Comparison of machine learning methods to predict udder health status based on somatic cell counts in dairy cows.
Ontology highlight
ABSTRACT:
SUBMITTER: Bobbo T
PROVIDER: S-EPMC8249463 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Mastitis is an inflammatory disease, affects the dairy industry and has a severe economic impact. During subclinical mastitis, milk production and milk quality deteriorates. Recently, rumen microbial composition has been linked to rumen health, but few studies have investigated the effect of rumen microbiota on mammary health in cows. This study was undertaken to identify the rumen microbial composition and associated microbial fermentation in cows with different somatic cell counts (SCC), with the speculation that cows with different health statuses of the mammary gland have different rumen bacterial composition and diversity. A total of 319 Holstein dairy cows fed the same diet and under the same management were selected and divided into four groups as SCC1 (N = 175), SCC2 (N = 49), SCC3 (N = 49), and SCC4 (N = 46) with < 200,000, 200,001-500,000, 500,001-1,000,000, and >1,000,000 somatic cells/mL, respectively. Further, 20 cows with the lowest SCC and 20 cows with the highest SCC were identified. The rumen microbial composition was profiled using 16S rRNA sequencing, along with measurement of rumen fermentation parameters and milking performance. Compared to low SCC, cows with high SCC showed poorer milk yield, milk composition, and rumen volatile fatty acids concentration, but higher rumen bacterial diversity. Although the predominant rumen bacterial taxa did not vary among the SCC groups, the relative abundance of phyla SR1 and Actinobacteria, unclassified family Clostridiales and genus Butyrivibrio were significantly different. In addition, Proteobacteria and family Succinivibrionaceae were enriched in cows with low SCC. Our results suggest that specific rumen microbes are altered in cows with high SCC.
| S-EPMC6312127 | biostudies-literature
Project description:The early detection of health problems in dairy cattle is crucial to reduce economic losses. Mid-infrared (MIR) spectrometry has been used for identifying the composition of cow milk in routine tests. As such, it is a potential tool to detect diseases at an early stage. Partial least squares discriminant analysis (PLS-DA) has been widely applied to identify illness such as lameness by using MIR spectrometry data. However, this method suffers some limitations. In this study, a series of machine learning techniques-random forest, support vector machine, neural network (NN), convolutional neural network and ensemble models-were used to test the feasibility of identifying cow sickness from 1909 milk sample MIR spectra from Holstein-Friesian, Jersey and crossbreed cows under grazing conditions. PLS-DA was also performed to compare the results. The sick cow records had a time window of 21 days before and 7 days after the milk sample was analysed. NN showed a sensitivity of 61.74%, specificity of 97% and positive predicted value (PPV) of nearly 60%. Although the sensitivity of the PLS-DA was slightly higher than NN (65.6%), the specificity and PPV were lower (79.59% and 15.25%, respectively). This indicates that by using NN, it is possible to identify a health problem with a reasonable level of accuracy.
| S-EPMC8388516 | biostudies-literature
Project description:This study investigated 960 Slovak and Czech spotted cattle from four different conventional (non-organic) dairy herds located in Eastern Slovakia and Czechia during early lactation (14-100 days after calving). Dairy cows were examined clinically; milk from fore-stripping of each udder quarter was subjected to sensory examination and assessed by the California mastitis test (CMT), and laboratory analyses of bacterial pathogens in milk, including virulence factors, were conducted. Positive CMT scores (1-3) for one or more quarters were detected in 271 (28.2%) of the examined animals. Out of 230 infected milk samples, representing 24.0% of all dairy cows, staphylococci (59.1% of positive findings) were the most commonly isolated organisms, followed by E. coli (11.3%), streptococci Str. uberis (9.1%) and Str. agalactiae (3.4%), and enterococci (6.1%). From 136 isolates of S. aureus (38 isolates) and non-aureus staphylococci (NAS; 98 isolates), virulence factors and their resistance to 14 antimicrobials were detected using the disk diffusion method, with PCR detection of the methicillin resistance gene, mecA. An increased incidence of clinical and chronic forms of mastitis has been reported in mastitic cows in which staphylococci, especially S. aureus and NAS (S. chromogenes, S. warneri, and S. xylosus), have been detected and compared to other isolated udder pathogens. From those species, S. aureus and isolates of NAS mentioned above showed multiple virulence factors that are more likely to hydrolyze DNA, hemolysis, produce gelatinase and biofilm, and have multi-drug resistance as compared to other less virulent staphylococci. Generally, the isolated staphylococci showed 77.2% resistance to one or more antimicrobials, in particular to aminoglycosides, β-lactams, macrolides, or cephalosporins. Isolates that showed the ability to form a biofilm were more resistant to more than one antimicrobial than isolates without biofilm production. Multi-drug resistance to three or more antimicrobial classes was recorded in 16 isolates (11.7%), and the presence of the mecA gene was also confirmed in two isolates of S. aureus and two species of NAS.
