CN111506881A - System for predicting Chinese Holstein cow mastitis onset risk - Google Patents

System for predicting Chinese Holstein cow mastitis onset risk Download PDF

Info

Publication number
CN111506881A
CN111506881A CN202010528636.9A CN202010528636A CN111506881A CN 111506881 A CN111506881 A CN 111506881A CN 202010528636 A CN202010528636 A CN 202010528636A CN 111506881 A CN111506881 A CN 111506881A
Authority
CN
China
Prior art keywords
mastitis
days
season
risk
predicting
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010528636.9A
Other languages
Chinese (zh)
Other versions
CN111506881B (en
Inventor
俞英
李文龙
史良玉
李锡智
肖炜
刘林
张毅
王雅春
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Agricultural University
Original Assignee
China Agricultural University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Agricultural University filed Critical China Agricultural University
Priority to CN202010528636.9A priority Critical patent/CN111506881B/en
Publication of CN111506881A publication Critical patent/CN111506881A/en
Application granted granted Critical
Publication of CN111506881B publication Critical patent/CN111506881B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Computational Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Public Health (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Algebra (AREA)
  • Biomedical Technology (AREA)
  • Operations Research (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Pathology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

The invention discloses a system for predicting the incidence risk of Chinese Holstein cow mastitis. The invention discloses a system for predicting Chinese Holstein cow mastitis onset risk, which comprises a parameter acquisition substance and a risk prediction module, wherein the acquired parameters are the field scale and season of a cow to be detected, the number of births, the lactation stage and SCC or SCS, and the risk prediction module can be used for predicting the mastitis onset risk. Experiments prove that the system for predicting the incidence risk of mastitis has high accuracy, high sensitivity, high specificity and simple and quick prediction, can realize the early discovery of cows with high incidence risk of mastitis, further prevent the cows from suffering from mastitis in production, has important guiding significance for the early detection and prevention of mastitis of cows, and is suitable for popularization and application.

