An intelligent diagnosis system for animal husbandry and veterinary medicine based on machine learning

Abstract

The present invention discloses an intelligent livestock and veterinary diagnosis system based on machine learning, which specifically relates to the technical field of intelligent livestock diagnosis. The system collects livestock vital sign data through multiple sensors configured on an intelligent collar, processes livestock surface temperature, livestock chewing data, and livestock movement data to obtain the livestock's true body temperature, rumination time, movement state, and movement speed, inputs the livestock's heart rate, respiratory rate, true body temperature, rumination time, and movement speed into a livestock vital sign abnormality recognition model for abnormality recognition, acquires livestock behavior data through a multi-angle camera, and after a series of processing, an abnormal image data group is labeled, and the received livestock vital sign abnormality data group and livestock behavior abnormality image data group are input into a livestock health intelligent diagnosis model for health diagnosis, which can effectively improve the efficiency and accuracy of livestock disease diagnosis.

Description

Technical Field

[0001] The present invention relates to the technical field of animal husbandry intelligent diagnosis, and more specifically, to an animal husbandry and veterinary intelligent diagnosis system based on machine learning. Background Art

[0002] Traditional livestock health diagnosis mainly relies on the sensory observation and experience judgment of veterinarians. These methods are simple and easy to use and can be implemented in any place. However, the limitations of traditional livestock health diagnosis are also obvious. The diagnostic accuracy is insufficient, the diagnostic efficiency is low, the diagnostic range is limited, and it cannot adapt to the needs of large-scale breeding. It is necessary to develop a more advanced and intelligent livestock diagnosis method to improve diagnostic efficiency and accuracy.

[0003] Existing livestock health diagnosis monitors the livestock's motion physiological data through a livestock physiological indicator intelligent monitoring system composed of multiple physiological indicator collection devices, wireless base stations, servers, etc., and establishes a standard physiological indicator model and livestock health index based on the motion physiological data. During the monitoring process, the health index of livestock that does not conform to the model rules and is abnormal is screened out and sent to the veterinary side to assist veterinarians in diagnosing livestock diseases. This greatly reduces the tedious work of diagnosing livestock one by one, accurately locks the range of livestock with abnormalities, and can improve diagnostic efficiency and accuracy.

[0004] Although the existing livestock health diagnosis system can improve diagnostic efficiency and accuracy to a certain extent, the following problems also exist: First, the current AI judgment for livestock optimization is mainly based on the comparative difference analysis of livestock movement physiological data monitored by intelligent monitoring equipment and the established model. However, when livestock are sick, there may be multiple abnormal physical parameters or behavioral data. The monitoring scope of the existing system is relatively narrow and cannot provide sufficient data support for veterinary diagnosis. The accuracy of disease diagnosis still needs to be improved; Second, due to limited experience, veterinarians need to rule out irrelevant causes one by one when receiving abnormal monitoring results. Due to insufficient monitoring indicators, the process of tracing the cause of the disease may consume a lot of time, which is not conducive to the subsequent treatment of livestock. The efficiency of cause diagnosis still needs to be further improved. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent livestock and veterinary diagnosis system based on machine learning, which collects livestock physical sign data and behavioral data and performs preprocessing and abnormality analysis respectively, and at the same time constructs a health intelligent diagnosis model. The livestock physical sign abnormality data group and livestock behavior abnormality image data group can be directly input as input into the trained and tested livestock health intelligent diagnosis model for health diagnosis, which can effectively solve the problems of low diagnostic efficiency and low diagnostic accuracy.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: a machine learning-based animal husbandry and veterinary intelligent diagnosis system, comprising:

[0007] Vital sign and behavior data collection module: collects livestock's vital sign data through multiple sensors configured on the smart collar, and obtains livestock's behavior data through multi-angle cameras;

[0008] Vital sign data preprocessing module: processes livestock surface temperature, livestock chewing data and livestock movement data to obtain livestock's true body temperature, rumination time, movement status and movement speed;

[0009] Behavior data preprocessing module: This module uses image processing tools to extract images of livestock's eating posture, eating motion, movement posture, and lying posture from the behavior data, identifies whether the extracted posture images are accurate, and updates the behavior data based on the identification results.

[0010] Abnormal signs analysis module: The livestock's heart rate, respiratory rate, actual body temperature, rumination time, and movement speed are input into a trained and tested livestock abnormal signs recognition model to identify abnormalities, mark abnormal signs data groups, and send the abnormal signs data groups to the health intelligent diagnosis module after confirming that the livestock's physical signs are abnormal.

[0011] Abnormal behavior recognition module: Calculates the proportion of correct postures for eating, exercising, and resting behaviors, then calculates the abnormality coefficient for each behavior, and then calculates the comprehensive abnormality index of the behavior. Abnormal image data sets are marked, and after confirming that the livestock behavior is abnormal, the abnormal livestock behavior image data sets are sent to the health intelligent diagnosis module;

[0012] Intelligent Health Diagnosis Module: This module inputs the received livestock physical sign abnormality data set and livestock behavior abnormality image data set as input into a trained and tested livestock health intelligent diagnosis model for health diagnosis, outputs the abnormality cause and corresponding treatment plan with the highest matching degree, and determines whether the model diagnosis result triggers a secondary diagnosis. If so, a secondary diagnosis is performed and the final output abnormality cause and treatment plan are determined based on the set rules. If not, the abnormality cause and treatment plan output by the model diagnosis are directly output;

[0013] Database: used to store data information of all modules in the system.

