Diagnosis of cases of Mycoplasma mastitis is particularly challenging due to their unique biological characteristics, which complicate diagnosis and treatment. Hence, accurate and quick diagnostic tests for early detection of Mycoplasma mastitis are essential to initiate appropriate interventions or culling. The objective of this research is to estimate the diagnostic performance of the molecular microarray assay (MMA) against bacterial culture for the diagnosis of bovine intramammary infections (IMI) with Mycoplasma spp., using a gold standard approach and the Kappa agreement coefficient. A total of 395 quarter milk samples were collected from cows in 31 dairy herds with conventional milking systems in California, USA. Following dairy personnel practices, milk samples were collected from the lactating cows showing abnormal milk characteristics and shipped within 24 hours to the laboratory for bacterial culture and MMA examination. Milk samples with positive growth were confirmed via PCR test to eliminate misdiagnosis of Acholeplasma spp. Eighty-seven cows (22%) were positive for Mycoplasma spp. IMI and the test accuracy was 88.4%. The sensitivity of MMA was 90.8% (95% CI (Confidence Interval): 82.68–95.95), and the specificity was 87.66% (95% CI: 83.46–91.12). The positive predictive value of MMA in these herds was 67.52% (95% CI: 60.51–73.83), and the negative predictive value was 97.12% (95% CI: 94.57–98.49). Calculated Kappa coefficient was 0.70 (95% CI: 0.618–0.778). The high estimates of sensitivity and specificity of MMA suggest its usefulness as a routine and quick test for accurate diagnosis of Mycoplasma spp. IMI in dairy cows. Our findings indicate that MMA holds promise for enhancing the detection of Mycoplasma spp. and could potentially revolutionize diagnostic practices in the dairy industry and supports udder health management.

Changes in waxed dry cheese during the ripening process, over periods of 7 and 30 days, were analysed using near-infrared spectroscopy (FT-NIR) and mid-infrared spectroscopy (FT-MIR) by attenuated total reflection (ATR). FT-NIR was employed to determine the proximate composition of the cheese (protein, fat, moisture, total solids, and salt content), identifying changes directly associated with the ripening process. FT-MIR data were used to identify spectral bands associated with chemical changes occurring during the cheese maturation. Additionally, chemometric techniques were applied to demonstrate the potential of FT-MIR infrared spectroscopy for cheese differentiation and fingerprint profiling. Subsequently, partial least squares discriminant analysis (PLS-DA) of the FT-MIR spectra was performed, revealing two distinct clusters representing the cheese ripening times. Functional groups related to lipids (–CH2 – and – CH3), proteins (amide bands I and II), and carbohydrates (C–O) were identified, correlating to lipolysis, proteolysis, and lactose catabolism. Infrared spectroscopy in combination with chemometric methods proved to be a robust and reliable tool for monitoring changes during the ripening of waxed dry cheese. The results obtained highlight its usefulness as an alternative approach for the analysis and fingerprinting of traditional Mexican foods, aiming to add value to local products.

Improving udder health on dairy farms requires knowledge about the prevailing mastitis pathogens in order to take appropriate measures. The aim of this research communication was to evaluate the association between two sampling approaches for determining the prevalence of mastitis pathogens in dairy herds. Sampling approaches tested included (a) bacteriological investigation of randomly selected cows independent from stage of lactation (random sampling) and (b) sampling of cows two weeks prior to drying off (dry-off sampling). Using linear regression, the prevalence of mastitis pathogens were compared on herd-level for groups of specific pathogens. Associations between the prevalence estimated by the two approaches were found for Staphylococcus aureus, Streptococcus dysgalactiae, Streptococcus uberis and Gram-negative bacteria but not for other esculin-positive streptococci. This study indicated that both sampling approaches provide farmers with an overview of the prevalence of mastitis pathogens in their herds, with the dry-off results also being used to target antibiotic use to infected quarters.

Buffalo are shy breeders and poor fertility traits are a major hindrance in exploiting the production potential of the animal. This study hypothesizes that polymorphisms in the luteinizing hormone beta (LHβ) gene can affect oestrus behaviour in buffaloes. A total of 100 animals were screened by calculating the heat index (threshold-50) and animals were categorized into two groups (Group1 > 50, Group2 < 50). Animals were subjected to blood sampling, genomic DNA isolation, specific primer based polymerization and sequencing of amplicons. A total of six genomic variations were identified in the gene. c.V15M was a non synonymous mutation found in line with the Hardy Weinberg Equilibrium and was significantly associated with the trait. Functional impact of the variation was determined by three-dimensional structure of the protein. Effect of c.V15M on the functionality of the gene was evident and hypothesis was supported so this can potentially be used as a marker for the future development of superior animal breed or regulating the expression of the gene to get the optimal oestrus cyclicity in river buffalo of Pakistan.