Description

System for predicting Chinese Holstein cow mastitis onset risk
Technical Field
The invention relates to a system for predicting the incidence risk of Chinese Holstein cow mastitis in the field of animal medicine.
Background
According to the existence of macroscopic change of breast or milk, the mastitis of the dairy cow is divided into two types of clinical mastitis and recessive mastitis, the index for measuring the mastitis reaction at present is the number of somatic cells (SCC) of milk in a DHI measurement record, the DHI is a complete dairy cow production performance recording system aiming at the lactation performance and milk components of the dairy cow, and the current international judgment standard of the SCC of the milk is 10 ten thousand m L-1About 50 thousand m L-1Are not equal. To overcome the deficiencies of SCC in statistical analysis, it is usually converted into a form of body cell score (SCS) following a normal distribution.
There are a variety of models for predicting the risk of mastitis in dairy cows, among which L logistic regression, deep learning, random forests, etc. models have no significant difference in accuracy and predictive value.
Disclosure of Invention
The invention aims to solve the technical problem of predicting the incidence risk of the mastitis of the dairy cows, particularly Chinese Holstein cows.
The invention firstly provides a system for predicting or assisting in predicting the incidence risk of mastitis of a dairy cow, which comprises a parameter acquisition substance;
the parameter collecting substance is used for collecting the field scale and season, the birth times, the lactation stage and the milk somatic cell number (SCC) or Somatic Cell Score (SCS) of the dairy cow to be detected.
In the system, the parameter collecting substance may be a device and/or a reagent.
In the system, the SCS value can be 0-9.9. SCS value log2(SCC/100,000) + 3.
In the above system, the cow mastitis may be clinical mastitis or recessive mastitis of the cow.
The clinical mastitis can satisfy milk SCC of milk cow>50 ten thousand m L-1。
The recessive mastitis can satisfy milk cow milk 10 ten thousand m L-1< SCC ≦ 50 k.m L-1。
In the system, the system may further comprise a risk prediction module or a computing device provided with the risk prediction module, and the risk prediction module may be capable of predicting the incidence risk of mastitis of the dairy cow according to the size and season of the farm where the dairy cow to be detected is located, the number of births, the lactation period, and the SCC or SCS.
In the above system, the cow mastitis is clinical cow mastitis, and the risk prediction module may predict the risk of the cow clinical mastitis according to formula I:
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I);
wherein, PFacePredicting the probability for clinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is SCS, which is log2(SCC/100,000) + 3.
In the above system, the cow mastitis is cow subclinical mastitis, and the risk prediction module may predict the risk of cow subclinical mastitis according to formula II:
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II);
wherein, PConcealedPredicting probability for subclinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is SCS, which is log2(SCC/100,000) + 3.
The system may further include a readable carrier; the readable carrier can predict the clinical mastitis onset risk of the dairy cow according to the field scale and season, the birth and abortion, the lactation stage and SCC or SCS of the dairy cow to be detected, and can be recorded as the following formula I:
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I);
wherein, PFacePredicting the probability for clinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is SCS, which is log2(SCC/100,000) + 3.
The system may further include a readable carrier; the readable vector can predict the recessive mastitis onset risk of the dairy cow according to the field scale and season, the birth and fetal times, the lactation stage and SCC or SCS of the dairy cow to be detected, and the readable vector is described as the following formula II:
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II);
wherein, PConcealedPredicting probability for subclinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is SCS, which is log2(SCC/100,000) + 3.
In the above system, the parameter-collecting substance may include an apparatus for measuring SCC by microscopy, electron particle counting (Coulter counting) or fluorescence photoelectric counting, such as a somatic cell detector (e.g., a somatic cell detector of Foss, Denmark or Bentley, USA) for microscopy, Coulter counting or fluorescence photoelectric counting.
The system can be composed of the parameter acquisition substance only, the parameter acquisition substance and the risk prediction module, the parameter acquisition substance and the computing equipment provided with the risk prediction module, and the parameter acquisition substance and the readability carrier.
The invention also provides a use method of the system for predicting or assisting in predicting the incidence risk of the mastitis of the dairy cattle, wherein the method comprises the following steps:
1) collecting the scale and season of a cow to be detected, the number of births and births, the lactation stage and SCC or SCS;
2) calculating a clinical mastitis prediction probability according to formula I or calculating a recessive mastitis prediction probability according to formula II;
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I), PFacePredicting the probability for clinical mastitis;
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II), PConcealedPredicting probability for subclinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is SCS, which is log2(SCC/100,000) + 3.
The invention also provides application of the system for predicting or assisting in predicting the incidence risk of the dairy cow mastitis in preparing a product for predicting or assisting in predicting the incidence risk of the dairy cow mastitis.
The cow mastitis can be cow clinical mastitis or recessive mastitis.
In the present invention, the system may be a product. The product may be a kit.
The dairy cow may be a Chinese Holstein cow.
In the invention, the parameter acquired by the parameter acquisition substance can be the parameter of one month of the dairy cow, and the system can be used for predicting the incidence risk of mastitis in the next month.
PConcealedThe larger the value, the greater the risk of occult mastitis; pFaceThe larger the number, the greater the risk of clinical mastitis onset.
Experiments prove that the system for predicting or assisting in predicting the incidence risk of the mastitis of the dairy cattle can predict the incidence risk of the clinical mastitis and the incidence risk of the recessive mastitis, and has the advantages of high accuracy, high sensitivity, high specificity and simplicity and quickness in prediction. The method can realize the early discovery of the dairy cows with high mastitis incidence risk, further prevent the dairy cows from suffering from mastitis in production, has important guiding significance for the early detection and prevention of the mastitis of the dairy cows, and is suitable for popularization and application.
Drawings
FIG. 1 is a ROC curve for a recessive mastitis risk assessment model.
FIG. 2 is a ROC curve of a clinical mastitis risk assessment model.