[0014] Technical effects and advantages of the present invention:

[0015] The present invention collects livestock vital sign data through multiple sensors configured on a smart collar, processes livestock surface temperature, livestock chewing data, and livestock movement data to obtain the livestock's true body temperature, rumination time, movement state, and movement speed, inputs the livestock's heart rate, respiratory rate, true body temperature, rumination time, and movement speed into a trained and tested livestock vital sign abnormality recognition model for abnormality recognition, marks the livestock vital sign abnormality data group, and outputs the livestock vital sign abnormality data group after confirming that the livestock vital sign is abnormal, thereby improving the accuracy and efficiency of vital sign abnormality recognition.

[0016] The present invention obtains livestock behavior data through a multi-angle camera, uses image processing tools to extract livestock eating posture images, eating action images, exercise posture images, and lying posture images from the behavior data, and identifies whether the extracted posture images are accurate. The behavior data is updated based on the identification results, and the proportion of correct postures for eating behavior, exercise behavior, and resting behavior is calculated. Then, the abnormality coefficient of each behavior is calculated, and then the comprehensive abnormality index of the behavior is calculated. The abnormal image data group is marked, and the livestock behavior abnormality data group is output after confirming that the livestock behavior is abnormal, thereby improving the accuracy and efficiency of behavior abnormality identification.

[0017] After receiving the livestock physical sign abnormality data group and the livestock behavior abnormality image data group, the present invention inputs the received livestock physical sign abnormality data group and the livestock behavior abnormality image data group as input ends into the livestock health intelligent diagnosis model that has been trained and tested for health diagnosis. The model automatically matches the physical sign abnormality parameters and the behavior abnormality images with the abnormal physical sign manifestations and abnormal behavior manifestations corresponding to different causes in the machine learning process. The ratio of the number of overlapping abnormal parameters to the actual number of abnormal parameters of the livestock is the matching degree. The abnormal cause with the highest matching degree and the corresponding treatment plan are output as the output end. The diagnosis result is then judged, that is, the abnormal cause matching degree is compared with the set matching degree credibility threshold. If the abnormal cause matching degree is greater than Or equal to the set matching degree trust threshold, the abnormal cause with the highest matching degree and the corresponding treatment plan can be directly sent to the livestock user end. Otherwise, information is sent to multiple veterinarians registered in the system to request a secondary diagnosis. If the secondary diagnosis result is consistent with the model diagnosis result, the abnormal cause with the highest matching degree and the corresponding treatment plan will be sent to the livestock user end. Otherwise, based on the same principle, the matching degree of the abnormal signs and abnormal behavior of the abnormal cause obtained by the secondary diagnosis with the abnormal signs parameters and abnormal behavior images of the livestock is calculated, and the matching degree of the abnormal cause obtained by the model diagnosis and the secondary diagnosis is compared. The abnormal cause and treatment plan corresponding to the one with a higher matching degree value is the abnormal cause and treatment plan finally output, which improves the efficiency and accuracy of disease diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a system structure diagram of the present invention.

[0019] Figure 2 A diagram showing the steps of the method of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] like Figure 1 The embodiment shown provides an intelligent livestock and veterinary diagnosis system based on machine learning, including a vital sign and behavior data acquisition module, a vital sign data preprocessing module, a behavior data preprocessing module, a vital sign abnormality analysis module, a behavior abnormality recognition module, a health intelligent diagnosis module and a database.

[0022] The vital sign and behavior data acquisition module is connected to the vital sign data preprocessing module and the behavior data preprocessing module. The vital sign data preprocessing module, the vital sign abnormality analysis module, and the health intelligent diagnosis module are connected in sequence. The behavior data preprocessing module, the behavior abnormality identification module, and the health intelligent diagnosis module are connected in sequence. All modules in the system are connected to the database.

[0023] The vital sign and behavior data acquisition module collects the vital sign data of the livestock through multiple sensors configured on the smart collar and obtains the behavior data of the livestock through multi-angle cameras;

[0024] Specifically, in this embodiment, the smart collar used integrates multiple sensors, including a temperature sensor, a heart rate sensor, an accelerometer, and a respiratory rate monitoring device. These devices use high-precision sensors and advanced algorithm technology to accurately and in real time monitor and record the physiological data of livestock. In addition, the smart collar has built-in electronic tags and positioning chips. The electronic tags store the species name, number, age, and weight of the livestock, and the positioning chip is used to locate the real-time location of the livestock.

