Animal biochemical detection management and diagnosis system for zoo

Abstract

The invention relates to the technical field of animal management, in particular to an animal biochemical detection management and diagnosis system for a zoo. Through a non-intrusive sampling technology, multi-parameter biochemical detection, intelligent data analysis and early warning, comprehensive monitoring, accurate evaluation and prospective management of the health condition of animals in the zoo are realized; the system realizes comprehensive acquisition and analysis of multi-source biochemical data, covers multi-dimensional indexes from basic biochemistry to microbiology, improves the coverage rate of detection indexes, combines multi-source data fusion and individualized baseline establishment, realizes prediction of the change trend of health indexes and early recognition of potential health problems, and improves the detection accuracy. Based on a customized prediction and dynamic adjustment mechanism of individual historical data, the disease prediction advance period is averagely prolonged by 7-14 days, the prediction accuracy is improved, precious time is won for preventive intervention, and the risk of sudden health events is reduced.

Description

Technical Field

[0001] This invention relates to the field of animal management technology, and more specifically to an animal biochemical detection management and diagnostic system for zoos. Background Technology

[0002] Currently, animal health management in zoos relies primarily on human experience and basic physiological indicator testing, lacking intelligent technological support. Since animals cannot describe their symptoms, veterinarians must assess their health through behavioral observation and routine examinations, which can easily lead to misdiagnosis or missed diagnosis due to individual differences or complex cases. Furthermore, existing monitoring equipment (such as handheld thermometers and manual activity recording devices) cannot achieve continuous 24-hour data collection, making it difficult to detect sudden abnormal signs (such as a sudden rise in body temperature or a sudden decrease in activity at night), resulting in delays in disease prevention and control.

[0003] Zoos currently store their health data (such as medical records, environmental parameters, and behavioral records) in scattered paper archives or independent systems, lacking a unified data integration platform. This makes it impossible to predict disease risks through historical data modeling, and also prevents the provision of a scientific basis for husbandry optimization. For example, the correlation between seasonal behavioral changes in animals and potential health problems is often overlooked.

[0004] Therefore, more intelligent solutions are still needed to automate the management of animal health in zoos. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, this invention aims to provide an animal biochemical detection management and diagnostic system for zoos, enabling intelligent detection, analysis, and management of animal health in zoos.

[0006] To solve the above problems, the present invention adopts the following technical solution:

[0007] An animal biochemical detection management and diagnostic system for zoos includes a sample collection module, a detection and analysis module, an analysis and diagnosis module, and a health management module.

[0008] The sample collection module is used to collect urine, feces, and blood samples from animals in the zoo, identify individual animals, and associate individuals with samples.

[0009] The detection and analysis module is used to perform biochemical index analysis and sample quality control on the collected urine, feces and blood samples of animals.

[0010] The analysis and diagnostic module is used to calculate a comprehensive health index based on multiple biochemical indicators and to conduct a comprehensive health assessment of animals. It calculates the Z-scores of multiple biochemical indicators to obtain a multi-parameter abnormality score and assesses the animal's systemic function based on the multi-parameter abnormality score. It calculates a risk score for specific diseases based on multiple biochemical indicators to assess the animal's disease risk. It uses an autoregressive integrated moving average (ARIMA) model to predict the future trend of biochemical indicators and pay attention to biochemical indicators that may become abnormal in the future. It performs multi-source data fusion on the same biochemical indicator from different sources to obtain the standard value of the biochemical indicator. It updates the individualized baseline based on the historical data of biochemical indicators and seasonal factors.

[0011] The health management module is used to provide decision support for animal health management for managers.

[0012] As one possible implementation, the sample collection module includes a urine sample collection unit, a fecal sample collection unit, a blood sample collection unit, and an individual identification unit.

[0013] The urine sample collection unit includes a smart collection pad, an automatic extraction device, and a urine sample pretreatment device. The smart collection pad is placed in the animal's urination area to absorb urine and clean excess urine from the pad. The automatic extraction device is connected to the smart collection pad and extracts the urine absorbed by the pad at regular intervals or after sensing urine. The urine sample pretreatment device receives the urine extracted by the automatic extraction device and performs quality management on the urine.

[0014] The fecal sample collection unit includes an intelligent fecal collector and a fecal storage device. The intelligent fecal collector integrates infrared sensors, weight sensors, or odor sensors to detect fresh feces. The intelligent fecal collector integrates a robotic arm to collect fresh feces. The fecal storage device is used to store, sample, and manage the quality of the collected feces.

[0015] The blood sample collection unit includes a minimally invasive blood collection device and an induction device. The induction device is used to induce an animal to come into contact with the minimally invasive blood collection device, and the minimally invasive blood collection device is used to collect blood samples when the animal comes into contact with it.

