Veterinary drug parameter regulation and control optimization method based on artificial intelligence integrated service

Abstract

The invention relates to the technical field of animal breeding intelligent management, and particularly discloses an artificial intelligence integrated service-based veterinary drug parameter regulation and control optimization method, which comprises the following steps of: acquiring drug concentration data, drug delivery frequency data, body temperature, heart rate and other physiological data in an animal body in real time in an animal drug delivery monitoring period; carrying out normalization processing and wavelet decomposition on the collected drug concentration data, calculating a drug concentration abnormal index, and evaluating a drug concentration fluctuation abnormal degree; performing frequency domain analysis on the drug delivery frequency data, and calculating a delivery frequency stability index for evaluating drug delivery stability; comprehensively analyzing the drug concentration anomaly index and the dosing frequency stability index through a Bayesian network, and calculating a comprehensive anomaly index to identify the anomaly in the dosing process; in combination with animal physiological data, a correlation model of drug concentration, dosing frequency and animal physiological response is constructed by using a maximum likelihood estimation method, drug dosing parameters are dynamically regulated and controlled, and personalized precise treatment is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management of animal breeding, and particularly relates to a method for regulating and optimizing veterinary drug parameters based on artificial intelligence integrated services. Background Art

[0002] With the rapid development of modern animal husbandry, the health management and drug treatment of animals have become important links in ensuring breeding efficiency and food safety. In the prevention and control of animal diseases, the scientific administration of veterinary drugs is crucial, which not only directly affects the treatment effect, but also has a profound impact on the health status of animals and breeding efficiency. Traditional veterinary drug management usually adopts a fixed-dose or empirical administration mode. However, factors such as individual differences in animals, disease complexity, and environmental changes often make it difficult for the fixed drug administration mode to meet the actual needs. At the same time, the rapid development of artificial intelligence technology and intelligent devices in recent years has provided new technical means for precise veterinary drug administration and personalized treatment. Dynamic regulation of the drug administration process through data collection and analysis has become an important research direction in the breeding industry.

[0003] The existing technologies have the following deficiencies:

[0004] The existing technologies lack the ability to monitor the physiological data of animals and drug concentrations in real time, and cannot comprehensively grasp the dynamic changes of drugs in animals. Secondly, traditional technologies cannot effectively evaluate the fluctuations in the drug administration process, such as abnormal fluctuations in drug concentrations and instability in drug administration frequencies, which may lead to a significant decline in the drug treatment effect or even cause adverse reactions. In addition, the existing technologies lack the ability to analyze and model the correlation between drug concentrations, drug administration frequencies, and animal physiological responses, and cannot make real-time adjustments according to the health status of animals, resulting in the difficulty of realizing personalization and precision in drug administration plans. The above deficiencies not only reduce the drug treatment effect, but also increase drug waste and animal health risks. Therefore, how to achieve precise regulation and dynamic optimization of veterinary drug administration through real-time monitoring and intelligent analysis is the core problem that the existing technologies urgently need to solve. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for regulating and optimizing veterinary drug parameters based on artificial intelligence integrated services to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A method for regulating and optimizing veterinary drug parameters based on artificial intelligence integrated services includes the following steps:

[0008] S1: During an animal drug administration monitoring period, collect in real time the drug concentration data, drug administration frequency data in the animal body, and the physiological data of the animal, where the physiological data of the animal includes: animal body temperature and heart rate;

[0009] S2: Analyze the drug concentration data in the animal body collected, and calculate the drug concentration anomaly index according to the fluctuation degree of the drug concentration data within the monitoring period, which is used to evaluate the anomaly degree of the drug concentration fluctuation;

[0010] S3: Analyze the drug delivery frequency data collected, and calculate the delivery frequency stability index according to the fluctuation amplitude of the drug delivery frequency within the monitoring period, which is used to evaluate the stability of the drug delivery;

[0011] S4: Conduct a comprehensive analysis of the fluctuation degree of the drug concentration data and the stability of the drug delivery within the monitoring period to identify whether there is an anomaly in the drug delivery;

[0012] S5: Combine the physiological data of the animal and the drug response results to construct an association model among the drug concentration, the drug delivery frequency and the animal physiological response, analyze the influence degree of the abnormal drug delivery on the animal physiology, and conduct dynamic regulation of the drug concentration and the drug delivery frequency according to the analysis results of the influence degree.