[0025] Furthermore, the vital signs and behavior data collection module includes a vital signs data collection unit, a behavior data collection unit and a data transmission unit. The vital signs data collection unit collects the surface temperature of the livestock in real time based on the temperature sensor on the smart collar, collects the heart rate of the livestock in real time through the heart rate sensor on the smart collar, collects the respiratory rate of the livestock in real time through the respiratory rate monitoring device on the smart collar, and collects the chewing data and livestock movement data of the livestock in real time through the acceleration sensor. The chewing data of the livestock is the Z-axis acceleration collected by the acceleration sensor fixed on the lower jaw of the livestock, and the livestock movement data is the X-axis acceleration collected by the acceleration sensor fixed on the smart collar. The acceleration sensor fixed on the livestock's lower jaw takes the livestock's lower jaw as the reference point, and the direction perpendicular to the livestock's head-tail axis and pointing to the side of the livestock's body is the Z axis. The acceleration sensor fixed on the smart collar takes the smart collar as the reference point, the direction of the livestock's movement is the X axis, and the vertical upward direction perpendicular to the X axis is the Y axis; the behavior data acquisition unit obtains the livestock's eating behavior images, exercise behavior images, and resting behavior images through a multi-angle camera; the data transmission unit synchronously transmits the collected livestock vital signs data to the vital signs data preprocessing module, and synchronously transmits the acquired livestock behavior images to the behavior data preprocessing module.

[0026] It should be specifically noted in this embodiment that body temperature is an important indicator reflecting the metabolic and immune status of livestock. Abnormal body temperature changes usually mean that there is infection, inflammation or other health problems in the livestock. By monitoring body temperature, fever symptoms of livestock can be detected in time, and it helps veterinarians to judge the type and severity of the disease and provide a basis for formulating treatment plans; movement speed is an important aspect of assessing the vitality and health of livestock. Livestock that are slow in movement or unable to move may be suffering from a disease or affected by other health problems. By monitoring movement speed, movement disorders of livestock can be detected in time; rumination is a digestive process unique to ruminants. A reduction or cessation of rumination time may mean that there is a problem with the livestock's digestive system, such as rumen food accumulation, acidosis, etc. By monitoring rumination time, digestion problems of livestock can be detected in time. Problem; Heart rate is an important indicator reflecting the health of the livestock's cardiovascular system. A heart rate that is too fast or too slow may mean that the livestock has cardiovascular problems or other health problems. By monitoring the heart rate, cardiovascular problems of the livestock, such as arrhythmia and heart failure, can be discovered in time; respiratory rate is an important indicator reflecting the health of the livestock's respiratory system. Abnormal respiratory rate may mean that the livestock has respiratory tract infection, pneumonia or other respiratory system problems. By monitoring the respiratory rate, respiratory problems of the livestock can be discovered in time, and the respiratory rate data can also help veterinarians evaluate the ventilation volume and oxygenation capacity of the livestock, and provide a basis for formulating treatment plans. Therefore, the real body temperature, movement speed, rumination time, heart rate and respiratory rate of the livestock are selected as the physical sign data for intelligent diagnosis of livestock health to determine whether the livestock has abnormal physical signs.

[0027] The vital sign data preprocessing module processes the livestock's body surface temperature, livestock chewing data, and livestock movement data to obtain the livestock's real body temperature, rumination time, movement state, and movement speed;

[0028] Furthermore, the vital sign data preprocessing module includes a data receiving unit, a body temperature correction unit, a rumination time calculation unit, a motion state analysis unit and a data output unit. The data receiving unit is used to receive livestock vital sign data; the body temperature correction unit inputs the received livestock surface temperature into the temperature linear regression equation for correction to obtain the livestock's true body temperature; the rumination time calculation unit uploads the received Z-axis acceleration data to MATLAB, uses the findpeaks function to extract the peak value of the Z-axis acceleration data to obtain local peaks and troughs, then sets positive and negative thresholds to extract peak data greater than the positive threshold and less than the negative threshold for statistics and mark as rumination, and the proportion of peak data that meets the requirements in all collected data is the livestock's rumination time; the motion state analysis unit uses the received X-axis acceleration and Y-axis acceleration as input, applies the K-means clustering algorithm to divide the livestock's motion state into three states: stillness, slow walking and running, and then calculates the real-time motion speed; the data output unit transmits the received livestock's heart rate, respiratory rate, true body temperature, rumination time and motion speed to the vital sign abnormality analysis module.

[0029] Specifically, in this embodiment, rectal temperature is typically used as the livestock's body temperature, typically measured manually using a veterinary thermometer. However, this method is labor-intensive, cannot obtain real-time data in a timely manner, and can easily spread disease. Therefore, an accurate and simple automated temperature measurement method is used to measure livestock's body temperature. The livestock's surface temperature, collected in real time by the temperature sensor on the smart collar, will have a certain error compared to the actual livestock's body temperature. However, because the collected data and the actual data are generally linearly related, the collected data can be corrected through linear regression analysis. The corrected predicted livestock body temperature is generally consistent with the actual livestock body temperature and can be directly used to represent the actual livestock body temperature.

[0030] Specifically, it should be noted that when constructing the temperature linear regression model, the least squares method can be applied to find the best matching function, and the linear regression model is defined as: , y, x, a0, b0 are the true body temperature value, the surface temperature value, the linear equation coefficient value, and the linear equation error respectively, then the expression of the average loss function Le is:

[0031] , N is the total number of control data sets of surface temperature and rectal temperature of the same livestock collected in the same environment and at the same time, i is the i-th control data set, yi is the rectal temperature in the i-th control data set, xi is the surface temperature in the i-th control data set, and the linear regression model corresponding to the minimum value of the average loss function Le is the best matching model. At this time, the partial derivatives of the average loss function Le with respect to a0 and b0 are both 0, that is:

[0032] , , after solving, we have:

[0033] , , retrieve the data in the control data group for simulation and establish a temperature linear regression model.