[0016] The individual identification unit is used to acquire the animal's RFID tag, biometrics and / or behavioral characteristics to identify the individual animal, and to associate the collected samples with the individual animal based on the sample collection time and animal activity data.

[0017] As one possible implementation, the biochemical indicators of urine analyzed in the detection and analysis module include:

[0018] Basic physicochemical indicators: pH value, specific gravity, color, and transparency.

[0019] Biochemical indicators: protein, glucose, ketone bodies, bilirubin, and urobilinogen.

[0020] Trace elements: sodium, potassium, chlorine, calcium, and phosphorus.

[0021] Cytological examination: red blood cells, white blood cells, epithelial cells, and bacteria.

[0022] Special indicators: Specific indicators are detected based on the characteristics of the animal species.

[0023] The biochemical indicators of feces analyzed in the detection and analysis module include:

[0024] Macroscopic characteristics: color, shape, hardness, and odor.

[0025] Digestive function indicators: undigested food residue, fat, and starch.

[0026] Parasite examination: eggs, cysts, and adult worms.

[0027] Microbiological examination: pathogen culture and microbial community analysis.

[0028] Occult blood test: detects bleeding in the gastrointestinal tract.

[0029] Inflammatory markers: fecal calprotectin and lactoferrin.

[0030] The biochemical indicators of the blood analyzed in the detection and analysis module include:

[0031] Complete blood count: red blood cell count, white blood cell count, platelet count, and hemoglobin.

[0032] Blood biochemistry: liver function, kidney function, blood glucose, blood lipids, and electrolytes.

[0033] Hormone levels: thyroid hormones, sex hormones, and adrenal hormones.

[0034] Immune markers: immunoglobulins, complement, and inflammatory factors.

[0035] Special markers: Specific markers are detected based on animal species and health risks.

[0036] As one possible implementation method, the comprehensive health index HI is calculated using the following formula:

[0037]

[0038]

[0039] Among them, S i w is the standardized score of the i-th indicator. i V is the weight of the i-th indicator, n is the number of indicators considered, and Vi It is the original index value, M i This is the ideal median value of the indicator, R. i It is within an acceptable range of deviation.

[0040] As one possible implementation, the Z-score is calculated using the following formula:

[0041]

[0042] Among them, Z i V is the Z-score of the i-th indicator. i This is the current measurement value, μ i It is the historical average of this indicator; σ i It is the historical standard deviation of this indicator.

[0043] The multi-parameter anomaly score A is calculated using the following formula:

[0044]

[0045] Where n is the number of indicators.

[0046] As one possible implementation method, the risk score for the specific disease is calculated using the following formula:

[0047]

[0048]

[0049] Among them, R d β is the risk probability of disease d, β0 is the intercept term, and β i X is the coefficient of the i-th biochemical indicator, where n is the number of biochemical indicators considered. i It is a standardized biochemical indicator, V i These are the raw measurements of biochemical indicators, μ i It is the average value of this biochemical indicator, σ i It is the standard deviation of the biochemical indicator.

[0050] As one possible implementation method, the method of using the autoregressive integrated moving average (ARIMA) model to predict the future trend of biochemical indicators and to monitor biochemical indicators that may show abnormalities in the future includes:

[0051] Predicting future trends of biochemical indicators using the autoregressive composite moving average (ARIMA) model:

[0052] X t =c+ϕ1X t−1 +ϕ2X t−2 +...+ϕ p X t−p +θ1ε t−1+θ2ε t−2 +...+θ q ε t−q +ε t

[0053] Among them, X t Here, ϕ is the biochemical index value at time t, c is a constant term, and ϕ is the value of the biochemical index at time t. i It is the coefficient of the autoregressive term, θ j It is the coefficient of the moving average term, ε t is the random error at time t, p is the autoregressive order, and q is the moving average order.

[0054] Short-term biochemical indicators were calculated using the exponential smoothing method:

[0055]

[0056] Among them, S t F is the smoothed value at time t, α is the smoothing coefficient, and F is the smoothing value at time t. t+m It is the predicted value at time t+m.

[0057] The Holt linear trend method was used to calculate long-term biochemical indicators.

[0058] As one possible implementation, updating the individualized baseline based on historical data of biochemical indicators and seasonal factors includes:

[0059] Obtain historical baseline data for biochemical indicators.

[0060] Calculate the individualized baseline of biochemical indicators based on the historical baseline data of the obtained biochemical indicators:

[0061]

[0062] Among them, B i new It is the updated baseline of biochemical indicator i, B i old It is the original baseline of biochemical indicator i, V i,j γ is the j-th measurement of biochemical indicator i in a healthy state, m is the number of new health state measurements, and γ is the update weight.