In the figure, Model represents a mastitis risk assessment Model, Herd _ scale represents a field scale, Parity represents a fetal number, Test _ Season represents a measurement Season, DIM represents a lactation stage, and SCS _ This _ Month represents a SCS value in This Month.
Detailed Description
The invention provides a risk assessment system for predicting the incidence of Chinese Holstein cow mastitis, which is used for assessing and predicting the risk of mastitis in the next month of cows; the system comprises:
a factor analysis module: the method is used for selecting and analyzing risk indexes for predicting cow mastitis;
the calculation module is used for calculating the prediction probability according to the record of the risk indexes and an L logistic regression equation;
a prediction module: and predicting the incidence of mastitis of the dairy cow in the next month according to the prediction probability.
A factor analysis module, further comprising:
a factor selection submodule: for selecting a mastitis risk factor; mastitis risk factors include: field scale, fetal frequency, season, lactation stage, SCS in this month;
a classification factor judgment submodule: the risk assessment model is used for judging and assigning the form of the classified risk factors entering the risk assessment model; the factor forms include: the field scale is divided into small size, medium size (1) and large size (2); the number of the tires is divided into a first tire (1), a second tire (2) and a third tire (3); measuring seasons which are divided into summer (1) and non-summer (2); the lactation stage comprises 1-100 days (1), 101-200 days (2), 201-300 days (3) and more than 300 days of pregnancy (4);
and the factor analysis submodule is used for analyzing the factor forms which are more prone to occuring recessive mastitis and clinical mastitis in the risk factor forms, selecting field scale, fetal times, measuring seasons, lactation stages and the like as variables to substitute into a multi-factor L logistic regression equation, and determining the factor forms which are more prone to occuring recessive mastitis and clinical mastitis according to the OR value, wherein the OR value of the forms indicates that the risk forms are more prone to occuring recessive mastitis OR clinical mastitis.
In the calculation module, the recessive mastitis L logistic regression equation is as follows:
the L logistic regression equation for clinical mastitis is:
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I), PFacePredicting the probability for clinical mastitis; wherein x1 is field size, x2 is fetal number, x3 is season, x4 is lactation stage, and x5 is SCS value;
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II), PConcealedPredicting probability for subclinical mastitis; wherein x1 is field size, x2 is fetal number, x3 is season, x4 is lactation stage, and x5 is SCS value;
PconcealedThe larger the value, the greater the risk of occult mastitis; pFaceThe larger the number, the greater the risk of clinical mastitis onset.
The present invention is described in further detail below with reference to specific embodiments, which are given for the purpose of illustration only and are not intended to limit the scope of the invention. The examples provided below serve as a guide for further modifications by a person skilled in the art and do not constitute a limitation of the invention in any way.
The experimental procedures in the following examples, unless otherwise indicated, are conventional and are carried out according to the techniques or conditions described in the literature in the field or according to the instructions of the products. Materials, reagents, instruments and the like used in the following examples are commercially available unless otherwise specified. The quantitative tests in the following examples, all set up three replicates and the results averaged.
Example 1 construction of Risk assessment System for Breast inflammation of Chinese Holstein cattle
1. Milk cow production performance measurement record collection
Selecting DHI data (from the cow center of Beijing city) in Beijing area, covering 78,386 Chinese Holstein cows in 121 cattle farms, wherein the time is 1998-2016 years, and 1,967,310 DHI determination records are counted. The data content comprises individual number, fetal times, calving date, measuring date, milk production days, SCC and the like. Among them, SCC can be measured by microscopy, electron particle counting (coulter counter) or fluorescence photoelectric counting.
Recording of the obtained raw DHI: deleting the SCC-missing cattle records; deleting the SCC value of the previous 4 days to avoid deviation caused by overhigh SCC value of the previous 4 days; screening records that the same cow has DHI for two consecutive months, and dividing Chinese Holstein cows into a healthy group, a recessive mastitis group and a clinical mastitis group according to SCC in the next month; using the international conversion formula SCS log2(SCC/100,000) +3 the SCC value is converted into SCS, and the SCS is taken as a data set of 0-9.9.
Wherein the standard for judging mastitis is that the standard for judging the mastitis is that the SCC is less than or equal to 10 ten thousand m L in a healthy group-1Group of recessive mastitis, 10 Wan.m L-1< SCC ≦ 50 k.m L-1(ii) a Clinical mastitis group, SCC>50 ten thousand m L-1。
After culling, a total of 781,611 DHI records were obtained.
2. Selection of mastitis risk factor
Using the DHI data of Beijing area, using SAS 9.2 software to perform single factor analysis on each factor influencing the cow mastitis prediction, primarily screening the single factor, researching the influence of the single factor on the cow mastitis, and determining indexes with statistical significance among a healthy group, a recessive mastitis group and a clinical mastitis group, namely, the risk factors of the cow mastitis.
3. Formal determination and assignment of mastitis risk factor
And (3) analyzing by using the DHI data in Beijing area, calculating the logit (p) of the recessive mastitis group and the clinical mastitis group of the dairy cows, and drawing a line drawing to determine the form of entering the classification variables into the model and assigning values at the same time. The classification factor forms include: the field scale is divided into small size, medium size (1) and large size (2); the number of the tires is divided into a first tire (1), a second tire (2) and a third tire (3); measuring seasons which are divided into summer (1) and non-summer (2); the lactation stage comprises 1-100 days (1), 101-200 days (2), 201-300 days (3) and more than 300 days of pregnancy (4), and numbers in brackets are assigned to different indexes.
Wherein, the small and medium size of the farm scale means that the number of the lactating cows is less than or equal to 800, and the large size of the farm scale means that the number of the lactating cows is more than 800.
The first, second and third fetus of the gestation is the gestation of the cow.
Summer means 6, 7 and 8 months, and non-summer means non-6, 7 and 8 months.
The lactation stage refers to the time of lactation after the production of the dairy cow.
The results of the single factor analyses are shown in table 2 and fig. 2.
4. Analysis of single factor in recessive mastitis group and clinical mastitis group
The risk assessment of cow mastitis was performed using DHI data provided by the beijing dairy center. And (3) analyzing by using the field scale, the fetal times, the season measurement and the lactation stage as variables, and defining and assigning the factors in the same step.