[0034] In this embodiment, it is specifically necessary to explain that in order to facilitate the understanding of the calculation rules of rumination time, a set of data is given for demonstration. Assume that the Z-axis acceleration after peak extraction is a1=10m / s 2 、a2=3m / s 2 、a3=-1m / s 2 、a4=-5m / s 2 、a5=2.5m / s 2 、a6=-3m / s 2 、a7=6m / s 2 、a8=4.5m / s 2 There are five sets of peak data and three sets of trough data. The positive and negative accelerations only represent the direction. The threshold is set to 5m / s. 2 and -5m / s 2 , then the acceleration is greater than 5m / s 2 and less than -5m / s 2 There are two groups of peak data, accounting for 25% of all the collected data. The collection time is 1 hour, and the rumination time is 15 minutes.

[0035] The behavior data preprocessing module uses image processing tools to extract images of the livestock's eating posture, eating movement, movement posture, and lying posture from the behavior data, identifies whether the extracted posture images are accurate, and updates the behavior data based on the identification results;

[0036] Furthermore, the behavior data preprocessing module includes a data receiving unit, a behavior extraction unit, an image recognition unit, a behavior data recording unit and a data output unit, wherein the data receiving unit is used to receive the acquired livestock behavior images; the behavior extraction unit uses image processing tools to extract the livestock's eating posture and eating movements from the eating behavior image video, extract the livestock's movement posture from the movement behavior image video, and extract the livestock's lying posture from the resting behavior image; the image recognition unit identifies whether the extracted posture image is correct based on the image recognition algorithm; the behavior data recording unit is used to record the cumulative duration of eating behavior, the cumulative duration of movement behavior, the cumulative duration of resting behavior, the duration of correct eating posture, the duration of correct eating movements, the duration of correct movement posture and the duration of correct lying posture; the data output unit transmits the recorded real-time cumulative duration of eating behavior, cumulative duration of movement behavior, cumulative duration of resting behavior, cumulative duration of correct eating posture, cumulative duration of correct eating movements, cumulative duration of correct movement posture and cumulative duration of correct lying posture to the behavior abnormality recognition module.

[0037] Specifically, in this embodiment, it should be noted that MultimodalVideoTag or VideoTag can be used to classify and extract livestock behavior data. The former is based on real short video business data and integrates three modalities: video text, image, and audio for video multimodal label classification. The model provides 25 first-level labels and more than 200 second-level labels, with a label accuracy rate exceeding 85%. The latter is based on tens of millions of short video business data from Baidu, supports 3,000 practical labels derived from industrial practice, has good generalization capabilities, and is very suitable for domestic large-scale (tens of millions, hundreds of millions, and billions) short video classification scenarios, with a label accuracy rate of 89%.

[0038] Specifically, in this embodiment, it should be noted that eating posture and movement can reflect the oral cavity, digestive system, and overall health of livestock. For example, an abnormal eating posture may indicate oral pain, digestive system disease, or other health issues. Movement posture can reveal the condition of the livestock's muscles, bones, nervous system, and overall mobility. Abnormal movement may indicate illness, pain, or movement disorders. For some livestock, such as dairy cows, lying posture is an important indicator for assessing their comfort, rest quality, and potential health issues. For example, reduced lying time in dairy cows may be related to heat stress, pain, or production stress. This behavioral data can often be easily acquired through video surveillance or on-site observation, without the need for complex equipment or technology. The recording and analysis of behavioral data is relatively intuitive, making it easier for veterinarians or livestock personnel to quickly identify potential health issues. Compared to traditional diagnostic methods such as blood tests and tissue biopsies, intelligent diagnosis based on behavioral data has the advantage of being non-invasive, reducing stress reactions and potential harm to livestock. By continuously monitoring and analyzing this behavioral data, potential diseases or health issues can be detected before clinical symptoms appear, allowing for timely intervention. Combining machine learning and artificial intelligence technologies allows for in-depth analysis and pattern recognition of these behavioral data, improving diagnostic accuracy and reliability. Therefore, feeding posture, eating movements, movement, and lying posture are selected as behavioral data for intelligent livestock health diagnosis. These data reflect health status, are easy to observe and record, and are non-invasive.