In this Research Communication we describe the application of single-step Bayesian regression (ssBR) models to predict milk yield of Murrah buffaloes. Milk production records of 2,026 cows in their first lactation were used. Using 270-day cumulative milk yield records as phenotype, genomic breeding values were predicted and their accuracies and dispersions were calculated by five methods: BayesA (ssBA), BayesB (ssBB), BayesC (ssBC); Bayesian Lasso (ssBL); and Bayesian ridge regression (ssBRR). For models based on mixture distributions (ssBB and ssBC), the proportions of markers having effect (π) were assumed as fixed, with respective values of 99% or 90%, or as unknown, where two approaches to estimate π were applied (ssBayesBπ and ssBayesCπ). The accuracy values found ranged from 0.550 (ssBBπ) to 0.584 (ssBCπ) and, the dispersion estimates ranged from 0.867 (ssBA) to 0.958 (ssBRR). The results indicated that Bayesian Lasso was the most suitable model for genetic evaluation of milk yield by buffaloes, considering accuracy and dispersion as criteria.

Colostrum (CM) and transition milk (TM) are crucial for neonatal development but knowledge about the changes in their composition in dairy goats is still limited. The aim of this study was to characterise the immunological, lipid and microbial changes from CM to TM in Guanzhong dairy goats, an indigenous, highly productive Chinese breed. Milk samples were collected from eight primiparous goats at day 0 (CM) and day 7 (TM) postpartum. Immunoglobulin concentrations (IgG, IgA, IgM) were measured by ELISA, long-chain saturated fatty acids (FA) were analysed by gas chromatography, and bacterial communities were profiled by 16S rRNA gene amplicon sequencing. Microbial functions were inferred using PICRUSt2 (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) software based on KEGG (Kyoto Encyclopedia of Genes and Genomes) orthologs. Immunoglobulin concentrations declined significantly from CM to TM, reflecting a transition from immunoprotection to food supply. The TM had increased concentrations of long-chain saturated FA, particularly palmitic acid (C16:0) and stearic acid (C18:0) and higher concentrations of branched-chain FA, suggesting an adaptation to the growing energy demands of the newborn. Although overall microbial diversity was stable, taxonomic shifts were observed. CM was enriched in Emticicia, which may be involved in oligosaccharide degradation, while TM had higher abundance of Faecalibacterium, Ruminococcus and Salinicoccus, genera associated with SCFA production and intestinal development. Functional predictions indicated enrichment of cold shock and haem biosynthetic pathways in CM and enhanced carbohydrate degradation pathways in TM. Correlation analyses revealed relationships between specific FA and bacterial genera, suggesting interactions between host, microbes and nutrients that may influence milk functionality. This integrative analysis of milk components during early lactation provides new insights into the compositional and functional dynamics of goat milk.

Mastitis is a major health problem in dairy industry as well as a major threat to profitability of dairy farms. Mastitis is also the main reason for the application of antibiotic treatment during lactation or at dry period. The aim of our study was to determine the prevalence of the most common mastitis pathogens in dairy cows and the antibiotic resistance under the conditions of Slovak dairy farms in 2017–2023. The samples came from 52 samplings in 2017 (47 farms), from 32 samplings in 2018 (29 farms), from 31 samplings in 2019 (29 farms), from 44 samplings in 2020 (41 farms), from 40 samplings in 2021 (35 farms), from 33 samplings in 2022 (31 farms) and form 38 samplings in 2023 (35 farms). A total of 2236 quarter udder milk samples were collected. The milk samples were taken from dairy cows based on high somatic cell count or California mastitis test or visible abnormalities in milk. Up to 88.62% of the identified isolates were the Coagulase-negative staphylococci (36.89%) followed by Escherichia coli (24.26%), Streptococcus uberis (16.21%), Staphylococcus aureus (8.41%) and Streptococcus agalactiae (2.86%). The most effective antibiotic was amoxicillin/clavulanic acid and antibiotic with the highest resistance was streptomycin. In conclusion, identification of mastitis pathogens in dairy cows and detection of antibiotic resistance is very important for the mastitis treatment and prevention of antibiotic resistance.