The results of the single factor analysis are shown in table 1 and fig. 1.
Determining the more likely form of the factor for occurence of occult, clinical mastitis from the OR value; where the magnitude of the OR value indicates that the at risk form is more prone to occurence of occult OR clinical mastitis. Wherein, when the field scale is small or medium, the risks of recessive mastitis and clinical mastitis of the dairy cows are obviously increased; the risks of subclinical mastitis and clinical mastitis of the dairy cattle after the second and third births are obviously increased; the risk of subclinical mastitis and clinical mastitis is obviously increased in summer; risks of recessive mastitis and clinical mastitis are obviously increased after 101-200 days, 201-300 days and 300 days of lactation (namely more than 300 days).
TABLE 1 Single factor analysis of occult mastitis groups
Figure BDA0002534422860000071
Note:1is a reference category; p<0.001。
TABLE 2 analysis of clinical mastitis group factors
Figure BDA0002534422860000072
Note:1is a reference category; p<0.001。
5. Establishment of mastitis risk assessment model
Aiming at the data set screened in the step 1, establishing a model by adopting a non-conditional multifactorial L logistic regression analysis method by taking each risk index (field scale, fetal time, measuring season, lactation stage and SCS value) as an independent variable and mastitis diseased condition as a dependent variable, and respectively obtaining L logistic regression equations (namely mastitis risk assessment models) of recessive mastitis and clinical mastitis as follows:
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I), PFacePredicting the probability for clinical mastitis;
wherein x1 is field scale, x2 is number of births, x3 is season, x4 is lactation stage, and x5 is dairy cow production performance record SCS value in this month;
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II), PConcealedPredicting probability for subclinical mastitis;
wherein x1 is the field size, x2 is the number of births, x3 is the season, x4 is the lactation stage, and x5 is the SCS value of the month recorded for the production performance of the cow.
P threshold determination
The method for predicting recessive mastitis and clinical mastitis according to the P value obtained by the regression equation is as follows:
PconcealedThe larger the value, the greater the risk of occult mastitis; pFaceThe larger the number, the greater the risk of clinical mastitis onset.
In particular, as PConcealedIf the milk cow is more than 0.5, the next month of the milk cow to be detected suffers from recessive mastitis, such as PConcealedLess than or equal to 0.5, the dairy cow to be detected does not suffer from recessive mastitis in the next month; such as PFaceIf the milk cow is more than 0.5, the milk cow to be detected suffers from clinical mastitis in the next month, such as PFaceLess than or equal to 0.5, the dairy cow to be detected does not suffer from clinical mastitis in the next month.
Such as PFace> 0.5 and PConcealedIf the milk cow is more than 0.5, the milk cow to be detected suffers from clinical mastitis in the next month; such as PFaceLess than or equal to 0.5 and PConcealedIf the milk cow is more than 0.5, the next month of the milk cow to be detected suffers from recessive mastitis; such as PFaceLess than or equal to 0.5 and PConcealedLess than or equal to 0.5, the dairy cow to be detected does not suffer from clinical mastitis or recessive mastitis in the next month; not seen in PFace> 0.5 and PConcealedLess than or equal to 0.5.
7. Testing verification model using actual data
The accuracy of the above mastitis risk assessment model for predicting cow mastitis was evaluated using 781,611 DHI records from step 1.
The result shows that the accuracy of predicting recessive mastitis by using the recessive mastitis risk assessment model is 67.6%, the sensitivity is 50.7%, the specificity is 80.3%, and the area under the curve is 0.7214; the area under the curve of recessive mastitis is predicted to be 0.5130 by using the field scale; the area under the curve of the subclinical mastitis is predicted to be 0.5645 by using the fetal number; the area under the curve of recessive mastitis is predicted to be 0.5112 by using the measured season; predicting the area under the curve of recessive mastitis by using the lactation stage to be 0.5749; the area under the curve for predicting recessive mastitis by SCS was 0.7133. The prediction of the recessive mastitis by using the recessive mastitis risk assessment model is superior to each individual index.
The accuracy of predicting clinical mastitis by using the clinical mastitis risk assessment model is 83.6 percent, the sensitivity is 42.5 percent, the specificity is 94.9 percent, and the area under the curve is 0.8251; the area under the curve for predicting clinical mastitis using field scale is 0.5356; the area under the curve of clinical mastitis is predicted to be 0.6105 by using the fetal number; the area under the curve of clinical mastitis is predicted to be 0.5282 by using the measuring season; the area under the curve of clinical mastitis is predicted to be 0.5762 by utilizing the lactation stage; the area under the curve for clinical mastitis was predicted using SCS to be 0.8157. The clinical mastitis risk assessment model is used for predicting clinical mastitis to be superior to each individual index.
Example 2 measuring, recording, testing and testing the accuracy of the test model by using the production performance of the dairy cow
The accuracy of predicting cow mastitis by the mastitis risk assessment model of example 1 was evaluated using DHI data of chinese holstein cows in the pasture. The DHI data content comprises individual number, fetal number, calving date, measuring date, milk production related records, SCC and the like. DHI data was sourced from another exemplary cattle farm in beijing.
The DHI data is sorted by using R5.4.1 software, an L logistic regression model is applied to analyze each measurement index in the DHI data of the pasture, and the consistency between the monthly prediction condition and the actual condition of the dairy cow mastitis in the next month is compared to verify the accuracy of the model, wherein the incidence judgment standard of the dairy cow mastitis is that SCC is more than 50 ten thousand per milliliter.
The results are shown in Table 3, and the results show that the accuracy is over 65%.
TABLE 3 measurement, record, test and test model accuracy for dairy cow production performance in pasture
Month of the year 2 month 3 month 4 month Month 5 6 month Average
Number of heads of lactating cattle 1234 1339 1329 1263 1165 1266
Rate of accuracy 67％ 69％ 81％ 66％ 65％ 70％
The present invention has been described in detail above. It will be apparent to those skilled in the art that the invention can be practiced in a wide range of equivalent parameters, concentrations, and conditions without departing from the spirit and scope of the invention and without undue experimentation. While the invention has been described with reference to specific embodiments, it will be appreciated that the invention can be further modified. In general, this application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. The use of some of the essential features is possible within the scope of the claims attached below.