[0039] The abnormal vital signs analysis module inputs the livestock's heart rate, respiratory rate, actual body temperature, rumination time, and movement speed into a trained and tested livestock abnormal vital signs recognition model to perform abnormality recognition, mark the abnormal vital signs data group, and send the abnormal vital signs data group to the health intelligent diagnosis module after confirming that the livestock's vital signs are abnormal;

[0040] Furthermore, after receiving the livestock's heart rate, respiratory rate, true body temperature, rumination time, and movement speed, the physical sign abnormality analysis module inputs the received physical sign data into a trained and tested livestock physical sign abnormality recognition model for abnormality recognition, and outputs the abnormality determination results and abnormality coefficients of each physical sign parameter and the physical sign comprehensive abnormality index. The abnormality coefficient βxi of the i-th physical sign parameter is determined by the positive and negative relationship between the actual value βci of the i-th physical sign parameter received and the corresponding upper limit βsai and lower limit βsbi of the latest normal physical sign parameter fluctuation in the model. The specific formula is: The specific calculation formula of the comprehensive abnormality index αz of physical signs is: , mzy and mzx are the number of abnormal physical sign parameters and the number of normal physical sign parameters respectively. The calculated physical sign comprehensive abnormality index is then compared with the preset value of physical sign abnormality. If the calculated value is greater than the set value, the group of livestock physical sign data is marked as an abnormal data group. If abnormal data groups exceeding the preset number of groups appear continuously, the livestock physical signs are determined to be abnormal, and the abnormal livestock physical sign data group is sent to the health intelligent diagnosis module.

[0041] It should be specifically noted in this embodiment that the training data set and validation data set of the livestock abnormal sign recognition model used are normal physical sign data and abnormal physical sign data of several groups of livestock throughout their life cycle. The specific model construction process is existing technology and will not be elaborated here.

[0042] What needs to be specifically explained in this embodiment is that after the model training is completed, the normal standards for the vital sign parameters of livestock corresponding to the age, weight and environmental conditions will be generated based on the training data. The received livestock heart rate, respiratory rate, real body temperature, rumination time and movement speed will be input into the livestock vital sign abnormality recognition model that has been trained and passed the test. When performing abnormality recognition, the model will automatically compare the vital sign parameters of the received data with the normal standards for the vital sign parameters of livestock corresponding to the age, weight and environmental conditions. If the actual vital sign parameters meet the standards, the name of the vital sign parameter is output, followed by the normal indicator and the abnormality coefficient with a value of 0. If the actual vital sign parameters do not meet the standards, the name of the vital sign parameter is output, followed by the abnormality coefficient of the indicator being too high or too low and the specific value.

[0043] In this embodiment, it is specifically necessary to explain that, in order to facilitate understanding of the process scheme of abnormal physical sign analysis, a set of examples is provided: a one-year-old calf weighing 116 kg has a set of processed heart rate, respiratory rate, true body temperature, rumination time and slow walking speed of 70 beats / minute, 40 beats / minute, 32.7°C, 15 minutes, and 50 meters / minute in summer, respectively. The model generates a normal heart rate range, normal respiratory rate range, normal true body temperature range, normal rumination time range for a one-year-old calf weighing 100-120 kg in summer. The normal exercise speed intervals are [60 times / minute, 75 times / minute], [45 times / minute, 55 times / minute], [31.5℃, 32.5℃], [20min, 45min], and [10m / min, 100m / min]. The corresponding abnormal recognition results are: the heart rate index is normal, the heart rate abnormality coefficient is 0, the respiratory rate index is low, the respiratory rate abnormality coefficient is (45-40) / 45=0.1111, the real body temperature index is high, and the real body temperature abnormality coefficient is (32.7-32.5) / 32.5=0 0.0062, the rumination time index is low, the abnormality coefficient of rumination time is (20-15) / 20=0.2500, the slow walking speed index is normal, and the abnormality coefficient of slow walking speed is 0, then the calf's physical sign comprehensive abnormality index is 4 / 3×(0.0062+0.2500+0+0+0.1111)=0.4897, and then the calculated calf's physical sign comprehensive abnormality index is compared with the preset value. If the calculated value is greater than the preset value, the actual physical sign parameters of the calf in this group are marked as abnormal data group, otherwise only the abnormal physical sign parameters of the calf are marked. Continuous abnormality exceeds the preset value. If there are a set number of abnormal data groups, the livestock vital signs are judged to be abnormal, and the livestock vital signs abnormal data groups are sent to the health intelligent diagnosis module. If the abnormal preset value is 0, the calf example data group given is the abnormal data group. If the preset number of groups is 5, assuming that there are 5 groups of calf vital signs comprehensive abnormality indexes are 0.4987, 0.2573, 0.1986, 0.2471, and 0.3976 respectively, then these five groups of vital sign parameters are all abnormal data groups and trigger the calf vital signs abnormal judgment condition. The calf vital signs abnormality and the 5 groups of calf vital signs abnormal data groups are directly output and sent to the health intelligent diagnosis module.

[0044] The abnormal behavior recognition module calculates the correct posture ratio of eating behavior, exercise behavior and resting behavior, and then calculates the abnormality coefficient of each behavior, and then calculates the comprehensive abnormality index of the behavior, marks the abnormal image data group, and sends the abnormal livestock behavior image data group to the health intelligent diagnosis module after confirming that the livestock behavior is abnormal;