[0063] Update individualized baselines based on seasonal factors:

[0064]

[0065] Among them, B i,s It is the adjusted baseline of biochemical indicator i under seasonal s, B i It is the updated baseline of biochemical indicator i, SF i,s It is the seasonality factor of biochemical indicator i in season s.

[0066] As one possible implementation, the multi-source data fusion of the same biochemical indicator from different sources to obtain the standard value of the biochemical indicator includes:

[0067] Multi-source data fusion of the same biochemical indicator from different sources was performed using a weighted average method.

[0068]

[0069]

[0070] Among them, V fused It is the merged index value, V i w is the measurement value from the i-th source. i σ is the weight of the i-th source, n is the number of data sources, and σ is the weight of the i-th source. i The measurement uncertainty is σ from the i-th source. j It is the measurement uncertainty from the j-th source.

[0071] As one possible implementation, it also includes a data management and analysis module, which is used to store and back up the biochemical index data obtained by the detection and analysis module, and to transmit it in real time with encryption. The health management module is used for single biochemical index analysis and display, multi-biochemical index correlation display, time series data analysis and display, and comparison display of individual and group biochemical indexes. Based on the analysis results of the analysis and diagnosis module, it provides recommendations for drugs, dosages and expected effects, compares changes in biochemical indicators before and after treatment, analyzes recovery trends, monitors side effects and analyzes the achievement rate of treatment goals, formulates health management plans for animals based on biochemical indicators and provides remote consultation.

[0072] The beneficial effects of this invention are as follows: This invention achieves comprehensive monitoring, accurate assessment, and proactive management of the health status of zoo animals through non-invasive sampling technology, multi-parameter biochemical detection, intelligent data analysis, and predictive early warning. The system realizes the comprehensive collection and analysis of multi-source biochemical data (urine, feces, blood), covering multi-dimensional indicators from basic biochemistry to microbiology, improving the coverage of detection indicators. Combined with multi-source data fusion and individualized baseline establishment, it improves the accuracy of detection results and the sensitivity of anomaly detection, enabling the capture of earlier and more subtle health changes. It enables the prediction of health indicator trends and the early identification of potential health problems. Based on individual historical data, the customized prediction and dynamic adjustment mechanism extends the average disease prediction lead time by 7-14 days, improves prediction accuracy, gains valuable time for preventive intervention, and significantly reduces the risk of sudden health events. The system's automated sampling, intelligent analysis, and decision support functions greatly improve management efficiency and reduce the workload of management personnel. Attached Figure Description

[0073] Figure 1 This is a schematic diagram of an animal biochemical detection management and diagnostic system for zoos according to the present invention. Detailed Implementation

[0074] The present invention will be further described in detail below with reference to specific embodiments.

[0075] It should be noted that these embodiments are only used to illustrate the present invention and are not intended to limit the present invention. Simple improvements to the method under the premise of the present invention are all within the scope of protection claimed by the present invention.

[0076] See Figure 1 This is an animal biochemical detection management and diagnostic system for zoos, including a sample collection module 100, a detection and analysis module 200, an analysis and diagnosis module 300, and a health management module 400.

[0077] The sample collection module 100 is used to collect urine, feces and blood samples from animals in the zoo, identify individual animals, and associate individuals with samples.

[0078] The sample collection module 100 includes a urine sample collection unit 110, a fecal sample collection unit 120, a blood sample collection unit 130, and an individual identification unit 140.

[0079] The urine sample collection unit 110 includes a smart collection pad, an automatic extraction device, and a urine sample pretreatment device. The smart collection pad is placed in the animal's urination area to absorb urine and clean excess urine from the smart collection pad. The automatic extraction device is connected to the smart collection pad and extracts the urine absorbed by the smart collection pad at regular intervals or after sensing urine. The urine sample pretreatment device is used to receive the urine extracted by the automatic extraction device and to perform quality management of the urine.

[0080] The urine sample pretreatment device employs a layered structure: a comfortable, permeable top layer, an absorbent middle layer, and a collection bottom layer. Immediately after collection, pH stabilization and contamination prevention treatments are performed to ensure sample quality.

[0081] The fecal sample collection unit 120 includes an intelligent fecal collector and a fecal storage device. The intelligent fecal collector integrates an infrared sensor, a weight sensor, or an odor sensor to detect fresh feces. The intelligent fecal collector integrates a robotic arm to collect fresh feces. The fecal storage device is used to store, sample, and manage the quality of the collected feces.