[0013] As a further solution of the present invention: The process of obtaining the drug concentration anomaly index is as follows:

[0014] Within the monitoring period, collect the drug concentration data in real time according to the time series. For the collected drug concentration time series data in the animal body, perform normalization processing, and the calculation expression is:

[0015]

[0016] In the formula, i represents the number of drug concentration data, i is a positive integer greater than 0, x i is the i-th real-time drug concentration data, x' i represents the i-th drug concentration data after normalization processing, max(X) represents the maximum drug concentration data in the drug concentration time series data, and min(X) represents the minimum drug concentration data in the drug concentration time series data;

[0017] Use discrete wavelet transform to perform multi-scale decomposition on the normalized drug concentration time series, and the decomposition formula is:

[0018]

[0019] In the formula, J represents the number of wavelet decomposition scales, j represents the j-th scale of wavelet decomposition, D j (t) represents the high-frequency component of the j-th layer, A J (t) represents the low-frequency component of the J-th layer, t represents the acquisition time point, and t is a positive integer greater than 0;

[0020] Calculate the high-frequency component D of each layerj (t) The fluctuating energy, the calculation expression is:

[0021]

[0022] In the formula, n represents the total number of acquisition time points, E j represents the fluctuating energy of the high-frequency component of the j-th layer;

[0023] According to the fluctuating energy E of the high-frequency components of each layer j , combined with a preset proportionality coefficient, calculate the drug concentration anomaly index, and the calculation expression is:

[0024]

[0025] In the formula, I f represents the drug concentration anomaly index, and α j represents the preset proportionality coefficient.

[0026] As a further solution of the present invention: The specific process of evaluating the abnormal degree of drug concentration fluctuation includes:

[0027] Judge whether the drug concentration anomaly index is greater than or equal to a preset threshold. If so, it indicates the abnormality of the corresponding drug concentration fluctuation. If not, it indicates the normality of the corresponding drug concentration fluctuation.

[0028] As a further solution of the present invention: The process of obtaining the drug delivery frequency stability index is as follows:

[0029] During the monitoring period, the drug delivery frequency is collected in real time according to the time series, and the drug delivery frequency data is normalized to obtain the normalized drug delivery frequency data. The calculation expression is:

[0030]

[0031] In the formula, a represents the number of drug delivery frequency data, a is a positive integer greater than 0, and f a represents the a-th real-time drug delivery frequency, and f′ a represents the a-th normalized drug delivery frequency, max(f) represents the maximum drug delivery frequency of the drug delivery frequency data, and min(f) represents the minimum drug delivery frequency of the drug delivery frequency data;

[0032] Perform Fourier transform on the normalized drug delivery frequency data to obtain the frequency domain representation. The calculation expression is:

[0033]

[0034] Wherein, F(ω) represents the Fourier transform result of the drug delivery frequency data, t represents the acquisition time point, t is a positive integer greater than 0, n represents the total number of acquisition time points, ω represents the frequency component, and z represents the imaginary unit;

[0035] Calculate the amplitude spectrum of the Fourier transform result, and the calculation expression is:

[0036]

[0037] Wherein, A(ω) is the amplitude of the corresponding frequency component, Re(F(ω)) represents the real part of F(ω), and Im(F(ω)) represents the imaginary part of F(ω);

[0038] Calculate the total energy in the frequency domain, and the calculation expression is:

[0039] E total = ∑ ω A(ω) 2 ;

[0040] Wherein, E total represents the total energy in the frequency domain, and A(ω) 2 is the energy of each frequency component;

[0041] Calculate the drug delivery frequency stability index, and the calculation expression is:

[0042]

[0043] Wherein, P s represents the drug delivery frequency stability index, and ω ∈ Ω represents the low-frequency fluctuation frequency band.

[0044] As a further solution of the present invention: The evaluation of the stability of drug delivery specifically includes:

[0045] Judge whether the drug delivery frequency stability index is greater than or equal to a preset threshold. If so, it means that the corresponding drug delivery is stable. If not, it means that the corresponding drug delivery is unstable.

[0046] As a further solution of the present invention: The comprehensive analysis of the fluctuation degree of drug concentration data within the monitoring period and the stability of drug delivery specifically includes:

[0047] Obtain the drug concentration anomaly index I f and the drug delivery frequency stability index P s ;

[0048] Process the drug concentration anomaly index I f and the drug delivery frequency stability index P s within the monitoring period through a Bayesian network, and calculate the expected value of the comprehensive anomaly index IP, specifically including:

[0049] Set the conditional probability relationship of the Bayesian network as: P(IP|I f ,P s );

[0050] Among them, P(IP|I f ,P s ) represents the conditional probability distribution of the comprehensive anomaly index IP;

[0051] Through the Bayesian formula, calculate the joint probability distribution, and the calculation expression is:

[0052] P(IP|I f ,P s ) = P(IP|I f ,P s ) · P(I f ) · P(P s );

[0053] Among them, P(I f ) represents the prior probability distribution of the drug concentration anomaly index, and P(P s ) represents the prior probability distribution of the dosing frequency stability index;

[0054] Calculate the expected value of the comprehensive anomaly index, and the calculation expression is:

[0055]

[0056] Among them, represents the expected value of the comprehensive anomaly index;

[0057] Judge whether the expected value of the comprehensive anomaly index is greater than or equal to the preset threshold. If so, it means that there is an anomaly in the dosing of the corresponding drug. If not, it means that there is no anomaly in the dosing of the corresponding drug.

