The method comprises: Construct a digital twin model of the metabolism of feed mycotoxins in target organs of livestock and poultry. The digital twin model is based on the kinetic mechanisms of toxin absorption, distribution, metabolism, and excretion coupled with multiple organs. Real-time collection of mycotoxin concentration data in feed and physiological status data of livestock and poultry; Inputting the mycotoxin concentration data and the physiological status data into the metabolic digital twin model to generate a dynamic concentration distribution of mycotoxins in the target organs of the livestock and poultry; Based on the dynamic concentration distribution and the preset organ-specific hazard threshold, a livestock and poultry health risk level prediction result is output.

2. The method for dynamically predicting the risk of livestock and poultry damage caused by mycotoxins in feed according to claim 1, characterized in that: The construction of a digital twin model of the metabolism of feed mycotoxins in target organs of livestock and poultry includes: The gastrointestinal tract, liver and kidney of livestock and poultry are connected through hemodynamic parameters to obtain a closed-loop topological network corresponding to the livestock and poultry target organs; The migration rate equation of the mycotoxin in the target organs of livestock and poultry is established based on the law of conservation of mass, and the bioavailability of the dynamic concentration distribution is verified by virtual mycotoxin injection.

3. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 2, characterized in that: The verifying the biological effectiveness of the dynamic concentration distribution by virtual mycotoxin injection includes: Setting a pulsed mycotoxin input at the inlet of the closed-loop topology network; Calculate the mycotoxin deviation between the total amount of mycotoxins at the outlet of all livestock and poultry target organs and the pulse mycotoxin input amount; If the mycotoxin deviation exceeds a preset tolerance value, reversely optimizing the hemodynamic parameters; When the mycotoxin deviation of three consecutive virtual injections is lower than the preset tolerance value, a biovalidity certification of the dynamic concentration distribution is issued.

4. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 3, characterized in that: The migration rate equation is: ； in, It is The amount of mycotoxins in the target organs of livestock and poultry, From the livestock and poultry target organs To the livestock target organs The migration rate constant, From the livestock and poultry target organs To the livestock target organs The migration rate constant, It is The amount of mycotoxins in the target organs of livestock and poultry, is an external input to the livestock target organ The mycotoxin rate, The livestock and poultry target organs The excretion rate, is a time stamp, Indicates the The target organs of livestock and poultry, Indicates the The target organs of livestock and poultry.

5. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 1, characterized in that: The real-time collection of mycotoxin concentration data in feed and physiological status data of livestock and poultry includes: The mycotoxin concentration data in the feed is obtained through spectral sensors, and the body temperature, heart rate and digestion rate of livestock and poultry are simultaneously collected as the physiological status data of livestock and poultry.

6. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 4, characterized in that: The dynamic concentration distribution of the generated mycotoxins in the target organs of the livestock and poultry includes: allocating the initial toxin load of each of the livestock and poultry target organs according to the physiological status data and the mycotoxin concentration data; Based on the initial toxin load, the dynamic concentration distribution of mycotoxins in the closed-loop topological network is iteratively solved in combination with the Runge-Kutta method.

7. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 2, characterized in that: Determination of the preset organ-specific hazard threshold includes: Screening the benchmark damage thresholds of the gastrointestinal tract, the liver, and the kidney from the species toxicology database; The gastrointestinal threshold was correlated with intestinal villus height, the liver threshold was correlated with ALT activity, and the kidney threshold was correlated with glomerular filtration rate; Determining the specific safety factors of the livestock and poultry target organs one by one according to the correlation relationship of the livestock and poultry target organs and the growth stage of the livestock and poultry; The baseline damage threshold is adjusted based on the specific safety factor to obtain the organ-specific hazard threshold of the livestock and poultry.

8. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 6, characterized in that: The output livestock and poultry health risk level prediction results include: The organ exposure index is generated by dividing the dynamic concentration distribution by the threshold value of the corresponding livestock and poultry target organ, wherein the calculation formula of the organ exposure index is: ； in, For the The first time point Organ exposure index of target organs of livestock and poultry, For the The first time point Toxin concentration in target organs of livestock and poultry, For the Preset thresholds for target organs of livestock and poultry; Screening the livestock and poultry target organs where the maximum value of the organ exposure index is located as key organs; If all of the organ exposure indexes are less than a first preset threshold, outputting a low risk; If the organ exposure index of any organ exceeds a second preset threshold, a high risk is output, and the key organs and toxin accumulation pathways causing the high risk are marked.

9. The method for dynamically predicting the risk of livestock and poultry damage caused by feed mycotoxins according to claim 6, characterized in that: The solution formula for the dynamic concentration distribution is: ； in, is the mycotoxin concentration at the current time step, is the mycotoxin concentration at the next time step, is the initial migration rate calculated based on the current state and the migration rate equation, is the migration rate correction value at half the time step, is the quadratic correction value at half the time step, is the final migration rate at the complete time step, is the time stamp, is the time step.

10. A dynamic prediction system for the risk of livestock and poultry damage caused by mycotoxins in feed, characterized in that: The system comprises: A twin model construction module is used to construct a digital twin model of the metabolism of feed mycotoxins in the target organs of livestock and poultry. The metabolic digital twin model is based on the kinetic mechanism of toxin absorption, distribution, metabolism and excretion coupled with multiple organs; Data acquisition module, used to collect real-time data on mycotoxin concentration in feed and physiological status of livestock and poultry; a concentration distribution generation module, configured to input the mycotoxin concentration data and the physiological status data into the metabolic digital twin model to generate a dynamic concentration distribution of mycotoxins in the target organs of the livestock and poultry; The risk level prediction module is used to output the livestock and poultry health risk level prediction result based on the dynamic concentration distribution and the preset organ-specific hazard threshold.

Images