| S-EPMC8868196 | biostudies-literature
Project description:Udder cleft dermatitis (UCD) is a skin condition affecting the fore udder attachment of dairy cows. UCD may be defined as mild (eczematous skin changes) or severe (open wounds, large skin changes). Our aims were to compare the microbiota of mild and severe UCD lesions with the microbiota of healthy skin from the fore udder attachment of control cows, and to investigate whether mastitis-causing pathogens are present in UCD lesions. Samples were obtained from cows in six dairy herds. In total, 36 UCD samples categorized as mild (n = 17) or severe (n = 19) and 13 control samples were sequenced using a shotgun metagenomic approach and the reads were taxonomically classified based on their k-mer content. The Wilcoxon rank sum test was used to compare the abundance of different taxa between different sample types, as well as to compare the bacterial diversity between samples. A high proportion of bacteria was seen in all samples. Control samples had a higher proportion of archaeal reads, whereas most samples had low proportions of fungi, protozoa and viruses. The bacterial microbiota differed between controls and mild and severe UCD samples in both composition and diversity. Subgroups of UCD samples were visible, characterized by increased proportion of one or a few bacterial genera or species, e.g. Corynebacterium, Staphylococcus, Brevibacterium luteolum, Trueperella pyogenes and Fusobacterium necrophorum. Bifidobacterium spp. were more common in controls compared to UCD samples. The bacterial diversity was higher in controls compared to UCD samples. Bacteria commonly associated with mastitis were uncommon. In conclusion, a dysbiosis of the microbiota of mild and severe UCD samples was seen, characterized by decreased diversity and an increased proportion of certain bacteria. There was no evidence of a specific pathogen causing UCD or that UCD lesions are important reservoirs for mastitis-causing bacteria.
| S-EPMC7710049 | biostudies-literature
Project description:Teat disinfection is a common pre- and post-milking mastitis prevention practice that is part of a mastitis control program in dairy herds. Commercially available teat disinfectants are generally chemical-based products. The use of these products has occasionally raised concerns about the risk of chemical residues in milk. An alternative treatment or prevention strategy based on probiotics has the potential to circumvent this risk. Two treatments were compared in a cross-over clinical trial in a single herd: a lactobacillus-based, post-milking teat spray (LACT), and a commercial iodine-based post-milking teat disinfectant product as (positive control, PC). The effect of the two treatments on cow somatic cell counts was quantified using a multivariate mixed-effects linear regression model with cow fitted as a random effect. The odds of teat end scores increasing from a low to a high score tended to be lower (OR = 0.74, 95% CI 0.54-1.01, P = 0.06) for cows receiving LACT treatment. On average, there was also a tendency for a lower somatic cell counts in the LACT treated cows (antilog of coefficient = 0.91, 95% CI 0.80-1.03, P = 0.13) compared with the PC treated cows. The application of the lactobacillus-based product to teats could reduce the rate of teat end scores progression from low to higher scores, and potentially improve teat end sphincter functions and udder health. Further, larger scale validation work is required to support the findings of the current study.