[0058] As a further solution of the present invention: Combine the physiological data of animals and the drug response results to construct an association model between drug concentration, dosing frequency and animal physiological response, and analyze the impact of abnormal drug dosing on animal physiology, specifically including:

[0059] Collect the physiological data of animals during the monitoring period, including the body temperature and heart rate of animals during the monitoring period;

[0060] Obtain the expected value of the comprehensive anomaly index of drug dosing during the monitoring period;

[0061] Use the body temperature and heart rate of animals during the monitoring period and the expected value of the comprehensive anomaly index of drug dosing during the monitoring period as the input of the association model to construct the association model, and the calculation expression is:

[0062]

[0063] Among them, \(t\) represents the time point of animal physiological data collection, and \(Y(t)\) represents the physiological data. represents the mean function. represents the kernel function.

[0064] Among them, the mean function

[0065] In the formula, \(\beta\) 0 represents the baseline value of the physiological response, and \(\beta\) 1 represents the influence coefficient of the comprehensive anomaly index on the physiological response. represents the expected value of the comprehensive anomaly index.

[0066] Among them, the kernel function

[0067] In the formula, \(\sigma\) 2 represents the amplitude parameter of the kernel function, and \(l\) represents the length scale parameter of the kernel function.

[0068] The model parameters \(\beta\) 0 , \(\beta\) 1 , \(\sigma\) 2 and \(l\) are fitted by the maximum likelihood estimation method, and the predicted value of the mean of the physiological response is calculated according to the fitting result. The calculation expression is:

[0069]

[0070] In the formula, represents the mean of the predicted physiological response. represents the new data point and the training data The covariance vector between them, \(K\) represents the covariance matrix of the training data, and \(Y\) represents the physiological data.

[0071] The prediction variance is calculated. The calculation expression is:

[0072]

[0073] In the formula, represents the prediction variance. represents the new data point The autocovariance of.

[0074] The influence coefficient of drug delivery anomaly is calculated. The calculation expression is:

[0075]

[0076] In the formula, \(I\) impact represents the influence coefficient of drug delivery anomaly.

[0077] Determine whether the influence coefficient of abnormal drug delivery is greater than or equal to a preset threshold. If so, it indicates that the abnormal drug delivery has an impact on the physiological state of the animal. If not, it indicates that the abnormal drug delivery has no impact on the physiological state of the animal.

[0078] As a further solution of the present invention: according to the analysis result of the influence degree, perform dynamic regulation on the drug concentration and dosing frequency, specifically including:

[0079] According to the analysis result of the influence degree, perform intelligent dynamic regulation on the drug concentration and dosing frequency, including: the system collects the drug concentration, dosing frequency and physiological data in the animal body in real time, combines the drug concentration abnormality index and dosing frequency stability index, analyzes the influence degree of drug delivery on the physiological state of the animal, constructs an optimized correlation model, and according to the relationship between the comprehensive abnormality index and the physiological response, the system predicts the optimal regulation range of drug delivery, and determines the adjustment direction and amplitude of the drug concentration and dosing frequency;

[0080] Adaptive adjustment of the drug concentration and dosing frequency according to the real-time data feedback. After each monitoring cycle, the system optimizes the dosing frequency and drug concentration through the reward function according to the animal physiological response and the historical data of drug delivery, to ensure the stability of drug concentration fluctuation and dosing frequency.

[0081] The beneficial effects of the present invention:

[0082] (1) Through intelligent data collection and analysis technologies, the present invention breaks through the limitations of traditional veterinary drug delivery monitoring. By utilizing advanced sensors and real-time data processing technologies, it comprehensively improves the accuracy and controllability of veterinary drug management. Through the combination of various sensing devices such as embedded biosensors, intelligent drug delivery pumps, and electrocardiogram sensors, it can collect real-time data on drug concentration, drug delivery frequency, and physiological status of animals within the monitoring period. These data are transmitted to the central processing system, and through multi-dimensional data analysis, including drug concentration anomaly index and drug delivery frequency stability index, a comprehensive monitoring of the drug delivery process is formed. This monitoring method can not only capture any fluctuations in drug concentration in real time but also detect the stability of drug delivery frequency. By evaluating drug fluctuations through a mathematical model, abnormal situations in the drug delivery process can be detected in a timely manner, thus providing strong support for precise regulation. By analyzing drug concentration fluctuations and drug delivery frequency stability in real time, the system of the present invention can timely identify unstable factors in drug delivery, thereby avoiding adverse reactions caused by too high or too low drug concentration. In this process, the stability of the drug is significantly enhanced, effectively reducing the animal health risks caused by drug fluctuations and improving the reliability and safety of veterinary drug treatment. In addition, the combination of real-time feedback and optimization algorithms can ensure that the physiological status of animals is always within a healthy range. This can not only significantly improve the accuracy of animal treatment but also provide a more efficient drug management method for veterinarians and the breeding industry, facilitating the realization of intelligent breeding and precise treatment;