[0082] Seal immediately after collection to prevent sample contamination and evaporation. Incubate or refrigerate samples as needed to maintain their activity. Perform preliminary processing (such as separation, grinding, dilution, etc.) according to testing requirements. Automatically clean and disinfect after each collection to ensure accuracy for the next collection.

[0083] The blood sample collection unit 130 includes a minimally invasive blood collection device and an induction device. The induction device is used to induce an animal to come into contact with the minimally invasive blood collection device, and the minimally invasive blood collection device is used to collect blood samples when the animal comes into contact with it.

[0084] The minimally invasive blood collection device uses disposable needles and an automatic disinfection mechanism to ensure a safe and hygienic collection process.

[0085] The induction device, combined with food rewards or other positive stimuli, encourages animals to actively contact the device. It can also be equipped with restraint devices to stabilize the animal to some extent when it comes into contact with the minimally invasive blood collection device, facilitating blood collection. The collection process is optimized to shorten collection time, reduce animal stress, and ensure the stability of blood samples using special preservation solutions or real-time analysis technologies.

[0086] The individual identification unit 140 is used to acquire the animal's RFID tag, biometrics (such as appearance, hair texture, iris, etc.), and / or behavioral characteristics to identify the individual animal, and to associate the collected samples with the individual animals based on the sample collection time and animal activity data. This ensures accurate matching between the collected samples and the animals. A unique identifier is added to each sample to ensure traceability throughout the entire process.

[0087] The detection and analysis module 200 is used to perform biochemical index analysis and sample quality control on collected animal urine, feces and blood samples.

[0088] The biochemical indicators of urine analyzed in the detection and analysis module include:

[0089] Basic physicochemical indicators: pH value, specific gravity, color, and transparency, etc.

[0090] Biochemical indicators: protein, glucose, ketone bodies, bilirubin, and urobilinogen, etc.

[0091] Trace elements: sodium, potassium, chlorine, calcium, and phosphorus, etc.

[0092] Cytological examination: red blood cells, white blood cells, epithelial cells, and bacteria, etc.

[0093] Special indicators: Specific indicators are detected based on the characteristics of the animal species.

[0094] The biochemical indicators of feces analyzed in the detection and analysis module include:

[0095] Macroscopic characteristics: color, shape, hardness, and odor, etc.

[0096] Digestive function indicators: undigested food residue, fat, and starch, etc.

[0097] Parasite examination: eggs, cysts, and adult worms, etc.

[0098] Microbiological examination: pathogen culture, microbial community analysis, etc.

[0099] Occult blood test: detects gastrointestinal bleeding, etc.

[0100] Inflammatory markers: fecal calprotectin and lactoferrin, etc.

[0101] The biochemical indicators of blood analyzed in the detection and analysis module include:

[0102] Complete blood count (CBC): Red blood cell count, white blood cell count, platelet count, and hemoglobin, etc.

[0103] Blood biochemistry: liver function, kidney function, blood glucose, blood lipids, and electrolytes, etc.

[0104] Hormone levels: thyroid hormones, sex hormones, and adrenal hormones, etc.

[0105] Immune markers: immunoglobulins, complement, and inflammatory factors, etc.

[0106] Special markers: Specific markers for detecting animal species and health risks, etc.

[0107] The analysis and diagnostic module 300 is used to calculate a comprehensive health index based on multiple derived biochemical indicators and to conduct a comprehensive health assessment of the animal.

[0108] The Comprehensive Health Index (HI) is calculated using the following formula:

[0109]

[0110]

[0111] Among them, S i w is the standardized score of the i-th indicator. i V is the weight of the i-th indicator, n is the number of indicators considered, and V i It is the original index value, M i This is the ideal median value of the indicator, R. i It is within an acceptable range of deviation.

[0112] For example, when conducting a health assessment of an adult male tiger, the following three key indicators are considered: hemoglobin (Hb), creatinine (Cr), a renal function indicator, and alanine aminotransferase (ALT), a liver function indicator. The ideal median, acceptable range of deviation, actual measured value, and weight of each indicator are as follows:

[0113] Hb: Ideal median 15 g / dL, deviation range ±3 g / dL, actual value 13.5 g / dL, weight 0.4.

[0114] Cr: Ideal median 1.5 mg / dL, deviation range ±0.5 mg / dL, actual value 1.8 mg / dL, weight 0.3.

[0115] ALT: Ideal median 40 U / L, deviation range ±20 U / L, actual value 55 U / L, weight 0.3.

[0116] Calculate the standardized score:

[0117]

[0118] Calculate the overall health index:

[0119] HI=0.4×0.5+0.3×0.4+0.3×0.25=0.2+0.12+0.075=0.395.