| S-EPMC7644449 | biostudies-literature
Project description:Intramammary infections (mastitis) of dairy cows, along with other periparturient diseases, have become problematic within the dairy industry as they lead to loss of milk production. The main objective of this study was to determine whether elevated somatic cell counts (SCC) in cows prior to drying off are related to the incidence of other periparturient diseases. Additionally, we determined whether milk composition and milk yield are affected by a high SCC prior to drying off. Somatic cell counts of milk samples were determined prior to dry off (n = 140) and were used to classify cows in the study as high (>200,000 cells/mL) or low (<200,000 cells/mL) SCC. The composition of milk was analyzed before drying off and at 1 and 2 weeks after calving. The results showed that an elevated SCC before drying off was related to the incidence of ketosis. Cows with a high SCC at drying off also showed an increased likelihood of retained placenta, metritis, and lameness postpartum; however, it was not statistically significant. Milk lactose was lower in cows with high SCC, whereas protein content was lower after parturition. Milk production was lower for cows with pre-drying elevated SCC, particularly for cows with retained placenta, ketosis, and mastitis. In conclusion, cows with pre-drying elevated SCC were more likely to develop disease after parturition and produce less milk and with lower lactose and protein content.
| S-EPMC9693442 | biostudies-literature
Project description:Complete blood counts (CBCs) measure the abundance of individual immune cells, red blood cells, and related measures such as platelets in circulating blood. These measures can indicate the health status of an animal; thus, baseline circulating levels in a healthy animal may be related to the productive life, resilience, and production efficiency of cattle. The objective of this study is to determine the heritability of CBC traits and identify genomic regions that are associated with CBC measurements in lactating Holstein dairy cattle. The heritability of CBCs was estimated using a Bayes C0 model. The study population consisted of 388 cows with genotypes at roughly 75,000 markers and 16 different CBC phenotypes taken at one to three time points (n = 33, 131, and 224 for 1, 2, and 3 time points, respectively). Heritabilities ranged from 0.00 ± 0.00 (red cell distribution width) to 0.68 ± 0.06 (lymphocytes). A total of 96 different 1-Mb windows were identified that explained more than 1% of the genetic variance for at least one CBC trait, with 10 windows explaining more than 1% of the genetic variance for two or more traits. Multiple genes in the identified regions have functions related to immune response, cell differentiation, anemia, and disease. Positional candidate genes include RAD52 motif-containing protein 1 (RDM1), which is correlated with the degree of immune infiltration of immune cells, and C-X-C motif chemokine ligand 12 (CXCL12), which is critically involved in neutrophil bone marrow storage and release regulation and enhances neutrophil migration. Since animal health directly impacts feed intake, understanding the genetics of CBCs may be useful in identifying more disease-resilient and feed-efficient dairy cattle. Identification of genes responsible for variation in CBCs will also help identify the variability in how dairy cattle defend against illness and injury.
| S-EPMC11004310 | biostudies-literature
Project description:BackgroundSomatic cell score (SCS) has been promoted as a selection criterion to improve mastitis resistance. However, SCS from healthy and infected animals may be considered as separate traits. Moreover, imperfect sensitivity and specificity could influence animals' classification and impact on estimated variance components. This study was aimed at: (1) estimating the heritability of bacteria negative SCS, bacteria positive SCS, and infection status, (2) estimating phenotypic and genetic correlations between bacteria negative and bacteria positive SCS, and the genetic correlation between bacteria negative SCS and infection status, and (3) evaluating the impact of imperfect diagnosis of infection on variance component estimates.MethodsData on SCS and udder infection status for 1,120 ewes were collected from four Valle del Belice flocks. The pedigree file included 1,603 animals. The SCS dataset was split according to whether animals were infected or not at the time of sampling. A repeatability test-day animal model was used to estimate genetic parameters for SCS traits and the heritability of infection status. The genetic correlation between bacteria negative SCS and infection status was estimated using an MCMC threshold model, implemented by Gibbs Sampling.ResultsThe heritability was 0.10 for bacteria negative SCS, 0.03 for bacteria positive SCS, and 0.09 for infection status, on the liability scale. The genetic correlation between bacteria negative and bacteria positive SCS was 0.62, suggesting that they may be genetically different traits. The genetic correlation between bacteria negative SCS and infection status was 0.51. We demonstrate that imperfect diagnosis of infection leads to underestimation of differences between bacteria negative and bacteria positive SCS, and we derive formulae to predict impacts on estimated genetic parameters.ConclusionsThe results suggest that bacteria negative and bacteria positive SCS are genetically different traits. A positive genetic correlation between bacteria negative SCS and liability to infection was found, suggesting that the approach of selecting animals for decreased SCS should help to reduce mastitis prevalence. However, the results show that imperfect diagnosis of infection has an impact on estimated genetic parameters, which may reduce the efficiency of selection strategies aiming at distinguishing between bacteria negative and bacteria positive SCS.