[0083] (2) The present invention combines the physiological data of animals with the drug response results, and uses advanced artificial intelligence technology to construct an accurate correlation model between drug concentration, dosing frequency and animal physiological responses. By collecting physiological indicators such as animal body temperature and heart rate in real time, and combining with historical drug delivery data, the system can analyze the actual impact of drug delivery on the physiological state of animals in real time. This process not only relies on traditional biomedical indicators, but also dynamically optimizes the model through the maximum likelihood estimation method to continuously improve the prediction accuracy. With the help of this model, the system can predict the changing trend of the physiological responses of animals to drug delivery in real time during each monitoring cycle, accurately evaluate the adjustment of drug concentration and dosing frequency, so as to ensure the personalization and precision of the drug delivery process. Through this dynamic regulation, the present invention can accurately adjust drug delivery among different animals, and automatically adjust the drug concentration and dosing frequency according to the health status, in-vivo drug concentration of each animal and the real-time changes of its physiological responses. The system comprehensively analyzes the continuously collected physiological data such as body temperature and heart rate and the historical records of drug delivery to customize a drug treatment plan for each animal. This precise dynamic regulation not only significantly improves the personalization and pertinence of veterinary drug treatment, but also reduces the risks of side effects and adverse reactions caused by drug overdose or insufficiency, and effectively improves the health level of animals. At the same time, the real-time adaptive adjustment ability of the system makes drug delivery more flexible, and can quickly adjust according to the physiological responses of animals, minimizing the uncertainty in treatment. This innovative technical solution not only promotes the development of the animal breeding industry towards a more intelligent, personalized and precise direction, but also provides strong support for the sustainable development of the industry, helping to improve breeding efficiency and animal health management level. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] The present invention will be further described below with reference to the accompanying drawings.

[0085] Figure 1 is a specific step flow block diagram of a method for optimizing veterinary drug parameters based on artificial intelligence integrated services according to the present invention;

[0086] Figure 2 is a calculation flow block diagram of the comprehensive anomaly index in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0087] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0088] Please refer to Figure 1As shown in the figure, the present invention is a method for regulating and optimizing veterinary drug parameters based on artificial intelligence integrated services, including the following steps:

[0089] S1: During an animal drug administration monitoring period, real-time collect data on the drug concentration in the animal's body, the drug administration frequency data, and the physiological data of the animal. The animal physiological data includes: animal body temperature and heart rate;

[0090] S2: Analyze the collected data on the drug concentration in the animal's body, and calculate the drug concentration anomaly index according to the fluctuation degree of the drug concentration data during the monitoring period, which is used to evaluate the anomaly degree of the drug concentration fluctuation;

[0091] S3: Analyze the collected drug administration frequency data, and calculate the administration frequency stability index according to the fluctuation range of the drug administration frequency during the monitoring period, which is used to evaluate the stability of the drug administration;

[0092] S4: Conduct a comprehensive analysis of the fluctuation degree of the drug concentration data during the monitoring period and the stability of the drug administration to identify whether there is an abnormality in the drug administration;

[0093] S5: Combine the physiological data of the animal and the drug response results to construct an association model between the drug concentration, the drug administration frequency, and the animal physiological response, analyze the impact degree of the abnormal drug administration on the animal physiology, and perform dynamic regulation on the drug concentration and the drug administration frequency according to the analysis results of the impact degree.

[0094] In S1, during an animal drug administration monitoring period, real-time collect data on the drug concentration in the animal's body, the drug administration frequency data, and the physiological data of the animal. The animal physiological data includes: animal body temperature and heart rate, specifically including:

[0095] Use a biosensor to monitor the drug concentration in the animal's body, and collect the drug concentration in the animal's body in real time according to the time series. The biosensor can be directly implanted into the animal's body;

[0096] Adopt a wireless transmission system to transmit the collected data to the central data processing system in real time. The sensor will collect drug concentration data at fixed time intervals and upload the data to the cloud through the wireless network.

[0097] Use an intelligent drug infusion pump to record the time, frequency, and dose of each drug administration. By precisely controlling the drug injection rate and frequency, ensure the stability of the drug administration process.

[0098] Real-time monitor the animal body temperature through an implantable sensor and transmit the data to the data processing system;

[0099] Real-time heart rate monitoring is performed through an electrocardiogram sensor. The electrocardiogram sensor can be directly attached to the animal's body surface to collect the animal's heart rate data in real time according to the time series.