Sanitation Standard Operating Procedures (SSOP) are critical in key stages of food production and processing. After manufacturing, slicing process can serve as a point of contamination, potentially compromising the quality and shelf life of mozzarella. The objective of this study was to determine the effect of SSOP on the quantification and diversity of psychrotrophic bacteria with proteolytic and lipolytic potential in mozzarella before and after industrial slicing. Psychrotrophic bacteria were isolated, phenotypically assessed for spoilage potential under mesophilic and psychrotrophic conditions, analysed for diversity using dendrograms of genetic similarity and identified by partial sequencing of the 16S rRNA gene. The mean psychrotrophic counts were 3.77 (±0.83) log CFU/mL before slicing and 3.58 (±0.51) log CFU/mL in the sliced product, indicating a non-significant reduction (p < 0.05). Regarding spoilage potential, none of the 233 isolates evaluated exhibited proteolytic activity under psychrotrophic conditions. However, psychrotrophic lipolytic activity was predominant both before and after slicing. The species Lactobacillus delbrueckii, which is part of the saccharolytic inoculum used to reduce the pH of the curd during cheese production, was the main proteolytic bacteria under mesophilic conditions (35°C) in both before and after sliced samples. Although the bacterial counts indicated the full efficiency of the slicer’s SSOP, the microbial diversity analysis revealed the inclusion of Staphylococcus succinus, Staphylococcus hominis, Enterococcus faecalis and Klebsiella pneumoniae during the slicing process, albeit at low levels. Therefore, relying solely on psychrotrophic quantification may not be sufficient to attest the efficiency of the slicer’s SSOP. Even under controlled industrial conditions, spoilage bacteria from handling and environmental sources may be introduced into sliced mozzarella. Methods for improving the microbiological quality of the mozzarella pieces prior to slicing, as well as the intensification of sanitary procedures, must be reviewed and implemented to improve the shelf life and commercial potential of sliced mozzarella.

This study evaluated the effects of replacing ground corn with cassava root silage (CRS) in the supplement of grazing dairy cows on production yield, physicochemical characteristics, sensory attributes, and profitability of Minas Frescal cheese. Ten primiparous Girolando cows, with a mean weight of 373.45 ± 63.55 kg, a mean milk production of 12.48 ± 1.58 kg/d, and 76 days of lactation, were distributed into two 5 × 5 Latin squares. The animals were placed in the following five treatments: I, grazing without supplementation (WOS); II to V, grazing receiving 5 kg of dry matter (DM) of supplement without CRS (0 g/kg DM CRS) and with 260, 520 and 780 g/kg DM of CRS. Inclusion level of CRS did not affect (P > 0.067) physicochemical characteristics, sensory attributes, and production yield of cheese. However, cheese produced from supplemented animals had greater levels of protein (P = 0.025) and individual cheese production (kg of cheese/animal/day; P < 0.001) compared to WOS animals. Finally, the inclusion of CRS at up to 520 g/kg DM maximized cheese production by 0.73 kg of cheese/animal/day and gross revenue by 3.49 US$/animal/day, compared to WOS animals. In conclusion, replacement of ground corn with CRS in the supplement of dairy cows did not impact physicochemical characteristics and sensory attributes of Minas Frescal cheese. In addition, inclusion of CRS at up to 520 g/kg DM replacing ground corn in the supplement may be a suitable strategy for enhancing the profitability of Minas Frescal cheese production.

Follicle-stimulating hormone (FSH) must be applied at 12-h intervals over 4–5 days in the traditional cattle superovulation protocol, which still needs to be improved. This research paper evaluated the superovulation results obtained by a traditional protocol or by a single administration of FSH dissolved in MontanideTM ISA-206 VG (MonISA-206). Control cows were superovulated with 10 mL of FSH (500 µg pFSH + 100 µg pLH) from day 7 to day 10 (for 4 days, twice daily i.m. injections, decreasing doses). Cows in the EG10 and EG7.5 groups were injected i.m. with 20 mL (100%, 10 mL + 10 mL) or 15 mL (75%, 7.5 mL + 7.5 mL) of the FSH and MonISA-206 mixture at once on day 7. All cows were inseminated 12 and 24 h after oestrus onset. The cows presented no pathology at the injection sites. Plasma FSH levels differed between the groups, but the interaction between hour and group × time was not different. Superstimulation and embryo quality results were similar between the groups. A single injection of FSH (both 100% and 75% doses) dissolved in MonISA-206 led to adequate plasma FSH levels and similar superovulation results to traditional FSH treatment, and caused no pathology at the injection sites.