| S-EPMC2915952 | biostudies-literature
Project description:Mastitis is one of the most important infectious diseases and one of the diseases that causes the greatest use of antibiotics in dairy cows. Therefore, updated information on the bacteria that cause mastitis and their antibiotic susceptibility properties is important. Here, for the first time in over 10 years, we updated the bacterial findings in clinical mastitis in Swedish dairy cows together with their antibiotic resistance patterns and risk factors for each bacterial species. During the period 2013-2018, samples from clinical mastitis were collected, together with information on the cows and herds of origin. The samples were cultured, and a total of 664 recovered bacterial isolates were subjected to susceptibility testing. Staphylococcus aureus (S. aureus) was the most common pathogen and accounted for 27.8% of diagnoses, followed by Streptococcus dysgalactiae (S. dysgalactiae) (15.8%), Escherichia coli (E. coli) (15.1%), Streptococcus uberis (S. uberis) (11.4%), Trueperella pyogenes (T. pyogenes) (7.7%), non-aureus staphylococci (NAS) (2.8%), Klebsiella spp. (2.7%), Enterococcus spp. (1.3%), and Streptococcus agalactiae (S. agalactiae) (1.2%). Various other bacteria accounted for 2.6%. Staphylococci were, in general, susceptible to most antibiotics, but 2.6% of S. aureus and 30.4% of NAS were resistant to penicillin. No methicillin-resistant staphylococci were found. All S. agalactiae were susceptible to penicillin. Bimodal and trimodal MIC distributions for penicillin in S. dysgalactiae and S. uberis, respectively, indicate acquired reduced susceptibility in some isolates. The mostly unimodal MIC distributions of T. pyogenes indicate that acquired resistance does usually not occur in this species. Among E. coli, 14.7% were resistant to at least one antibiotic, most often ampicillin (8.7%), streptomycin (7.8%), or sulphamethoxazole (6.9%). Klebsiella spp. had low resistance to tetracycline (9.1%) but is considered intrinsically resistant to ampicillin. Pathogen-specific risk factors were investigated using multivariable models. Staphylococcus aureus, S. dysgalactiae, and T. pyogenes were more common, while E. coli was less common in quarters with more than one pathogen. S. aureus and T. pyogenes were mostly seen in early lactation, while E. coli was more common in peak to mid lactation and S. dysgalactiae in early to peak lactation. Trueperella pyogenes and Klebsiella spp. were associated with a previous case of clinical mastitis in the current lactation. Staphylococcus aureus was associated with tie stalls and T. pyogenes with loose housing. All pathogens except E. coli and S. dysgalactiae had a seasonal distribution. In conclusion, the aetiological agents for clinical bovine mastitis have remained relatively stable over the last 10-15 years, S. aureus, S. dysgalactiae, E. coli and S. uberis being the most important. Resistance to penicillin among Gram-positive agents was low, and in general, antibiotic resistance to other compounds was low among both Gram-positive and Gram-negative agents.
| S-EPMC8300163 | biostudies-literature
Project description:Milk production in dairy cow udders is a complex and dynamic physiological process that has resisted explanatory modelling thus far. The current standard model, Wood's model, is empirical in nature, represents yield in daily terms, and was published in 1967. Here, we have developed a dynamic and integrated explanatory model that describes milk yield at the scale of the milking session. Our approach allowed us to formally represent and mathematically relate biological features of known relevance while accounting for stochasticity and conditional elements in the form of explicit hypotheses, which could then be tested and validated using real-life data. Using an explanatory mathematical and biological model to explore a physiological process and pinpoint potential problems (i.e., "problem finding"), it is possible to filter out unimportant variables that can be ignored, retaining only those essential to generating the most realistic model possible. Such modelling efforts are multidisciplinary by necessity. It is also helpful downstream because model results can be compared with observed data, via parameter estimation using maximum likelihood and statistical testing using model residuals. The process in its entirety yields a coherent, robust, and thus repeatable, model.
| S-EPMC5567198 | biostudies-literature