[0120] The health index is 0.395 (out of 1), indicating that the tiger's health condition is below average and requires further examination and possible intervention.

[0121] Z-scores for multiple biochemical indicators are calculated to obtain a multi-parameter anomaly score, which is then used to assess the animal's systemic function. Animal systems include the digestive, urinary, and immune systems.

[0122] The Z-score is calculated using the following formula:

[0123]

[0124] Among them, Z i V is the Z-score of the i-th indicator. i This is the current measurement value, μ i It is the historical average of this indicator; σ i It is the historical standard deviation of this indicator.

[0125] The multi-parameter anomaly score A is calculated using the following formula:

[0126]

[0127] Where n is the number of indicators.

[0128] For example, analyzing a urine sample from a giant panda and testing for the following four indicators: pH, protein, glucose, and white blood cells. The historical averages, standard deviations, and current measurements for each indicator are as follows:

[0129] pH value: historical average 6.5, standard deviation 0.3, current value 7.2.

[0130] Protein: Historical mean 10 mg / dL, standard deviation 5 mg / dL, current value 35 mg / dL.

[0131] Glucose: Historical mean 5 mg / dL, standard deviation 2 mg / dL, current value 4 mg / dL.

[0132] White blood cells: historical mean 2 / HPF, standard deviation 1 / HPF, current value 8 / HPF.

[0133] Calculate the Z-scores for each indicator:

[0134]

[0135] Calculate multi-parameter anomaly scores:

[0136] .

[0137] The abnormal score was 4.08, which is far above the normal range (a score above 3.0 is generally considered significantly abnormal), indicating that the giant panda may have a urinary system disease, especially a high risk of proteinuria and urinary tract infection, requiring immediate further examination and intervention.

[0138] Animals are assessed for disease risk by calculating risk scores for specific diseases based on multiple biochemical indicators.

[0139] The risk score for a specific disease is calculated using the following formula:

[0140]

[0141]

[0142] Among them, R d β is the risk probability of disease d, β0 is the intercept term, and β i X is the coefficient of the i-th biochemical indicator, where n is the number of biochemical indicators considered. i It is a standardized biochemical indicator, V i These are the raw measurements of biochemical indicators, μ i It is the average value of this biochemical indicator, σ i It is the standard deviation of the biochemical indicator.

[0143] For example, to assess the risk of kidney disease in a female African lion, the following biochemical indicators are considered: age, serum creatinine level, urine protein level, and blood pressure. The model parameters obtained through historical data analysis are as follows:

[0144] Intercept β0 = -3.5, age coefficient β1 = 0.4 (after standardization), serum creatinine coefficient β2 = 0.8, urine protein coefficient β3 = 0.6, blood pressure coefficient β4 = 0.3.

[0145] The actual measurements of the African lion (standardized):

[0146] Standardized age X1 = 1.2 (higher than average age), standardized serum creatinine X2 = 2.5 (significantly higher than normal), standardized urine protein X3 = 1.8 (higher than normal), and standardized blood pressure X4 = 0.5 (slightly higher than normal).

[0147] Calculate the kidney disease risk score:

[0148]

[0149] A risk score of 0.55 (or 55%) indicates that the African lion has a moderate to high risk of developing kidney disease, and more frequent monitoring and appropriate preventive interventions are recommended.

[0150] The Autoregressive Integrated Moving Average (ARIMA) model is used to predict future trends in biochemical indicators and to monitor for any abnormal biochemical indicators that may emerge in the future.

[0151] Predicting future trends of biochemical indicators using the autoregressive composite moving average (ARIMA) model:

[0152] X t =c+ϕ1X t−1 +ϕ2X t−2 +...+ϕ p X t−p +θ1ε t−1 +θ2ε t−2 +...+θ q ε t−q +ε t

[0153] Among them, X t Here, ϕ is the biochemical index value at time t, c is a constant term, and ϕ is the value of the biochemical index at time t. i It is the coefficient of the autoregressive term, θ j It is the coefficient of the moving average term, ε t is the random error at time t, p is the autoregressive order, and q is the moving average order.

[0154] Short-term biochemical indicators were calculated using the exponential smoothing method:

[0155]

[0156] Among them, S tF is the smoothed value at time t, α is the smoothing coefficient, and F is the smoothing value at time t. t+m It is the predicted value at time t+m.

[0157] The Holt linear trend method was used to calculate long-term biochemical indicators.

[0158] For example, predicting the blood glucose level trend of a male giraffe. The blood glucose measurements (in mg / dL) for the past 10 days are as follows: [95, 98, 102, 105, 108, 112, 115, 118, 122, 125]. Using a simple exponential smoothing method (α=0.3), predict the blood glucose level for the next 3 days: First, calculate the smoothed value sequence:

[0159] S1=X1=95.