The aim of this Research Communication was to develop new flow cytometric tools for the fine identification and characterization of milk somatic cells in water buffalo (Bubalus bubalis). Four multicolour panels of antibodies were designed to identify different subsets of live leukocytes and epithelial cells in bulk milk samples. Panel 1, including the CD18/CD172a/CD14/CD16 markers and Live/Dead vitality dye, allowed us to identify total lymphocytes, polymorphonuclear neutrophils (PMN) and monocyte/macrophage subsets. Panel 2 (CD18/CD4/CD8/δ chain/CD335 and Live/Dead dye) allowed us to identify T helper (CD4+), T cytotoxic (CD8+), γδ lymphocytes and NK cells. Panel 3 (CD18/CD79a/CD21 and Live/Dead dye) allowed us to identify total and CD21+ B lymphocytes. Finally, with Panel 4 (CD18/MHC-I/pan Cytokeratin and Live/Dead dye) the epithelial cells were distinguished from leukocytes. In conclusion, we propose a fine characterization of live milk somatic cell (live differential cell count (LDCC)) in buffalo species. In the future the determination of LDCC could used to identify new markers for detecting early inflammatory states of the mammary gland or for monitoring the technological properties of milks of different somatic cell composition.

This study evaluated the effect of sweet whey fermented by Lactobacillus delbrueckii subsp. bulgaricus NCFB 2772, an exopolysaccharide (EPS)-producing lactic acid bacterium (LAB), on the yield and quality of Requeson, a traditional Mexican fresh cheese, further supplemented with the probiotic strain Lactobacillus casei 431. Fermentation of the whey at 42°C for 48 hours resulted in the production of 1.126 ± 0.12 mgeq dextran/mL of EPS. The incorporation of this EPS-rich whey led to a marked (∼230%) increase in cheese yield and significantly enhanced textural properties, including greater hardness, cohesiveness and elasticity. Furthermore, L. casei 431 maintained viability above 107 cfu/g throughout 21 days of refrigerated storage (4°C), with no detectable microbial contamination. Although formulations containing EPS were slightly preferred in sensory evaluations, they exhibited improved stability and retained overall acceptable sensory quality. These results highlight the potential of EPS-producing LAB and probiotics to improve the technological, nutritional and functional attributes of fresh cheese, while contributing to the valorisation of dairy by-products.

This research communication hypothesizes that superstimulation with follicle-stimulating hormone (FSH) would not impact behavior and performance of Holstein cows. The objectives were to investigate the effect of FSH superstimulation on follicular dynamics, animal behaviour, body surface temperature and milk yield of Holstein cows. Cows were blocked by parity and body condition score (BCS), and within a block, they were assigned randomly to receive either normal saline (CON = 7) or 500 IU of FSH (n = 8). The estrous cycle of cows was synchronized by using two injections of prostaglandin F2-alpha (PG), 11 days apart. The dominant follicle was ablated at the time of the second PG injection, and an intra-vaginal controlled internal drug release (CIDR) was inserted (day 0). Two days later, FSH treatment was initiated and continued for 3 days in six equal doses of 83.33 IU after 12 h. Follicles were counted and their sizes were measured from day-0 to day-5. Behavior, including activity and feeding time, was recorded using SmartTag Neck from day 0 to 6. The surface temperatures of the eye, shoulder, flank, and vulva were measured by using infrared thermal imaging every 12 h from day-2 to day-5. Milk was recorded from day-0 to day-6. Mixed effects models were used to analyse the data using SAS statistical software. The number of small and medium follicles did not differ between treatments. However, FSH-treated cows had a more (P = 0.01) large and total follicles compared with CON cows. FSH treatment did not affect activity, feeding time, body surface temperature, or milk yield. In conclusion, FSH superstimulation increased the number of large follicles but did not influence behaviour, body surface temperature, or performance in dairy cows.