[0100] In S2, the collected data of the drug concentration in the animal's body is analyzed, and according to the degree of fluctuation of the drug concentration data within the monitoring period, the drug concentration anomaly index is calculated to evaluate the anomaly degree of the drug concentration fluctuation, specifically including:

[0101] The process of obtaining the drug concentration anomaly index is as follows:

[0102] Within the monitoring period, the drug concentration data is collected in real time according to the time series. For the collected time series data of the drug concentration in the animal's body, normalization processing is performed, and the calculation expression is:

[0103]

[0104] In the formula, i represents the number of drug concentration data, i is a positive integer greater than 0, x i is the i-th real-time drug concentration data, x' i represents the i-th drug concentration data after normalization processing, max(X) represents the maximum drug concentration data in the time series data of the drug concentration, and min(X) represents the minimum drug concentration data in the time series data of the drug concentration;

[0105] The normalized time series of the drug concentration is decomposed by multi-scale using discrete wavelet transform, and the decomposition formula is:

[0106]

[0107] In the formula, J represents the number of scales of wavelet decomposition, j represents the j-th scale of wavelet decomposition, D j (t) represents the high-frequency component of the j-th layer, A J (t) represents the low-frequency component of the J-th layer, t represents the acquisition time point, and t is a positive integer greater than 0;

[0108] Calculate the fluctuation energy of each layer of high-frequency component D j (t), and the calculation expression is:

[0109]

[0110] In the formula, n represents the total number of acquisition time points, and E j represents the fluctuation energy of the high-frequency component of the j-th layer;

[0111] According to the fluctuation energy E j of each layer of high-frequency component, combined with a preset proportional coefficient, calculate the drug concentration anomaly index, and the calculation expression is:

[0112]

[0113] In the formula, I f represents the drug concentration abnormality index, and α j represents the preset proportionality coefficient;

[0114] Judge whether the drug concentration abnormality index is greater than or equal to the preset threshold. If so, it indicates the abnormality of the corresponding drug concentration fluctuation. If not, it indicates the normality of the corresponding drug concentration fluctuation.

[0115] It should be noted that: the drug concentration abnormality index reflects the situation of drug concentration fluctuation, and when the value of the drug concentration abnormality index is larger, the corresponding drug concentration fluctuation is more abnormal.

[0116] In S3, analyze the collected drug delivery frequency data, and calculate the delivery frequency stability index according to the fluctuation range of the drug delivery frequency within the monitoring period to evaluate the stability of drug delivery, specifically including:

[0117] The acquisition process of the delivery frequency stability index is as follows:

[0118] During the monitoring period, collect the drug delivery frequency in real time according to the time series, normalize the drug delivery frequency data to obtain the normalized drug delivery frequency data, and the calculation expression is:

[0119]

[0120] In the formula, a represents the number of drug delivery frequency data, a is a positive integer greater than 0, and f a represents the a-th real-time drug delivery frequency, and f′ a represents the a-th normalized drug delivery frequency, max(f) represents the maximum drug delivery frequency of the drug delivery frequency data, and min(f) represents the minimum drug delivery frequency of the drug delivery frequency data;

[0121] Perform Fourier transform on the normalized drug delivery frequency data to obtain the frequency domain representation, and the calculation expression is:

[0122]

[0123] In the formula, F(ω) represents the Fourier transform result of the drug delivery frequency data, t represents the acquisition time point, t is a positive integer greater than 0, n represents the total number of acquisition time points, ω represents the frequency component, and z represents the imaginary unit;

[0124] Calculate the amplitude spectrum of the Fourier transform result, and the calculation expression is:

[0125]

[0126] Where, A(ω) is the amplitude of the corresponding frequency component, Re(F(ω)) represents the real part of F(ω), and Im(F(ω)) represents the imaginary part of F(ω);

[0127] Calculate the total energy in the frequency domain, and the calculation expression is:

[0128] E total = ∑ ω A(ω) 2 ;

[0129] Where, E total represents the total energy in the frequency domain, and A(ω) 2 is the energy of each frequency component;

[0130] Calculate the drug delivery frequency stability index, and the calculation expression is:

[0131]

[0132] Where, P s represents the drug delivery frequency stability index, and ω ∈ Ω represents the low-frequency fluctuation frequency band;

[0133] Judge whether the drug delivery frequency stability index is greater than or equal to the preset threshold. If so, it means that the corresponding drug delivery is stable. If not, it means that the corresponding drug delivery is unstable;

[0134] It should be noted that: The drug delivery frequency stability index reflects the stability of drug delivery during the drug delivery process of animals, and the larger the value of the drug delivery frequency stability index, the more stable the corresponding delivery process.

[0135] Please refer to Figure 2 As shown, in S4, comprehensively analyze the fluctuation degree of drug concentration data within the monitoring period and the stability of drug delivery, and identify whether there is any abnormality in drug delivery, specifically including:

[0136] Obtain the drug concentration anomaly index I f and the drug delivery frequency stability index P s ;

[0137] Process the drug concentration anomaly index I f and the drug delivery frequency stability index P s within the monitoring period through a Bayesian network, and calculate the expected value of the comprehensive anomaly index IP, specifically including:

[0138] Set the conditional probability relationship of the Bayesian network as: P(IP|I f ,P s );

[0139] Where, P(IP|I f,P s ) represents the conditional probability distribution of the comprehensive anomaly index IP;

[0140] Calculate the joint probability distribution through Bayes' formula. The calculation expression is:

[0141] P(IP|I f ,P s ) = P(IP|I f ,P s )·p(I f )·P(P s );

[0142] Among them, P(I f ) represents the prior probability distribution of the drug concentration anomaly index, and P(P s ) represents the prior probability distribution of the dosing frequency stability index;

[0143] Calculate the expected value of the comprehensive anomaly index. The calculation expression is:

[0144]

[0145] Among them, represents the expected value of the comprehensive anomaly index;

[0146] Judge whether the expected value of the comprehensive anomaly index is greater than or equal to the preset threshold. If so, it indicates that there is an anomaly in the dosing of the corresponding drug. If not, it indicates that there is no anomaly in the dosing of the corresponding drug;

[0147] It should be noted that: The comprehensive anomaly index reflects the comprehensive anomaly degree in the drug dosing process, and the greater the value of the comprehensive anomaly index, the higher the anomaly degree in the corresponding drug dosing process.