[0160]

[0161] Assuming S10 = 115.2, the predicted values ​​for the next 3 days are:

[0162] F11=F12=F13=S10=115.2.

[0163] However, since the data clearly shows an upward trend, simple exponential smoothing may underestimate future values. Using the Holt linear trend method (which considers trend factors) may be more accurate: Assuming the calculation using the Holt method yields:

[0164] .

[0165] This indicates that the giraffe's blood sugar levels are showing a continuous upward trend, and further examination and intervention may be necessary.

[0166] Multi-source data fusion is performed on the same biochemical indicator from different sources to obtain the standard value of the biochemical indicator.

[0167] Multi-source data fusion of the same biochemical indicator from different sources was performed using a weighted average method.

[0168]

[0169]

[0170] Among them, V fused It is the merged index value, V i w is the measurement value from the i-th source. i σ is the weight of the i-th source, n is the number of data sources, and σ is the weight of the i-th source. i The measurement uncertainty is σ from the i-th source. j It is the measurement uncertainty from the j-th source.

[0171] For example, measuring the body temperature of a male white tiger using three different methods:

[0172] Infrared thermal imaging: 38.5℃, measurement uncertainty ℃.

[0173] Minimally invasive body temperature sensor: 38.9℃, measurement uncertainty 0.2℃.

[0174] Fecal thermometer: 38.7℃, measurement uncertainty 0.3℃.

[0175] Calculate the weights of each measurement method:

[0176]

[0177] Calculate the body temperature after fusion:

[0178] V fused =0.15×38.5+0.59×38.9+0.26×38.7=5.775+22.951+10.062=38.79℃.

[0179] The estimated body temperature after fusion was 38.79℃, which is closer to the reading of a minimally invasive body temperature sensor with lower measurement uncertainty, but also takes into account the contributions of other measurement methods.

[0180] Individualized baselines are updated based on historical biochemical data and seasonal factors, including:

[0181] Obtain historical baseline data for biochemical indicators.

[0182] Calculate the individualized baseline of biochemical indicators based on the historical baseline data of the obtained biochemical indicators:

[0183]

[0184] Among them, B i new It is the updated baseline of biochemical indicator i, B i old It is the original baseline of biochemical indicator i, V i,j γ is the j-th measurement of biochemical indicator i in a healthy state, m is the number of new health state measurements, and γ is the update weight.

[0185] Update individualized baselines based on seasonal factors:

[0186]

[0187] Among them, B i,s It is the adjusted baseline of biochemical indicator i under seasonal s, B i It is the updated baseline of biochemical indicator i, SF i,sIt is the seasonality factor of biochemical indicator i in season s.

[0188] For example, updating the individualized baseline of hemoglobin (Hb) for a female red panda. The original baseline was 13.5 g / dL, and the five most recent measurements in a healthy state were: [13.8, 14.0, 13.7, 14.2, 13.9] g / dL, with an update weight γ=0.3. Calculate the average of the newly added measurements:

[0189]

[0190] Calculate the updated baseline:

[0191]

[0192] The updated baseline hemoglobin level is 13.63 g / dL.

[0193] Assuming winter affects the red panda's hemoglobin, with a winter seasonality factor of 1.05 (winter hemoglobin is typically 5% higher), then the winter-adjusted baseline is:

[0194] BHb, winter = 13.63 × 1.05 = 14.31.

[0195] This means that during winter, the red panda's normal baseline hemoglobin level should be 14.31 g / dL, instead of the annual average of 13.63 g / dL.

[0196] The analysis and diagnosis module is also used to assess nutritional status such as protein, fat, vitamins, and minerals; evaluate the match between the animal's physiological age and actual age; use decision trees for preliminary screening; use support vector machines to identify complex disease patterns; use deep neural networks to analyze multi-source data; and integrate the diagnostic results of multiple models based on ensemble learning. It can predict changes in health indicators over the next 7-30 days, assess the probability of developing a specific disease, provide different levels of early warning based on the risk level, and provide preventive intervention recommendations based on the prediction results.

[0197] The health management module 400 provides decision support for animal health management for managers. This includes features such as single biochemical indicator analysis and display, multi-biochemical indicator correlation display, time-series data analysis and display, comparison of individual and group biochemical indicators, recommendations for drugs, dosages, and expected effects based on the analysis results from the analysis and diagnosis module, comparison of changes in biochemical indicators before and after treatment, recovery trend analysis, side effect monitoring, and treatment goal achievement rate analysis. It also enables the development of health management plans for animals based on biochemical indicators and provides remote consultations.