Claims (9)

1. A system for predicting or aiding in predicting the risk of developing mastitis in a dairy cow, comprising a parameter acquisition substance;
the parameter acquisition substance is used for acquiring the site scale and season, the birth and fetal times, the lactation stage and SCC or SCS of the dairy cow to be detected.
2. The system of claim 1, wherein: the system also comprises a risk prediction module or a computing device provided with the risk prediction module, wherein the risk prediction module can predict the incidence risk of the mastitis of the dairy cows according to the field size and season, the birth and gestation times, the lactation stage and the SCC or SCS of the dairy cows to be detected.
3. The system of claim 2, wherein: the cow mastitis is clinical cow mastitis, and the risk prediction module predicts the incidence risk of the cow clinical mastitis according to formula I:
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I);
wherein, PFacePredicting the probability for clinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is the SCS value.
4. The system of claim 2, wherein: the cow mastitis is cow recessive mastitis, and the risk prediction module predicts the incidence risk of the cow recessive mastitis according to a formula II:
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II);
wherein, PConcealedPredicting probability for subclinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is the SCS value.
5. The system of claim 1, wherein: the system further comprises a readability carrier;
the readable carrier can predict the clinical mastitis onset risk of the dairy cow according to the field scale and season, the birth and fetal times, the lactation stage and SCC or SCS of the dairy cow to be detected, and the readable carrier is described as the following formula I:
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I);
wherein, PFacePredicting the probability for clinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is the SCS value.
6. The system of claim 1, wherein: the system further comprises a readability carrier;
the readable vector can predict the recessive mastitis onset risk of the dairy cow according to the field scale and season, the birth and fetal times, the lactation stage and SCC or SCS of the dairy cow to be detected, and the readable vector is described as the following formula II:
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II);
wherein, PConcealedPredicting probability for subclinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is the SCS value.
7. The system according to any one of claims 1-6, wherein: the parameter acquisition material comprises an instrument for measuring SCC by using a microscopy method, an electron particle counting method or a fluorescence photoelectric counting method.
8. A method of using the system of any of claims 1-7, comprising:
1) collecting the scale and season of a cow to be detected, the number of births and births, the lactation stage and SCC or SCS;
2) calculating a clinical mastitis prediction probability according to formula I or calculating a recessive mastitis prediction probability according to formula II;
Logit Pface＝-4.3759-0.3926x1+0.4070x2+0.2340x3+0.2367x4+0.6596x5(formula I), PFacePredicting the probability for clinical mastitis;
Logit Pconcealed＝-2.3966-0.1683x1+0.2618x2+0.0879x3+0.2507x4+0.4432x5(formula II), PConcealedPredicting probability for subclinical mastitis;
x1 is the field scale, x1 is 1 when the number of lactating cows is less than or equal to 800, and x1 is 2 when the number of lactating cows is more than 800;
x2 is the number of births, and the values of x2 are 1, 2 and 3 when the number of births is one, two and three;
x3 is a measuring season, x3 takes a value of 1 when the measuring season is 6, 7 or 8 months, and x3 takes a value of 2 when the measuring season is not 6, 7 or 8 months;
x4 is a lactation stage, and the values of x4 are 1, 2, 3 and 4 when the lactation stage is 1-100 days, 101-200 days, 201-300 days and more than 300 days;
x5 is the SCS value.
9. Use of the system of any one of claims 1 to 7 for the manufacture of a product for predicting or aiding in the prediction of the risk of developing mastitis in a dairy cow.
CN202010528636.9A 2020-06-11 2020-06-11 System for predicting Chinese Holstein cow mastitis onset risk Active CN111506881B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010528636.9A CN111506881B (en) 2020-06-11 2020-06-11 System for predicting Chinese Holstein cow mastitis onset risk

