scholarly journals Rule Discovery in Milk Content towards Mastitis Diagnosis: Dealing with Farm Heterogeneity over Multiple Years through Classification Based on Associations

Animals ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 1638
Author(s):  
Esmaeil Ebrahimie ◽  
Manijeh Mohammadi-Dehcheshmeh ◽  
Richard Laven ◽  
Kiro Petrovski
Keyword(s):  
Machine Learning ◽  
Early Detection ◽  
Milk Fat ◽  
Cost Effective ◽  
Subclinical Mastitis ◽  
Clinical Mastitis ◽  
Holstein Friesian ◽  
Milk Volume ◽  
First Time ◽  
Friesian Cows

Subclinical mastitis, an economically challenging disease of dairy cattle, is associated with an increased use of antimicrobials which reduces milk quantity and quality. It is more common than clinical mastitis and far more difficult to detect. Recently, much attention has been paid to the development of machine-learning expert systems for early detection of subclinical mastitis from milking features. However, differences between animals within a farm as well as between farms, particularly across multiple years, are major obstacles to the generalisation of machine learning models. Here, for the first time, we integrated scaling by quartiling with classification based on associations in a multi-year study to deal with farm heterogeneity by discovery of multiple patterns towards mastitis. The data were obtained from one farm comprising Holstein Friesian cows in Ongaonga, New Zealand, using an electronic automated monitoring system. The data collection was repeated annually over 3 consecutive years. Some discovered rules, such as when the milking peak flow is low, electrical conductivity (EC) of milk is low, milk lactose is low, milk fat is high, and milk volume is low, the cow has subclinical mastitis, reached high confidence (>70%) in multiple years. On averages, over 3 years, low level of milk lactose and high value of milk EC were part of 93% and 83.8% of all subclinical mastitis detecting rules, offering a reproducible pattern of subclinical mastitis detection. The scaled year-independent combinational rules provide an easy-to-apply and cost-effective machine-learning expert system for early detection of hidden mastitis using milking parameters.

scholarly journals L-selectin gene polymorphism and its association with clinical mastitis, somatic cell score, and milk production in Polish Holstein-Friesian cattle

2018 ◽  
Vol 63 (No. 7) ◽  
pp. 256-262 ◽  
Author(s):  
M. Dusza ◽  
J. Pokorska ◽  
J. Makulska ◽  
D. Kulaj ◽  
M. Cupial
Keyword(s):  
Somatic Cell ◽  
Milk Production ◽  
Milk Fat ◽  
Bacterial Infections ◽  
Clinical Mastitis ◽  
Somatic Cell Score ◽  
Production Traits ◽  
Milk Production Traits ◽  
Holstein Friesian ◽  
Friesian Cows

Bovine mastitis is a widespread disease of the mammary gland, highly contributing to the increase in veterinary costs in dairy industry. In the present study, the genetic polymorphism within bovine L-selectin gene was analysed and its impact on clinical mastitis occurrence, somatic cell score (SCS), and milk production traits in Polish Holstein-Friesian cows was examined. Polymorphism within L-selectin gene, molecule responsible for neutrophil attachment to endothelium, might have a potential role in immune response to bacterial infections and udder health. Two hundred and six Polish Holstein-Friesian cows were genotyped by polymerase chain reaction-restriction fragment length polymorphism method. Two single nucleotide polymorphisms mutations within the coding sequence of L-selectin gene were identified (c.165G>A and c.567C>T). The effect of c.165G>A and c.567C>T mutations on SCS was highly significant (P = 0.0019 and P = 0.0003, respectively). Strong associations (P ≤ 0.0001) were also observed between L-selectin polymorphism and milk production traits (milk yield, milk fat percentage, and milk protein percentage). However, the polymorphism in the analysed gene had no influence on the resistance or susceptibility of cows to clinical mastitis (only the tendency toward significance, P = 0.06 for c.567C>T mutation was found). Potential exploitation of the information on the identified associations in genetic selection needs to confirm the obtained results in further investigations.

scholarly journals Determination of Body Parts in Holstein Friesian Cows Comparing Neural Networks and k Nearest Neighbour Classification

Animals ◽  
2020 ◽  
Vol 11 (1) ◽  
pp. 50
Author(s):  
Jennifer Salau ◽  
Jan Henning Haas ◽  
Wolfgang Junge ◽  
Georg Thaller
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Body Parts ◽  
Nearest Neighbour ◽  
Data Set ◽  
3D Data ◽  
Holstein Friesian ◽  
Knn Classification ◽  
Friesian Cows