[0148] In S5, combine the physiological data of the animal and the drug response results to construct an association model between drug concentration, dosing frequency and animal physiological response, analyze the impact degree of the abnormal drug dosing on the animal physiology, and perform dynamic regulation on drug concentration and dosing frequency according to the analysis result of the impact degree, specifically including:

[0149] Collect the animal physiological data within the monitoring period, including the body temperature and heart rate of the animal within the monitoring period;

[0150] Obtain the expected value of the comprehensive anomaly index of the drug dosing within the monitoring period;

[0151] Take the body temperature and heart rate of the animal within the monitoring period and the expected value of the comprehensive anomaly index of the drug dosing within the monitoring period as the input of the association model, construct the association model, and the calculation expression is:

[0152]

[0153] Among them, \(t\) represents the time point of animal physiological data collection, and \(Y(t)\) represents the physiological data. represents the mean function, represents the kernel function;

[0154] Among them, the mean function

[0155] In the formula, \(\beta\) 0 represents the baseline value of the physiological response, and \(\beta\) 1 represents the influence coefficient of the comprehensive anomaly index on the physiological response, represents the expected value of the comprehensive anomaly index;

[0156] Among them, the kernel function

[0157] In the formula, \(\sigma\) 2 represents the amplitude parameter of the kernel function, and \(l\) represents the length scale parameter of the kernel function;

[0158] The model parameters \(\beta\) 0 , \(\beta\) 1 , \(\sigma\) 2 and \(l\) are fitted by the maximum likelihood estimation method, and the predicted value of the mean of the physiological response is calculated according to the fitting result. The calculation expression is:

[0159]

[0160] In the formula, represents the mean of the predicted physiological response, represents the new data point and the covariance vector between the training data , \(K\) represents the covariance matrix of the training data, and \(Y\) represents the physiological data;

[0161] The predicted variance is calculated. The calculation expression is:

[0162]

[0163] In the formula, represents the predicted variance, represents the autocovariance of the new data point ;

[0164] The influence coefficient of drug delivery anomaly is calculated. The calculation expression is:

[0165]

[0166] In the formula, \(I\) impact represents the influence coefficient of drug delivery anomaly;

[0167] Determine whether the influence coefficient of abnormal drug delivery is greater than or equal to a preset threshold. If so, it indicates that the abnormal drug delivery has an impact on the physiological state of the animal. If not, it indicates that the abnormal drug delivery has no impact on the physiological state of the animal;

[0168] According to the analysis results of the influence degree, perform intelligent dynamic regulation of drug concentration and dosing frequency, including: The system collects the drug concentration, dosing frequency, and physiological data in the animal's body in real time, combines the drug concentration abnormality index and dosing frequency stability index, analyzes the influence degree of drug delivery on the animal's physiological state, constructs an optimized correlation model, and predicts the optimal regulation range of drug delivery according to the relationship between the comprehensive abnormality index and physiological response, and determines the adjustment direction and amplitude of drug concentration and dosing frequency;

[0169] Adaptive adjustment of drug concentration and dosing frequency according to real-time data feedback. After each monitoring cycle, the system optimizes the dosing frequency and drug concentration through a reward function based on the animal's physiological response and historical data of drug delivery to ensure the stability of drug concentration fluctuations and dosing frequency.

[0170] The working principle of the present invention: The present invention realizes precise control and optimization in the process of veterinary drug delivery through intelligent means. The method of the present invention collects the drug concentration data, drug delivery frequency data, and physiological data (such as body temperature and heart rate) in the animal's body in real time during the animal drug delivery monitoring cycle, and conducts in-depth analysis in combination with these data to evaluate the stability and abnormalities of drug delivery. First, use devices such as biosensors, intelligent drug infusion pumps, and electrocardiogram sensors to collect data and monitor the fluctuations of drug concentration and dosing frequency in real time. Then, process the collected data through discrete wavelet transform and Fourier transform, calculate the drug concentration abnormality index and dosing frequency stability index respectively to evaluate the stability of drug concentration fluctuations and dosing frequency. On this basis, comprehensively analyze the relationship between drug concentration and dosing frequency, combine Bayesian network to infer the comprehensive abnormality index, and further judge whether there is an abnormality in drug delivery.