[0198] The system integrates the diagnostic experience and knowledge of veterinary experts, collects typical disease cases and their biochemical characteristics, integrates relevant findings from the latest research literature, continuously optimizes the diagnostic model based on diagnostic feedback, constructs an ontology in the field of animal health, establishes a disease-symptom-indicator relationship network, performs inference based on similar cases, and continuously updates the model based on new data and feedback.

[0199] It also includes a data management and analysis module 500, which is used to store and back up the biochemical indicator data obtained by the detection and analysis module, and to transmit it in real time with encryption.

[0200] Supports data caching and resumption in the event of network interruption. Performs data integrity and validity verification during transmission. Stores detection data in a structured format according to a standard format. Maintains complete historical detection data and supports long-term trend analysis.

[0201] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described with reference to preferred embodiments, those skilled in the art should understand that various changes in form and detail can be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. An animal biochemical detection, management, and diagnostic system for zoos, characterized in that, It includes a sample collection module, a detection and analysis module, an analysis and diagnosis module, and a health management module; The sample collection module is used to collect urine, feces and blood samples from animals in the zoo, identify individual animals, and associate individuals with samples. The detection and analysis module is used to perform biochemical index analysis and sample quality control on the collected urine, feces and blood samples of animals. The analysis and diagnosis module is used to calculate a comprehensive health index based on multiple biochemical indicators and to conduct a comprehensive health assessment of animals. It calculates the Z-scores of multiple biochemical indicators to obtain a multi-parameter abnormality score and assesses the system function of animals based on the multi-parameter abnormality score. It calculates the risk score of specific diseases based on multiple biochemical indicators to assess the disease risk of animals. It uses an autoregressive integrated moving average (ARIMA) model to predict the future trend of biochemical indicators and pay attention to biochemical indicators that will become abnormal in the future. It performs multi-source data fusion on the same biochemical indicator from different sources to obtain the standard value of the biochemical indicator. It updates the individualized baseline based on the historical data of biochemical indicators and seasonal factors. The health management module is used to provide decision support for animal health management for managers.

2. The animal biochemical detection management and diagnostic system for zoos according to claim 1, characterized in that, The sample collection module includes a urine sample collection unit, a fecal sample collection unit, a blood sample collection unit, and an individual identification unit; The urine sample collection unit includes an intelligent collection pad, an automatic extraction device, and a urine sample pretreatment device. The intelligent collection pad is placed in the animal's urination area to absorb urine and clean excess urine from the pad. The automatic extraction device is connected to the intelligent collection pad and extracts the urine absorbed by the pad at regular intervals or after sensing urine. The urine sample pretreatment device receives the urine extracted by the automatic extraction device and performs quality management on the urine. The fecal sample collection unit includes an intelligent fecal collector and a fecal storage device. The intelligent fecal collector integrates an infrared sensor, a weight sensor, or an odor sensor to detect fresh feces. The intelligent fecal collector integrates a robotic arm to collect fresh feces. The fecal storage device is used to store, sample, and manage the quality of the collected feces. The blood sample collection unit includes a minimally invasive blood collection device and an induction device. The induction device is used to induce an animal to come into contact with the minimally invasive blood collection device, and the minimally invasive blood collection device is used to collect blood samples when the animal comes into contact with it. The individual identification unit is used to acquire the animal's RFID tag, biometrics and / or behavioral characteristics to identify the individual animal, and to associate the collected samples with the individual animal based on the sample collection time and animal activity data.

3. The animal biochemical detection management and diagnostic system for zoos according to claim 2, characterized in that, The biochemical indicators of urine analyzed in the detection and analysis module include: Basic physicochemical properties: pH value, specific gravity, color, and transparency; Biochemical indicators: protein, glucose, ketone bodies, bilirubin, and urobilinogen; Trace elements: sodium, potassium, chlorine, calcium, and phosphorus; Cytological examination: red blood cells, white blood cells, epithelial cells, and bacteria; Special indicators: Specific indicators are detected based on the characteristics of different animal species; The biochemical indicators of feces analyzed in the detection and analysis module include: Macroscopic characteristics: color, shape, hardness, and odor; Digestive function indicators: undigested food residue, fat, and starch; Parasite examination: eggs, cysts, and adult worms; Microbiological examination: pathogen culture, microbial community analysis; Occult blood test: detects gastrointestinal bleeding; Inflammatory markers: fecal calprotectin and lactoferrin; The biochemical indicators of the blood analyzed in the detection and analysis module include: Complete blood count: red blood cell count, white blood cell count, platelet count, and hemoglobin; Blood biochemistry: liver function, kidney function, blood glucose, blood lipids, and electrolytes; Hormone levels: thyroid hormones, sex hormones, and adrenal hormones; Immune markers: immunoglobulins, complement, and inflammatory factors; Special markers: Specific markers are detected based on animal species and health risks.