[0045] Furthermore, the abnormal behavior recognition module receives the real-time cumulative duration of eating behavior, cumulative duration of exercise behavior, cumulative duration of resting behavior, cumulative duration of correct eating posture, cumulative duration of correct eating action, cumulative duration of correct exercise posture, and cumulative duration of correct lying posture, and calculates the abnormality coefficient and the comprehensive abnormality index of behavior. The abnormality coefficient βyi of the i-th behavior type is determined by the positive and negative relationship between the difference between the received correct posture ratio βzi of the i-th behavior type and the set correct posture ratio βri. The specific formula is: The correct posture ratio of eating behavior is the product of the cumulative duration of correct eating posture, the cumulative duration of correct eating movements and the cumulative duration of eating behavior. The correct posture ratio of exercise behavior is the product of the cumulative duration of correct exercise posture and the cumulative duration of exercise behavior. The correct posture ratio of resting behavior is the cumulative duration of correct lying posture and the cumulative duration of resting behavior. The specific calculation formula of the comprehensive abnormal behavior index αx is: , mxa and mxb are the number of abnormal behavior types and the number of normal behavior types respectively. The calculated comprehensive abnormality index of behavior is then compared with the preset value of behavior abnormality. If the calculated value is greater than the set value, the existing incorrect posture images of various behaviors are extracted and marked as abnormal image data groups. If the marking operation occurs continuously for more than the preset number of times, the livestock behavior is determined to be abnormal, and the abnormal livestock behavior image data group is sent to the health intelligent diagnosis module.

[0046] The health intelligent diagnosis module inputs the received livestock physical sign abnormality data group and livestock behavior abnormality image data group as input into the trained and tested livestock health intelligent diagnosis model for health diagnosis, outputs the abnormal cause and corresponding treatment plan with the highest matching degree, and determines whether the model diagnosis result triggers a secondary diagnosis. If triggered, a secondary diagnosis is performed and the final output abnormal cause and treatment plan are determined based on the set rules. If not triggered, the abnormal cause and treatment plan output by the model diagnosis are directly output.

[0047] Furthermore, after receiving the livestock physical sign abnormality data group and the livestock behavior abnormality image data group, the health intelligent diagnosis module inputs the received livestock physical sign abnormality data group and livestock behavior abnormality image data group as input to the livestock health intelligent diagnosis model that has been trained and tested for health diagnosis. The model automatically matches the physical sign abnormality parameters and behavior abnormality images with the abnormal physical sign manifestations and abnormal behavior manifestations corresponding to different causes in the machine learning process. The ratio of the number of overlapping abnormal parameters to the actual number of abnormal parameters of the livestock is the matching degree. The abnormal cause with the highest matching degree and the corresponding treatment plan are output as the output end, and then the diagnosis result is judged, that is, the matching degree of the abnormal cause is compared with the set matching degree to determine whether it is reliable. Threshold, if the matching degree of the abnormal cause is greater than or equal to the set matching degree trust threshold, the abnormal cause with the highest matching degree and the corresponding treatment plan can be directly sent to the livestock user end. Otherwise, information is sent to multiple veterinarians registered in the system to request a secondary diagnosis. If the secondary diagnosis result is consistent with the model diagnosis result, the abnormal cause with the highest matching degree and the corresponding treatment plan are sent to the livestock user end. Otherwise, based on the same principle, the matching degree of the abnormal signs and abnormal behavior of the abnormal cause obtained by the secondary diagnosis with the abnormal signs parameters and abnormal behavior images of the livestock is calculated, and the matching degree of the abnormal cause obtained by the model diagnosis and the secondary diagnosis is compared. The abnormal cause and treatment plan corresponding to the one with the higher matching degree value is the final output abnormal cause and treatment plan.

[0048] It should be specifically noted in this embodiment that the learning data and verification data used in the livestock health intelligent diagnosis model are existing livestock disease types and abnormal manifestation data.

[0049] The database is used to store data information of all modules in the system.

[0050] It should be specifically noted in this embodiment that the preset values ​​or standard values ​​used are selected based on actual needs or data patterns, and are not limited to specific values ​​here.

[0051] like Figure 2 This embodiment provides an intelligent diagnosis method for animal husbandry and veterinary medicine based on machine learning, which specifically includes the following steps:

[0052] S1: The smart collar collects livestock vital signs data through multiple sensors and uses multi-angle cameras to obtain livestock behavioral data;

[0053] S2: Processing the livestock's body surface temperature, livestock chewing data, and livestock movement data to obtain the livestock's true body temperature, rumination time, movement state, and movement speed;

[0054] S3: Using image processing tools to extract images of the livestock's eating posture, eating action, movement posture, and lying posture from the behavioral data, and identifying whether the extracted posture images are accurate, and updating the behavioral data based on the identification results;

[0055] S4: inputting the livestock's heart rate, respiratory rate, actual body temperature, rumination time, and movement speed into a trained and tested livestock vital sign abnormality recognition model for abnormality recognition, marking a vital sign abnormality data group, and outputting the vital sign abnormality data group after confirming that the livestock's vital signs are abnormal;

[0056] S5: Calculate the correct posture ratios for eating, exercising, and resting behaviors, then calculate the abnormality coefficient for each behavior, and then calculate the comprehensive abnormality index for the behavior. Label the abnormal image data set, and output the abnormal behavior image data set after confirming that the livestock behavior is abnormal.

[0057] S6: The received livestock physical sign abnormality data group and livestock behavior abnormality image data group are input as input to the trained and tested livestock health intelligent diagnosis model for health diagnosis, and the abnormal cause and corresponding treatment plan with the highest matching degree are output. It is determined whether the model diagnosis result triggers a secondary diagnosis. If it is triggered, a secondary diagnosis is performed and the final output abnormal cause and treatment plan are determined based on the set rules. If it is not triggered, the abnormal cause and treatment plan output by the model diagnosis are directly output.

[0058] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A machine learning-based intelligent diagnostic system for animal husbandry and veterinary medicine, characterized by: include: Vital sign and behavior data collection module: collects livestock's vital sign data through multiple sensors configured on the smart collar, and obtains livestock's behavior data through multi-angle cameras; Vital sign data preprocessing module: processes livestock surface temperature, livestock chewing data and livestock movement data to obtain livestock's true body temperature, rumination time, movement status and movement speed; Behavior data preprocessing module: This module uses image processing tools to extract images of livestock's eating posture, eating motion, movement posture, and lying posture from the behavior data, identifies whether the extracted posture images are accurate, and updates the behavior data based on the identification results. Abnormal signs analysis module: The livestock's heart rate, respiratory rate, actual body temperature, rumination time, and movement speed are input into a trained and tested livestock abnormal signs recognition model to identify abnormalities, mark abnormal signs data groups, and send the abnormal signs data groups to the health intelligent diagnosis module after confirming that the livestock's physical signs are abnormal. After receiving the livestock's heart rate, respiratory rate, actual body temperature, rumination time, and movement speed, the physical sign abnormality analysis module inputs the received physical sign data into a trained and tested livestock physical sign abnormality recognition model for abnormality recognition, and outputs the abnormality determination results and abnormality coefficients of each physical sign parameter and a comprehensive physical sign abnormality index. The abnormality coefficient βxi of the i-th physical sign parameter is determined by the positive and negative relationship between the actual value βci of the i-th physical sign parameter received and the corresponding upper limit βsai and lower limit βsbi of the latest normal physical sign parameter fluctuation in the model. The specific formula is: ， The specific calculation formula of the comprehensive abnormality index αz of physical signs is: ， mzy and mzx are the number of abnormal and normal physical sign parameters, respectively. The calculated physical sign comprehensive abnormality index is then compared with the preset value of physical sign abnormality. If the calculated value is greater than the set value, the group of livestock physical sign data is marked as an abnormal data group. If the number of abnormal data groups exceeding the preset number appears continuously, the livestock physical sign is determined to be abnormal, and the abnormal livestock physical sign data group is sent to the health intelligent diagnosis module. Abnormal behavior recognition module: Calculates the proportion of correct postures for eating, exercising, and resting behaviors, then calculates the abnormality coefficient for each behavior, and then calculates the comprehensive abnormality index of the behavior. Abnormal image data sets are marked, and after confirming that the livestock behavior is abnormal, the abnormal livestock behavior image data sets are sent to the health intelligent diagnosis module; The abnormal behavior recognition module receives the real-time cumulative duration of eating behavior, cumulative duration of exercise behavior, cumulative duration of resting behavior, cumulative duration of correct eating posture, cumulative duration of correct eating action, cumulative duration of correct exercise posture, and cumulative duration of correct lying posture, and calculates the abnormality coefficient and the comprehensive abnormality index of behavior. The abnormality coefficient βyi of the i-th behavior type is determined by the positive and negative relationship between the received correct posture ratio βzi of the i-th behavior type and the set correct posture ratio βri. The specific formula is: ， The correct posture ratio of eating behavior is the product of the cumulative duration of correct eating posture, the cumulative duration of correct eating movements, and the cumulative duration of eating behavior. The correct posture ratio of exercise behavior is the product of the cumulative duration of correct exercise posture and the cumulative duration of exercise behavior. The correct posture ratio of resting behavior is the cumulative duration of correct lying posture and the cumulative duration of resting behavior. The specific calculation formula of the comprehensive abnormal behavior index αx is: ， mxa and mxb are the number of abnormal behavior types and the number of normal behavior types respectively. The calculated comprehensive abnormality index of behavior is then compared with the preset value of behavior abnormality. If the calculated value is greater than the set value, the existing incorrect posture images of various behaviors are extracted and marked as abnormal image data sets. If the marking operation occurs continuously for more than the preset number of times, the livestock behavior is determined to be abnormal, and the abnormal livestock behavior image data set is sent to the health intelligent diagnosis module; Intelligent health diagnosis module: The received livestock physical sign abnormality data group and livestock behavior abnormality image data group are input as input into the trained and tested livestock health intelligent diagnosis model for health diagnosis, and the abnormal cause and corresponding treatment plan with the highest matching degree are output. It is determined whether the model diagnosis result triggers a secondary diagnosis. If it is triggered, a secondary diagnosis is performed and the final output abnormal cause and treatment plan are determined based on the set rules. If it is not triggered, the abnormal cause and treatment plan output by the model diagnosis are directly output.