This research paper evaluated the effects of different concentrations of pitanga (Eugenia uniflora L.) bagasse (PB) on the chemical, technological, and sensory properties of Petit Suisse. Petit Suisse cheeses were formulated without PB (control) and with the enrichment of 0.5%, 1.0%, and 1.5% PB, respectively. The cheeses were evaluated over a 21-day storage period. PB significantly increased Petit Suisse cheeses’ fibre content and yield. There was also an increase in antioxidant activity. Higher levels of PB resulted in lower hardness and greater syneresis. In rheological measurements, the cheeses showed typical behaviour of a viscoelastic solid; However, the inclusion of PB altered the casein network, making the structure softer than the control. The PB does not affect the viability of Lactococcus and Streptococcus, maintaining the probiotic character of the product. The Petit Suisse with 1% PB resulted in greater sensory acceptance. In conclusion, adding pitanga bagasse to Petit Suisse cheese has been demonstrated to provide attractive, functional characteristics and sensory properties, opening new possibilities for product development in the dairy industry.

This Research Communication presents a cross-sectional study to compare the prevalence of contagious, environmental, opportunistic and other intramammary infection pathogens in mountain dairy herds. In the Italian Apennine mountains, areas where dairy farming is thriving are interspersed among areas where only a few dairy herds remain. The disappearance of some dairy farms relates to the reduction of agricultural and veterinary services in a process that can jeopardize dairy herd profitability. Sixteen herds were screened for intramammary infection (IMI) pathogens. Antimicrobial susceptibility testing of mastitis pathogens was performed, and data on antimicrobial use was collected from the herd treatment registry. The prevalence of contagious IMI pathogens was significantly higher in areas where dairy farming is reducing, whereas environmental and opportunistic infections were more abundant in herds in thriving areas where farmers had a more consistent relationship with veterinarians. Antimicrobial resistance levels were low throughout all areas and did not relate to antimicrobial use, although it was significantly higher in areas where dairy herds were thriving.

This study aimed to develop a predictive model to investigate the effect of temperature, pH, NaCl concentration, initial inoculum concentration and time on enterotoxin A (SEA) production by Staphylococcus aureus. Combinations of three levels of temperature (10, 15 and 25°C ), five levels of pH (5.3, 5.5, 6.0, 6.5 and 6.7), five levels of NaCl (0.8, 1.0, 1.5, 2.0 and 2.2%), three levels of inoculum concentration (0, 3 and 5 log CFU/mL) in brain heart infusion (BHI) broth were studied. Colonies were counted and SEA production was assessed at 24 h intervals for up to 240 h. A probabilistic logistic regression model was used to describe the production of SEA by Staphylococcus aureus. SEA production was influenced by all factors, except NaCl concentration. S. aureus produced SEA in all samples at 25°C, while the temperature of 10°C delayed the growth and SEA production of S. aureus at initial contamination levels of 3 log CFU/mL and 5 log CFU/mL and prevented it at 0 log CFU/mL. The model was statistically and experimentally validated, demonstrating a good fit, with a high percentage agreement, Nagelkerke's R2 and the Hosmer and Lemeshow test for the SEA production model. The experimental validation confirmed the effectiveness of the models for predicting the probability of SEA production by S. aureus in Minas Frescal cheese.

This study investigated yeast diversity and physicochemical changes during the production of Kargı Tulum cheese, a traditional Turkish cheese. Samples were collected at six key stages, from raw milk to the final product, and analysed for parameters including pH, titratable acidity, dry matter, fat, salt, protein, water-soluble protein and maturation index. During ripening, pH values ranged from 3.03 ± 0.017 to 3.70 ± 0.017, while titratable acidity increased from 0.91% ± 0.16% to 3.33% ± 0.17%. Dry matter and fat content increased significantly, reaching 50.41% ± 1.56% and 33.50% ± 5.41%, respectively. Salt content ranged from 0.58% ± 0.08% to 3.03% ± 0.58%, and protein content from 3.70% ± 0.93% to 20.63% ± 1.16%. The maturation index increased from 5.54% ± 0.71% to 16.26% ± 4.35%, indicating ongoing proteolysis. A total of 42 yeast isolates were phenotypically characterized and grouped by sugar fermentation ability, growth and salt tolerance. Internal transcribed spacer region sequencing identified key yeast species, including Kazachstania unispora, Pichia fermentans, Kluyveromyces lactis, Pichia membranifaciens and Geotrichum candidum. These species play significant roles in cheese maturation, contributing to flavour and texture. The results emphasize the importance of indigenous yeast populations in traditional cheese production and offer insights for improving fermentation and ripening processes to enhance cheese quality.