Machine learning methods have become increasingly important in animal science, and the success of an automated application using machine learning often depends on the right choice of method for the respective problem and data set. The recognition of objects in 3D data is still a widely studied topic and especially challenging when it comes to the partition of objects into predefined segments. In this study, two machine learning approaches were utilized for the recognition of body parts of dairy cows from 3D point clouds, i.e., sets of data points in space. The low cost off-the-shelf depth sensor Microsoft Kinect V1 has been used in various studies related to dairy cows. The 3D data were gathered from a multi-Kinect recording unit which was designed to record Holstein Friesian cows from both sides in free walking from three different camera positions. For the determination of the body parts head, rump, back, legs and udder, five properties of the pixels in the depth maps (row index, column index, depth value, variance, mean curvature) were used as features in the training data set. For each camera positions, a k nearest neighbour classifier and a neural network were trained and compared afterwards. Both methods showed small Hamming losses (between 0.007 and 0.027 for k nearest neighbour (kNN) classification and between 0.045 and 0.079 for neural networks) and could be considered successful regarding the classification of pixel to body parts. However, the kNN classifier was superior, reaching overall accuracies 0.888 to 0.976 varying with the camera position. Precision and recall values associated with individual body parts ranged from 0.84 to 1 and from 0.83 to 1, respectively. Once trained, kNN classification is at runtime prone to higher costs in terms of computational time and memory compared to the neural networks. The cost vs. accuracy ratio for each methodology needs to be taken into account in the decision of which method should be implemented in the application.

Effect of timing of protein supplementation on the performance of lactating dairy cows

2005 ◽  
Vol 45 (4) ◽  
pp. 337
Author(s):  
B. C. Granzin
Keyword(s):  
Dairy Cows ◽  
Milk Production ◽  
Milk Fat ◽  
Protein Supplementation ◽  
Net Energy ◽  
Maize Silage ◽  
Holstein Friesian ◽  
Grazing Dairy Cows ◽  
And Control ◽  
Friesian Cows

Two experiments were undertaken to determine the effect of timing of protein supplementation on performance of grazing, lactating Holstein–Friesian cows fed maize silage and grain-based concentrate. In experiment 1, 36 cows were fed 0.8 kg DM/day of solvent-extracted cottonseed meal (CSM) either as 1 meal at 1200 hours with maize silage (CSM 1200) or at 1530 hours with concentrate (CSM 1530), or in 2 meals at 0600 and 1530 hours with concentrate (CSM 600 + 1530). In experiment 2, 36 cows were either fed no CSM (control) or fed 1.0 kg DM/day as either CSM 1200 or CSM 600 + 1530. In experiment 1, daily yields of 4% fat-corrected milk (FCM) and milk fat for CSM 600 + 1530 were significantly higher than for CSM 1530 with respective means of 22.8 v. 20.7 L and 895 v. 804 g. Daily yields of FCM and milk fat for CSM 1200 were intermediate (21.7 L and 841 g/cow, respectively). A similar trend for daily protein yield per cow was noted (712, 695 and 666 g for CSM 600 + 1530, 1200 and 1530, respectively). In experiment 2, milk yield differed numerically between CSM 600 + 1530 and other treatments, with means (± s.e.d.) of 24.7 ± 0.78, 22.9 ± 0.78 and 22.9 ± 0.78 L/cow.day for CSM 600 + 1530, CSM 1200 and control, respectively. Mean (± s.e.d.) net energy requirements for milk production and liveweight change tended to be lower for the control (68 ± 3.6 MJ/cow.day) as opposed to CSM 600 + 1530 (79 ± 3.6 MJ/cow.day) and CSM 1200 (76 ± 3.6 MJ/cow.day). Cumulative time where rumen degradable nitrogen:rumen degradable dry matter was less than 22 g/kg were 2, 2 and 3 h for CSM 600 + 1530, CSM 1200 and CSM 1530, respectively, in experiment 1, and 6, 4 and 2 h for the control, 1200 and CSM 600 + 1530, respectively, in experiment 2. No differences in rumen ammonia-N concentrations were noted between treatments in experiment 1. In experiment 2, a significantly lower mean (± s.e.d.) rumen ammonia-N concentration was recorded for the control at 1530 hours (62 ± 14.1 mg/dL) in comparison to CSM 600 + 1530 (114 ± 14.1 mg/dL) and CSM 1200 (119 ± 14.1 mg/dL). These experiments show that for grazing dairy cows supplemented with maize silage and grain-based concentrate, feeding a daily aliquot of CSM as 2 meals at 0600 and 1530 hours rather than 1 meal at 1200 or 1530 hours improves milk production.