[0171] Next, in combination with the physiological responses of the animals and the changes in drug concentration and dosing frequency, an association model is constructed. The model parameters are fitted by the maximum likelihood estimation method, the mean predicted value of the physiological response is calculated, and the impact degree of abnormal drug delivery on the animals' physiology is judged. If the abnormal drug delivery affects the physiological state of the animals, the drug concentration and dosing frequency are optimized and adjusted through intelligent dynamic regulation. After each monitoring cycle, the system optimizes the drug delivery strategy through a reward function based on the feedback of historical data and real-time data, ensuring the stability of the drug concentration and dosing frequency, and minimizing the potential risks of drug concentration fluctuations to the animals' health. Through the intelligent regulation mechanism, the present invention not only improves the accuracy and stability of drug delivery, but also provides an innovative solution for veterinary drug management and animal health, can effectively achieve personalized regulation, improve the drug delivery effect, reduce drug side effects, and promote the sustainable development of healthy animal breeding.

[0172] The above formulas are all calculated by taking the numerical values without dimensions. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0173] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other arbitrary combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center containing one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0174] It should be understood that the term "and / or" in this text is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this text generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context before and after.

[0175] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0176] The above has described in detail one embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A veterinary drug parameter control optimization method based on artificial intelligence integrated service, characterized in that: The following steps are involved: S1: During an animal medication monitoring cycle, real-time data on drug concentration in the animal, drug administration frequency data, and animal physiological data are collected, wherein the animal physiological data includes: animal body temperature and heart rate; S2: Analyze the collected drug concentration data in animals, and calculate the drug concentration abnormality index according to the fluctuation degree of drug concentration data during the monitoring period, which is used to evaluate the abnormal degree of drug concentration fluctuation; S3: Analyze the collected drug delivery frequency data, and calculate the delivery frequency stability index according to the fluctuation range of the drug delivery frequency during the monitoring period, so as to evaluate the stability of drug delivery; S4: Comprehensively analyze the fluctuation of drug concentration data during the monitoring period and the stability of drug delivery to identify whether there is any abnormality in drug delivery; S5: Combine the animal's physiological data and drug response results to construct a correlation model between drug concentration, administration frequency and animal physiological response, analyze the impact of abnormal drug administration on animal physiology, and dynamically regulate drug concentration and administration frequency based on the analysis results of the impact.

2. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The process of obtaining the abnormal drug concentration index is as follows: During the monitoring period, drug concentration data is collected in real time according to the time series, and the collected drug concentration time series data in the animal body is normalized. The calculation expression is: In the formula, i represents the number of drug concentration data, i is a positive integer greater than 0, and x i is the ith real-time drug concentration data, x′ i represents the drug concentration data after the i-th normalization process, max(X) represents the maximum drug concentration data in the drug concentration time series data, and min(X) represents the minimum drug concentration data in the drug concentration time series data; The normalized drug concentration time series is decomposed into multiple scales using discrete wavelet transform. The decomposition formula is: In the formula, J represents the scale number of wavelet decomposition, j represents the jth scale of wavelet decomposition, and D j (t) represents the high frequency component of the jth layer, A J (t) represents the low-frequency component of the Jth layer, t represents the acquisition time point, and t is a positive integer greater than 0; Calculate the high frequency component D of each layer j The wave energy of (t) is calculated as: In the formula, n represents the total number of acquisition time points, E j represents the fluctuation energy of the high-frequency component of the jth layer; According to the high-frequency component fluctuation energy E of each layer j , combined with the preset proportional coefficient, the drug concentration abnormality index is calculated, and the calculation expression is: In the formula, I f represents the abnormal drug concentration index, α j Indicates the preset scale factor.

3. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The abnormal degree of the evaluation of drug concentration fluctuation specifically includes: It is determined whether the drug concentration abnormality index is greater than or equal to a preset threshold value. If so, it indicates that the corresponding drug concentration fluctuation is abnormal; if not, it indicates that the corresponding drug concentration fluctuation is normal.

4. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The process of obtaining the delivery frequency stability index is as follows: During the monitoring period, the drug delivery frequency is collected in real time according to the time series, and the drug delivery frequency data is normalized to obtain the normalized drug delivery frequency data. The calculation expression is: In the formula, a represents the number of drug delivery frequency data, a is a positive integer greater than 0, and f a represents the ath real-time drug delivery frequency, f′ a represents the ath normalized drug delivery frequency, max(f) represents the maximum drug delivery frequency of the drug delivery frequency data, and min(f) represents the minimum drug delivery frequency of the drug delivery frequency data; The normalized drug delivery frequency data is Fourier transformed to obtain the frequency domain representation. The calculation expression is: In the formula, F(ω) represents the Fourier transform result of the drug delivery frequency data, t represents the acquisition time point, t is a positive integer greater than 0, n represents the total number of acquisition time points, ω represents the frequency component, and z represents the imaginary unit; Calculate the amplitude spectrum of the Fourier transform result. The calculation expression is: Where A(ω) is the amplitude of the corresponding frequency component, Re(F(ω)) represents the real part of F(ω), and Im(F(ω)) represents the imaginary part of F(ω); Calculate the total energy in the frequency domain. The calculation expression is: E total =∑ ω A(oh) 2 ; In the formula, E total represents the total energy in the frequency domain, A(ω) 2 is the energy of each frequency component; Calculate the drug delivery frequency stability index, the calculation expression is: Where P s It represents the stability index of drug delivery frequency, and ω∈Ω represents the low-frequency fluctuation frequency segment.

5. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The evaluation of the stability of drug delivery specifically includes: It is determined whether the drug delivery frequency stability index is greater than or equal to a preset threshold value. If so, it means that the corresponding drug delivery is stable; if not, it means that the corresponding drug delivery is unstable.

6. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The comprehensive analysis of the fluctuation degree of drug concentration data within the monitoring period and the stability of drug delivery specifically includes: Obtain the abnormal drug concentration index I during the monitoring period f and drug administration frequency stability index P s ; The abnormal drug concentration index I during the monitoring period was calculated by Bayesian network. f and drug administration frequency stability index P s Processing is performed to calculate the expected value of the comprehensive abnormality index IP, including: The conditional probability relationship of the Bayesian network is set as: P(IP|I f ,P s ); Among them, P(IP|I f ,P s ) represents the conditional probability distribution of the comprehensive anomaly index IP; By using the Bayesian formula, the joint probability distribution is calculated, and the calculation expression is: P(IP|I f ,P s )=P(IP|I f ,P s )·P(I f )·P(P s ); Among them, P(I f ) represents the prior probability distribution of the abnormal drug concentration index, P(P s ) represents the prior probability distribution of the delivery frequency stability index; Calculate the expected value of the comprehensive abnormality index, the calculation expression is: in, represents the expected value of the comprehensive anomaly index; It is determined whether the expected value of the comprehensive abnormality index is greater than or equal to a preset threshold value. If so, it indicates that there is an abnormality in the administration of the corresponding drug. If not, it indicates that there is no abnormality in the administration of the corresponding drug.

7. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The method combines the physiological data of the animal and the drug response results to construct a correlation model between drug concentration, drug administration frequency and animal physiological response, and analyzes the impact of abnormal drug administration on animal physiology, specifically including: Collect animal physiological data during the monitoring period, including the animal's body temperature and heart rate during the monitoring period; Obtaining the expected value of the comprehensive abnormal index of drug delivery during the monitoring period; The expected value of the comprehensive abnormal index of the animal's body temperature and heart rate during the monitoring period and the drug delivery during the monitoring period is used as the input of the association model to construct the association model. The calculation expression is: Where t represents the time point of animal physiological data collection, Y(t) represents physiological data, represents the mean function, represents the kernel function; Among them, the mean function In the formula, β0 represents the baseline value of physiological response, β1 represents the influence coefficient of comprehensive abnormal index on physiological response, represents the expected value of the comprehensive anomaly index; Among them, the kernel function In the formula, σ 2 represents the amplitude parameter of the kernel function, and l represents the length scale parameter of the kernel function; The model parameters β0, β1, σ are fitted by maximum likelihood estimation. 2 and l, and calculate the predicted value of the mean physiological response based on the fitting results. The calculation expression is: In the formula, represents the mean of the predicted physiological response, Represents a new data point With training data The covariance vector between them, K represents the covariance matrix of the training data, and Y represents the physiological data; Calculate the prediction variance, the calculation expression is: In the formula, represents the prediction variance, Represents a new data point The autocovariance of Calculate the influence coefficient of abnormal drug delivery, the calculation expression is: In the formula, I impact Indicates the influence coefficient of abnormal drug delivery; It is determined whether the influence coefficient of abnormal drug administration is greater than or equal to a preset threshold value. If so, it means that the abnormal drug administration has an impact on the physiological state of the animal. If not, it means that the abnormal drug administration has no impact on the physiological state of the animal.

8. The veterinary drug parameter control optimization method based on artificial intelligence integrated service according to claim 1 is characterized in that: The method of dynamically regulating the drug concentration and dosing frequency according to the analysis results of the degree of impact specifically includes: According to the analysis results of the degree of influence, the drug concentration and dosing frequency are intelligently and dynamically regulated, including: the system collects the drug concentration, dosing frequency and physiological data in the animal in real time, combines the drug concentration abnormality index and the dosing frequency stability index, analyzes the degree of influence of drug administration on the physiological state of the animal, and constructs an optimized correlation model. According to the relationship between the comprehensive abnormality index and the physiological response, the system predicts the optimal control range of drug administration and determines the adjustment direction and amplitude of drug concentration and dosing frequency; The drug concentration and administration frequency are adaptively adjusted according to real-time data feedback. After each monitoring cycle, the system optimizes the administration frequency and drug concentration through a reward function based on the animal's physiological response and historical data of drug administration to ensure the stability of drug concentration fluctuations and administration frequency.

Images