The objectives of the study were to determine somatic cell count (SCC) and evaluate the presence of pathogens (IMI – intramammary infection) in late lactation (LL), followed by the start (colostrum, CL) and approximate peak (established lactation, EL) of the next lactation, as well as to assess the possible transmission of IMI from lactation to lactation. The study was performed on a dairy farm in northern Slovakia. A total of 489 half udder milk samples (242, 80 and 167 in LL, CL and EL, respectively) were collected. Pathogens were identified using MALDI-TOF MS and PathoProof (the latter only in LL). SCC was determined only in LL and EL. Samples were divided according to SCC in four groups from lowest (SCC1 < 500 × 103 cells mL−1) to highest (SCC4 ≥ 2000 × 103 cells mL−1). SCC was higher in LL than in EL. The prevalence of pathogens identified using MALDI-TOF MS was 16.5, 38.8 and 12.6% in LL, CL and EL, respectively. Non-aureus staphylococci and mammaliicocci (NASM) were the most common isolated pathogens in goat milk and colostrum. Staphylococcus (S.) caprae and S. epidermidis species tended to cause persistent IMI in the next lactation. The identification of pathogens using PathoProof was higher than with MALDI-TOF MS. Of all the pathogens (n = 262) identified using PathoProof, the most common were Staphylococcus spp. (86.7%) of which 65.8% exhibited the β-lactamase gene. Additionally, Escherichia coli (4.2%), S. aureus (2.7%), Enterococcus spp. (2.3%), Streptococcus uberis (1.9%), Mycoplasma spp., Protetheca spp. (0.8% each), Arconabacterium pyogenes/Peptoniphilus indolicus and yeast (0.4% each) were also detected using PathoProof. Better identification of pathogen presence in samples with high SCC could contribute to the discussion about SCC as an indicator of subclinical mastitis in goats.

Cheese-whey is a valuable byproduct of the dairy industry, rich in various nutritional components such as minerals, lactose, and proteins. Whey proteins, often used in concentrate form, are widely applied in the food industry due to their diverse chemical, physical, and techno-functional properties. This study aimed to investigate the physicochemical composition and biochemical characteristics of camel and bovine whey after partial demineralization at a laboratory scale. Camel whey exhibited lower pH values compared to bovine whey, while showing comparable levels of total solids, ash, and lactose, but significantly higher protein content. Analysis of both types of whey, before and after dialysis filtration, demonstrated partial demineralization, a significant reduction in lactose content, and a decrease in β-lactoglobulin levels in bovine whey. These findings suggest that demineralized camel and bovine whey hold significant potential for applications in the agricultural and food industries.

This study aimed to identify the best automatic milking system (AMS) parameters and monitoring data for early detection of clinical mastitis in dairy cows and to determine the earliest possible detection within 30 days with the highest predictive accuracy. From August 2021 to February 2022, 55 Holstein cows were monitored for mastitis using physical examination, positive California mastitis test (CMT) and the AMS manufacturer’s software (Delpro®) criteria: milk electrical conductivity ≥ 5.37 mS/cm, milk yield ≤ 80%, somatic cell count (SCC) > 200,000 cells/mL and Mastitis Detection Index (MDi) ≥ 2.0. For every cow suspected of mastitis, two other lactating cows were randomly chosen for evaluation to provide a comparison with healthy herd companions. In total, 129 inspections were evaluated: 39 with clinical mastitis and 90 without. Data on milking, milk composition and production from the AMS, and behavioural data from monitoring collars were summarized for the 30 days leading up to the mastitis diagnosis. Thirty measurement parameters were analysed using generalized linear models. Sensitivity, specificity, accuracy, positive predictive value and negative predictive value were calculated. In the final model, significant parameters included: milk production per day (kg), SCC (cells/mL), average flow mean (kg/min), average conductivity (mS/cm), average flow peak (kg/min), average production per milking (kg), milking duration (s), rumination (min/day), panting (min/day) and feeding activity (min/day). From −30 to −10 days, accuracy, sensitivity and specificity varied without a defined pattern. However, from day −9, there was stabilization of the evaluated parameters. Results showed an average accuracy of 79.2%, a sensitivity of 82.5%, a specificity of 78.7%, a positive predictive value of 41.5% and a negative predictive value of 92.2% in predicting mastitis occurrence. In conclusion, using AMS parameters and behavioural data from monitoring collars, it was possible to predict clinical mastitis in dairy cows in an AMS with a 9-day advance notice.