Using dairy herd improvement records and clinical mastitis history to identify subclinical mastitis infections at dry-off

2008 ◽  
Vol 75 (2) ◽  
pp. 240-247 ◽  
Author(s):  
Audrey H Torres ◽  
Päivi J Rajala-Schultz ◽  
Fred J DeGraves ◽  
Kent H Hoblet
Keyword(s):  
Dairy Herd ◽  
Cost Effective ◽  
Subclinical Mastitis ◽  
Clinical Mastitis ◽  
Somatic Cell Counts ◽  
Dry Cow Therapy ◽  
Cell Counts ◽  
On Farm ◽  
First 90 Days ◽  
Major Pathogens

Interest in selective dry cow therapy (SDCT) has been increasing owing to concerns over development of antimicrobial resistance. Implementation of SDCT, however, requires a quick and cost-effective on-farm method for identifying cows for treatment and cows that can be left without treatment. The objective of the present study was to evaluate the use of clinical mastitis (CM) history and somatic cell counts (SCC) from monthly Dairy Herd Improvement (DHI) records in identification of infected and uninfected cows at dry-off. A total of 647 Holstein cows were classified as uninfected or infected at dry-off based on CM history and varying number of monthly SCC records (with three different SCC cut-offs). Cows were considered uninfected based on the following criteria: (1) SCC <100 000 cells/ml and no CM during the lactation; (2) SCC <200 000 cells/ml and no CM during the lactation; (3) as criterion two, but additionally a cow was also considered uninfected if it experienced a case of CM during the first 3 months of the lactation and the SCC was <100 000 cells/ml for the rest of the lactation; (4) SCC <300 000 cells/ml and no CM during the lactation; otherwise they were considered infected. Infected and uninfected cows at dry-off were most efficiently identified using three months' SCC records with a threshold of 200 000 cells/ml for cows without CM during the lactation and a threshold of 100 000 cells/ml during the rest of lactation for cows with CM during the first 90 days in milk. Moreover, this criterion also most efficiently identified cows infected with major pathogens only at dry-off. The success of the criteria used for identifying infected and uninfected cows will, however, depend on herd characteristics, such as prevalence of infection and type of pathogens present in the herd.

Milk production by dairy cows fed legume grains or barley grain with or without urea as supplements to a cereal hay based diet

1990 ◽  
Vol 30 (1) ◽  
pp. 7 ◽  
Author(s):  
SC Valentine ◽  
BD Bartsch
Keyword(s):  
Milk Production ◽  
Faba Bean ◽  
Milk Fat ◽  
Dry Matter ◽  
Milk Protein ◽  
Barley Grain ◽  
Early Lactation ◽  
Milk Protein Content ◽  
Holstein Friesian ◽  
Friesian Cows

Milk production and composition was determined in Holstein-Friesian cows fed either 3.5 or 7.0 kg dry matter (DM) daily of lupin grain, pea grain, faba bean grain or barley grain with or without 1.5% added urea, as supplements to an oaten hay based diet. All the grains were hammermilled. Daily yields of milk (L), fat (kg) and protein (kg) were significantly (P<0.05) higher for cows fed lupin (20.0, 0.81, 0.57), pea (18.9, 0.80, 0.56) and faba bean (18.9, 0.79,0.55) grains compared with those of cows fed barley grain with (17.8, 0.73, 0.51) or without (18.0, 0.71, 0.52) urea. Yields of milk (L), fat (kg) and protein (kg) and milk protein content (g/kg) were significantly (P<0.05) higher when 7.0 kg DM (19.5, 0.80, 0.57, 29.6) compared with 3.5 kg DM (18.0,0.73,0.51,29.0) of grain was fed. There were no significant differences between treatments in hay DM intake by cows fed 3.5 kg DM of grain daily. A significantly (P<0.05) higher milk fat production per unit DM intake was recorded for cows fed legume compared with barley grain. There were generally higher returns above grain cost for cows fed legume compared with barley grain and for cows fed 4 kg compared to 8 kg daily of grain. It was concluded that it was more economical to feed 3.5 kg DM of legume grain compared with 3.5 kg DM of barley grain, with or without urea, as supplements for cows offered cereal hay in early lactation. However, the economics of feeding 7.0 kg DM of legume grain compared with 7.0 kg DM of barley grain and 7.0 kg DM compared with 3.5 kg DM of grain will depend on the costs of relative changes in hay intake associated with these practices.

scholarly journals Effect of clinical mastitis at Holstein-Friesian cows on the farm economic efficiency

2020 ◽  
Vol 12 (4) ◽  
pp. 375-378
Author(s):  
K. Stankov
Keyword(s):  
Economic Performance ◽  
Cost Efficiency ◽  
Milk Protein ◽  
High Capacity ◽  
Clinical Mastitis ◽  
Lactation Period ◽  
Purchase Price ◽  
The Third ◽  
Holstein Friesian ◽  
Friesian Cows