4. The animal biochemical detection management and diagnostic system for zoos according to claim 3, characterized in that, The comprehensive health index HI is calculated using the following formula: Among them, S i w is the standardized score of the i-th indicator. i V is the weight of the i-th indicator, n is the number of indicators considered, and V i It is the original index value, M i This is the ideal median value of the indicator, R. i It is within an acceptable range of deviation.

5. The animal biochemical detection management and diagnostic system for zoos according to claim 4, characterized in that, The Z-score is calculated using the following formula: Among them, Z i V is the Z-score of the i-th indicator. i This is the current measurement value, μ i It is the historical average of this indicator; σ i This is the historical standard deviation of the indicator; The multi-parameter anomaly score A is calculated using the following formula: Where n is the number of indicators.

6. The animal biochemical detection management and diagnostic system for zoos according to claim 5, characterized in that, The risk score for the specific disease is calculated using the following formula: Among them, R d β is the risk probability of disease d, β0 is the intercept term, and β i X is the coefficient of the i-th biochemical indicator, where n is the number of biochemical indicators considered. i It is a standardized biochemical indicator, V i These are the raw measurements of biochemical indicators, μ i It is the average value of this biochemical indicator, σ i It is the standard deviation of the biochemical indicator.

7. The animal biochemical detection management and diagnostic system for zoos according to claim 6, characterized in that, The use of the Autoregressive Integrated Moving Average (ARIMA) model to predict future trends of biochemical indicators and to monitor for abnormal biochemical indicators in the future includes: Predicting future trends of biochemical indicators using the autoregressive composite moving average (ARIMA) model: X t =c+ϕ1X t−1 +ϕ2X t−2 +...+φ p X t−p +θ1ε t−1 +θ2ε t−2 +...+θ q e t−q +e t Among them, X t Here, ϕ is the biochemical index value at time t, c is a constant term, and ϕ is the value of the biochemical index at time t. i It is the coefficient of the autoregressive term, θ j It is the coefficient of the moving average term, ε t is the random error at time t, p is the autoregression order, and q is the moving average order; Short-term biochemical indicators were calculated using the exponential smoothing method: Among them, S t X is the smoothed value at time t. t It is the actual observed value at time t, α is the smoothing coefficient, and F t+m It is the predicted value at time t+m; The Holt linear trend method was used to calculate long-term biochemical indicators.

8. The animal biochemical detection management and diagnostic system for zoos according to claim 7, characterized in that, The method of updating the individualized baseline based on historical data of biochemical indicators and seasonal factors includes: Obtain historical baseline data for biochemical indicators; Calculate the individualized baseline of biochemical indicators based on the historical baseline data of the obtained biochemical indicators: Among them, B i new It is the updated baseline of biochemical indicator i, B i old It is the original baseline of biochemical indicator i, V i,j is the j-th measurement of biochemical indicator i in a healthy state, m is the number of new health state measurements, and γ is the update weight; Update individualized baselines based on seasonal factors: Among them, B i,s It is the adjusted baseline of biochemical indicator i under seasonal s, B i It is the updated baseline of biochemical indicator i, SF i,s It is the seasonality factor of biochemical indicator i in season s.

9. The animal biochemical detection management and diagnostic system for zoos according to claim 8, characterized in that, The method of fusing multi-source data of the same biochemical indicator from different sources to obtain the standard value of the biochemical indicator includes: Multi-source data fusion of the same biochemical indicator from different sources was performed using a weighted average method. Among them, V fused It is the merged index value, V i w is the measurement value from the i-th source. i σ is the weight of the i-th source, n is the number of data sources, and σ is the weight of the i-th source. i The measurement uncertainty is σ from the i-th source. j It is the measurement uncertainty from the j-th source.

10. The animal biochemical detection management and diagnostic system for zoos according to claim 9, characterized in that, It also includes a data management and analysis module, which stores and backs up the biochemical index data obtained by the detection and analysis module and transmits it in real time with encryption. The health management module is used for single biochemical index analysis and display, multi-biochemical index correlation display, time series data analysis and display, and comparison display of individual and group biochemical indexes. Based on the analysis results of the analysis and diagnosis module, it provides recommendations for drugs, dosages and expected effects, compares changes in biochemical indicators before and after treatment, analyzes recovery trends, monitors side effects and analyzes the achievement rate of treatment goals, formulates health management plans for animals based on biochemical indicators and provides remote consultation.

Images