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010528636.9A CN111506881B (en) 2020-06-11 2020-06-11 System for predicting Chinese Holstein cow mastitis onset risk

Publications (2)

Publication Number Publication Date
CN111506881A true CN111506881A (en) 2020-08-07
CN111506881B CN111506881B (en) 2022-11-04

Family

ID=71878826

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010528636.9A Active CN111506881B (en) 2020-06-11 2020-06-11 System for predicting Chinese Holstein cow mastitis onset risk

Country Status (1)

Country Link
CN (1) CN111506881B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105137011A (en) * 2015-07-23 2015-12-09 山东省农业科学院奶牛研究中心 System for screening dairy cow having recessive mastitis among cattle
CN105861657A (en) * 2016-04-11 2016-08-17 中国农业大学 Cow mastitis resistance related CNV fragment and applications thereof
CN106355033A (en) * 2016-09-27 2017-01-25 无锡金世纪国民体质与健康研究有限公司 Life risk assessment system
WO2017065708A1 (en) * 2015-10-12 2017-04-20 Nehir Biyoteknoloji Ar-Ge Hizm. Dan. Bils. Paz. San. Tic. Ltd. Sti On-line automatic subclinical mastitis detection device based on optical scattering and an automatic milk sampling system comprising this device
CN109711549A (en) * 2018-12-27 2019-05-03 中国农业大学 A kind of mastitis for milk cows detection method based on genetic algorithm optimization BP neural network
CN110699448A (en) * 2019-11-18 2020-01-17 中国农业大学 Method for detecting cow secondary nipple occurrence risk based on LGR5 gene polymorphism