Abstract. A study was conducted on the effect of clinical mastitis at Holstein-Friesian cows on the economic performance of dairy farms with different capacity. The study included three high-capacity farms in Bulgaria. The cows in all three farms are free-stall reared (freely in groups) and fed total mixed rations (TMR), in accordance with the animals’ milk yields, with milking performed at a milking parlor. In each farm, the cows were separated into two groups – ones affected by clinical mastitis and healthy cows throughout the entire 305-day lactation period. The results from the study indicated that dairy capacity was high for both groups, with 7567.5 kg for the healthy cows during the 305-day period, and 7429.7 kg for the afflicted ones. The fat content of milk from healthy cows was 3.5% with 3.24% milk protein, with the values being 3.43% and 3.18%, respectively, for the afflicted cows. The cows with mastitis in the three studied farms exhibited lowered economic results. Profit from the healthy cows was higher, by 8.4% in the third farm up to 21.2% for the second farm, compared to the results for the affected cows. Cost-efficiency is also higher for the healthy cows, and the production cost of 1 kg of milk was lower, which was due to the higher marketing of milk for processing and the better purchase price. Apart from the worse economic performance of the cows affected by clinical mastitis, their productive longevity and total lifetime utilization registered at the time of culling was reduced from 8.2% for the second up to 12.6% for the first farm.

scholarly journals Endophyte toxins and performance of spring-calving dairy cows in Northland

1999 ◽  
Vol 7 ◽  
pp. 45-50
Author(s):  
M.B. Blackwell ◽  
R.G. Keogh
Keyword(s):  
Milk Production ◽  
Milk Samples ◽  
Holstein Friesian ◽  
Milk Volume ◽  
Group A ◽  
Lolitrem B ◽  
And Performance ◽  
Milk Solids ◽  
On Farm ◽  
Friesian Cows

An on-farm trial was carried out at Te Hana, in Northland, to measure milk production responses in 2 groups of 16 spring-calving, 3-year-old Holstein-Friesian cows maintained throughout the trial on pastures with or without the ryegrass endophyte toxins, ergovaline and lolitrem B. The trial began in October 1997, with second calving cows balanced for calving date and production worth. Milk volumes of all cows were recorded for 10 consecutive days each month followed by a herd test in which milk volume, protein, fat and lactose contents were determined from milk samples. There were no differences in milk volume or milk solids production in the October and November measurement periods. In December, the group grazing toxin-free (-Ev) pastures produced 24% more milk than the group grazing toxincontaining (+Ev) pastures. These differences increased progressively as the trial proceeded, until terminated in mid-April. Throughout the trial period the -Ev group produced 23% and 19% more milk and milk solids, respectively, than the +Ev group. A rise in levels of the endophyte toxins in ryegrass coincided with the start of differences in milking performance. A toxin-free maize based supplement was fed to both groups during the January to March period to maintain the trial when pasture growth was insufficient. A negative correlation was found between milk production of the +Ev group in January and the prevailing temperature and humidity conditions during the night. Keywords: endophyte toxins, ergovaline, lolitrem B, milk production, Neotyphodium, Northland, ryegrass

scholarly journals RELATIONSHIP BETWEEN SOMATIC CELL COUNT AND ZINC BLOOD SERUM CONCENTRATION IN DAIRY COWS

2012 ◽  
Vol 5 (1) ◽  
pp. 59-68
Author(s):  
Ivana Davidov ◽  
Miodrag Radinović ◽  
Mihajlo Erdeljan ◽  
Ivan Stančić ◽  
Dragica Stojanović ◽  
...  
Keyword(s):  
Blood Serum ◽  
Somatic Cell ◽  
Cell Count ◽  
Somatic Cell Count ◽  
Somatic Cells ◽  
Subclinical Mastitis ◽  
Holstein Friesian ◽  
Spring And Autumn Period ◽  
Experimental Group ◽  
Friesian Cows

The study was conducted on 15 Holstein-Friesian cows aged between 3 and 5 years kept in the free grazing conditions in summer. The samples of milk were taken twice a year for monitoring subclinical mastitis. In addition to milk samples, the blood samples were taken to determine the concentration of zinc in serum. The occurrence of subclinical mastitis in high yielding cows was estimated based on the values of zinc concentration in blood serum and somatic cells of milk. By analyzing the samples of blood serum, a variation in the concentration of zinc in blood serum was noted depending on the season. Somatic cell count in the spring and autumn period in average was over 400.000/ml milk. Based on statistical analysis, the correlation test, it was concluded that the concentration of zinc in blood serum of cows in the experimental group had no significant effect on the somatic cells count in milk.

Sign in / Sign up

Export Citation Format

Share Document