2. The animal husbandry and veterinary intelligent diagnostic system based on machine learning according to claim 1, characterized in that: The vital sign and behavior data acquisition module includes a vital sign data acquisition unit, a behavior data acquisition unit, and a data transmission unit. The vital sign data acquisition unit collects the livestock's body surface temperature in real time based on the temperature sensor on the smart collar, collects the livestock's heart rate in real time through the heart rate sensor on the smart collar, collects the livestock's respiratory rate in real time through the respiratory rate monitoring device on the smart collar, and collects the livestock's chewing data and livestock movement data in real time through the acceleration sensor. The livestock chewing data is the Z-axis acceleration collected by the acceleration sensor fixed on the livestock's lower jaw, and the livestock movement data is the X-axis acceleration and Y-axis acceleration collected by the acceleration sensor fixed on the smart collar. The acceleration sensor fixed on the livestock's lower jaw uses the livestock's lower jaw as a reference point, and the direction perpendicular to the livestock's head and tail axis and pointing to the side of the livestock's body is the Z-axis. The acceleration sensor fixed on the smart collar uses the smart collar as a reference point, and the direction of the livestock's forward movement when the livestock moves is the X-axis, and the direction perpendicular to the X-axis and vertically upward is the Y-axis. The behavior data acquisition unit uses a multi-angle camera to obtain images of the livestock's eating behavior, exercise behavior, and resting behavior. The data transmission unit synchronously transmits the collected livestock vital sign data to the vital sign data preprocessing module, and synchronously transmits the acquired livestock behavior images to the behavior data preprocessing module.

3. The animal husbandry and veterinary intelligent diagnosis system based on machine learning according to claim 1, characterized in that: The vital sign data preprocessing module includes a data receiving unit, a body temperature correction unit, a rumination time calculation unit, a motion state analysis unit and a data output unit, wherein the data receiving unit is used to receive the vital sign data of the livestock; The body temperature correction unit inputs the received livestock body surface temperature into the temperature linear regression equation to correct it and obtain the livestock's real body temperature; The rumination time calculation unit uploads the received Z-axis acceleration data to MATLAB, uses the findpeaks function to extract the peak value of the Z-axis acceleration data to obtain local peaks and troughs, then sets positive and negative thresholds to extract peak data greater than the positive threshold and less than the negative threshold for statistics and mark them as rumination. The proportion of peak data that meets the requirements in the total collected data is the rumination time of the livestock; The motion state analysis unit takes the received X-axis acceleration and Y-axis acceleration as input, applies the K-means clustering algorithm to classify the livestock's motion state into three states: still, walking, and running, and then calculates the real-time motion speed; The data output unit transmits the received livestock heart rate, respiratory rate, real body temperature, rumination time and movement speed to the abnormal vital signs analysis module.

4. The animal husbandry and veterinary intelligent diagnostic system based on machine learning according to claim 1, characterized in that: The behavior data preprocessing module includes a data receiving unit, a behavior extraction unit, an image recognition unit, a behavior data recording unit and a data output unit. The data receiving unit is used to receive the acquired livestock behavior images; the behavior extraction unit uses image processing tools to extract the livestock's eating posture and eating movements from the eating behavior image video, extract the livestock's movement posture from the movement behavior image video, and extract the livestock's lying posture from the resting behavior image; the image recognition unit identifies whether the extracted posture image is correct based on the image recognition algorithm; the behavior data recording unit is used to record the cumulative duration of eating behavior, the cumulative duration of movement behavior, the cumulative duration of resting behavior, the duration of correct eating posture, the duration of correct eating movements, the duration of correct movement posture and the duration of correct lying posture; the data output unit transmits the recorded real-time cumulative duration of eating behavior, cumulative duration of movement behavior, cumulative duration of resting behavior, cumulative duration of correct eating posture, cumulative duration of correct eating movements, cumulative duration of correct movement posture and cumulative duration of correct lying posture to the behavior abnormality recognition module.

5. The animal husbandry and veterinary intelligent diagnosis system based on machine learning according to claim 1, characterized in that: After receiving the livestock physical sign abnormality data group and the livestock behavior abnormality image data group, the health intelligent diagnosis module inputs the received livestock physical sign abnormality data group and the livestock behavior abnormality image data group as input to the livestock health intelligent diagnosis model that has been trained and tested for health diagnosis. The model automatically matches the abnormal physical sign parameters and abnormal behavior images with the abnormal physical sign manifestations and abnormal behavior manifestations corresponding to different causes in the machine learning process. The ratio of the number of overlapping abnormal parameters to the actual number of abnormal parameters of the livestock is the matching degree. The abnormal cause with the highest matching degree and the corresponding treatment plan are output as the output end, and then the diagnosis result is judged, that is, the abnormal cause matching degree is compared with the set matching degree credibility threshold. If the matching degree of the abnormal cause is greater than or equal to the set matching degree trust threshold, the abnormal cause with the highest matching degree and the corresponding treatment plan can be directly sent to the livestock user end. Otherwise, information is sent to multiple veterinarians registered in the system to request a secondary diagnosis. If the secondary diagnosis result is consistent with the model diagnosis result, the abnormal cause with the highest matching degree and the corresponding treatment plan will be sent to the livestock user end. Otherwise, based on the same principle, the matching degree of the abnormal signs and abnormal behavior of the abnormal cause obtained by the secondary diagnosis with the abnormal physical signs parameters and abnormal behavior images of the livestock is calculated, and the matching degree of the abnormal cause obtained by the model diagnosis and the secondary diagnosis is compared. The abnormal cause and treatment plan corresponding to the one with a higher matching degree value is the final output abnormal cause and treatment plan.

Images