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105137011A (en) * 2015-07-23 2015-12-09 山东省农业科学院奶牛研究中心 System for screening dairy cow having recessive mastitis among cattle
WO2017065708A1 (en) * 2015-10-12 2017-04-20 Nehir Biyoteknoloji Ar-Ge Hizm. Dan. Bils. Paz. San. Tic. Ltd. Sti On-line automatic subclinical mastitis detection device based on optical scattering and an automatic milk sampling system comprising this device
CN105861657A (en) * 2016-04-11 2016-08-17 中国农业大学 Cow mastitis resistance related CNV fragment and applications thereof
CN106355033A (en) * 2016-09-27 2017-01-25 无锡金世纪国民体质与健康研究有限公司 Life risk assessment system
CN109711549A (en) * 2018-12-27 2019-05-03 中国农业大学 A kind of mastitis for milk cows detection method based on genetic algorithm optimization BP neural network
CN110699448A (en) * 2019-11-18 2020-01-17 中国农业大学 Method for detecting cow secondary nipple occurrence risk based on LGR5 gene polymorphism

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
史良玉 等: "中国荷斯坦牛***炎风险评估模型建立及预测分析", 《中国畜牧杂志》 *
张瑞强 等: "京津冀地区中小规模奶牛场隐性***炎状况调查", 《中国奶牛》 *
李博通 等: "河北某牛场奶牛***炎的防治", 《今日畜牧兽医》 *

Also Published As

Publication number Publication date
CN111506881B (en) 2022-11-04

Similar Documents

Publication Publication Date Title
Benedet et al. Invited review: β-hydroxybutyrate concentration in blood and milk and its associations with cow performance
Krogh et al. Latent class evaluation of a milk test, a urine test, and the fat-to-protein percentage ratio in milk to diagnose ketosis in dairy cows
Sherman et al. Toward elimination of mother–to–child transmission of HIV in South Africa: how best to monitor early infant infections within the Prevention of Mother–to–Child Transmission Program
CN111539308B (en) Embryo quality comprehensive evaluation device based on deep learning
CN108717867A (en) Disease forecasting method for establishing model and device based on Gradient Iteration tree
CN109072479A (en) Spontaneous pre-term risk is layered using circulation particle
Brouwer et al. The validity of a monitoring system based on routinely collected dairy cattle health data relative to a standardized herd check
Farschtschi et al. Development of an advanced flow cytometry based high-resolution immunophenotyping method to benchmark early immune response in dairy cows
LU101772B1 (en) Non-invasive biomarkers for identification/prewarning of dairy cows with fatty liver disease
Pakrashi et al. Early detection of subclinical mastitis in lactating dairy cows using cow-level features
CN111506881B (en) System for predicting Chinese Holstein cow mastitis onset risk
CN111261298B (en) Medical data quality prejudging method and device, readable medium and electronic equipment
CN113539473A (en) Method and system for diagnosing brucellosis only by using blood routine test data
Jaques et al. Follow up and evaluation of the Victorian first‐trimester combined screening programme for Down syndrome and trisomy 18
Rodriguez et al. Exploring the role of milk yield in the first week of lactation on the association between hyperketonemia and reproductive performance in dairy cattle
EP2809798B1 (en) Method and apparatus for detecting cancer in mammals
CN105137011B (en) System for screening dairy cow having recessive mastitis among cattle
CN112557283A (en) Diagnosis and treatment immune marker for recurrent pregnancy loss diseases and application thereof
CN113469227A (en) Forced expiration total amount prediction method
CN113066584A (en) Prediction method and system for early septicemia
De Jong et al. Milk β-hydroxybutyrate metrics and its consequences for surveillance of hyperketonaemia on commercial dairy farms
Van Loo et al. Enhancing bovine abortion surveillance: A learning experience
Andrade et al. Identifying pregnancy in cows using ovarian ultrasound images and convolutional neural networks-a proof-of-concept study
CN117906726B (en) Abnormal detection system for weight data of live cattle body ruler
CN116741384B (en) Bedside care-based severe acute pancreatitis clinical data management method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant