In der vorliegenden Arbeit wurde die Bedeutung des Einflusses von Laktationsnummer, Laktationsstadium, Körperkondition und Absetzverfahren auf Merkmale der Milchleistung und Eutergesundheit einer Herde Ostfriesischer Milchschafe untersucht. Die Tiere wurden nach dem Ablammen zufällig einer Früh- bzw. Spätabsetzergruppe zugeordnet (Absetzen der Lämmer 3 bzw. 42 Tage post partum). An insgesamt 40 Terminen wurden über eine vollständige Laktation Vorgemelks-, Hälftenanfangsgemelks- und Einzeltiergemelksproben gewonnen. Das Spektrum der untersuchten Merkmale umfasste die Milchmenge, die Milchinhaltsstoffe (Fett, Eiweiß, Laktose), die klinische Untersuchung des Euters, die somatische Zellzahl, die elektrische Leitfähigkeit und bakteriologische Untersuchungen. Zusätzlich wurden das Körpergewicht und die Körperkondition der Mutterschafe erfasst. Das durchschnittliche Leistungsniveau der Herde betrug 301±101,3 kg Milch, bei mittleren Milchfett-, Milcheiweiß-, und Laktosegehalten von 5,00, 5,14 bzw. 5,00 % (150-Tageleistung). Die Laktationsnummer hatte keinen signifikanten Effekt auf die Milchleistung. Schafe der Spätabsetzer-Gruppe produzierten im Anschluss an die Säugephase eine um 300 g signifikant höhere Testtagsmilchmenge als Tiere der Frühabsetzer-Gruppe. Das Absetzverfahren hatte keinen nachweisbaren Effekt auf die Milchinhaltsstoffe. Die Herde wies eine gute Eutergesundheit auf. Der Anteil bakteriologisch positiver Schafmilchproben betrug 28,5 %. Als dominante Erreger wurden in 96,6 % der bakteriologisch positiven Schafmilchproben Koagulasenegative Staphylokokken nachgewiesen. Der Anteil bakteriologisch positiver Euterhälftenbefunde stieg signifikant mit dem Fortschreiten der Laktation und zunehmender Laktationsnummer. Eine Euterinfektion mit Koagulasenegativen Staphylokokken sowie das Verfahren des Spätabsetzens beeinflussten alle drei untersuchten Eutergesundheitsparameter (logarithmierte Zellzahl, elektrische Leitfähigkeit und Laktosegehalt) negativ (p < 0,001).

La mammite subclinique est un problème de santé fréquent et coûteux. Les infections intra-mammaires (IIM) sont souvent détectées à l’aide de mesures du comptage des cellules somatiques (CCS). La culture bactériologique du lait est cependant requise afin d’identifier le pathogène en cause. À cause de cette difficulté, pratiquement toutes les recherches sur la mammite subclinique ont été centrées sur la prévalence d’IIM et les facteurs de risque pour l’incidence ou l’élimination des IIM sont peu connus. L’objectif principal de cette thèse était d’identifier les facteurs de risque modifiables associés à l’incidence, l’élimination et la prévalence d’IIM d’importance dans les troupeaux laitiers Canadiens. En premier lieu, une revue systématique de la littérature sur les associations entre pratiques utilisées à la ferme et CCS a été réalisée. Les pratiques de gestion constamment associées au CCS ont été identifiées et différentiées de celles faisant l’objet de rapports anecdotiques. Par la suite, un questionnaire bilingue a été développé, validé, et utilisé afin de mesurer les pratiques de gestion d’un échantillon de 90 troupeaux laitiers canadiens. Afin de valider l’outil, des mesures de répétabilité et de validité des items composant le questionnaire ont été analysées et une évaluation de l’équivalence des versions anglaise et française a été réalisée. Ces analyses ont permis d’identifier des items problématiques qui ont du être recatégorisés, lorsque possible, ou exclus des analyses subséquentes pour assurer une certaine qualité des données. La plupart des troupeaux étudiés utilisaient déjà la désinfection post-traite des trayons et le traitement universel des vaches au tarissement, mais beaucoup des pratiques recommandées n’étaient que peu utilisées. Ensuite, les facteurs de risque modifiables associés à l’incidence, à l’élimination et à la prévalence d’IIM à Staphylococcus aureus ont été investigués de manière longitudinale sur les 90 troupeaux sélectionnés. L’incidence d’IIM semblait être un déterminant plus important de la prévalence d’IIM du troupeau comparativement à l’élimination des IIM. Le port de gants durant la traite, la désinfection pré-traite des trayons, de même qu’une condition adéquate des bouts de trayons démontraient des associations désirables avec les différentes mesures d’IIM. Ces résultats viennent souligner l’importance des procédures de traite pour l’obtention d’une réduction à long-terme de la prévalence d’IIM. Finalement, les facteurs de risque modifiables associés à l’incidence, à l’élimination et à la prévalence d’IIM à staphylocoques coagulase-négatif (SCN) ont été étudiés de manière similaire. Cependant, afin de prendre en considération les limitations de la culture bactériologique du lait pour l’identification des IIM causées par ce groupe de pathogènes, une approche semi-Bayesienne à l’aide de modèles de variable à classe latente a été utilisée. Les estimés non-ajusté de l’incidence, de l’élimination, de la prévalence et des associations avec les expositions apparaissaient tous considérablement biaisés par les imperfections de la procédure diagnostique. Ce biais était en général vers la valeur nulle. Encore une fois, l’incidence d’IIM était le principal déterminant de la prévalence d’IIM des troupeaux. Les litières de sable et de produits du bois, de même que l’accès au pâturage étaient associés à une incidence et une prévalence plus basse de SCN. / Subclinical mastitis is a very frequent and costly health issue that can be detected using somatic cell count (SCC) measurements, but requires the use of milk bacteriological culture for identification of the causal pathogen. Because of this latter difficulty, nearly all subclinical mastitis research has focused on prevalent intramammary infections (IMI) and less is known on risk factors for IMI incidence or elimination. The main objective of this thesis was to identify the manageable risk factors associated with the incidence, elimination, and prevalence of IMI of importance in Canadian dairy herds. First, a systematic review of the literature on the associations between management practices used on farms and SCC was carried out. Management practices consistently associated with SCC were identified and differentiated from other practices for which anecdotic reports were available. Then, a bilingual questionnaire was developed, validated, and employed to measure the practices used on a sample of 90 Canadian dairy herds. To validate this tool, measures of repeatability and of validity of the questionnaire’s items were analyzed and an evaluation of the equivalence of the English and French versions was conducted. These analyses indicated that the questionnaire was, in general, acceptable, but also pinpointed some problematic items. These items were recategorized when possible or otherwise excluded from subsequent analyses to ensure good data quality. Most of the herds studied were already using post-milking teat disinfection and blanket dry cow therapy. Many other frequently recommended practices were not widely adopted. Next, manageable risk factors associated with Staphylococcus aureus IMI incidence, elimination, and prevalence were investigated on the 90 selected herds in a longitudinal fashion. The S. aureus IMI incidence appeared to be a stronger determinant of the herd prevalence than the elimination rate. Among other practices, wearing gloves during milking, using pre-milking teat disinfection, and having an adequate teat end condition showed desirable associations with the outcomes. These results highlight the importance of good milking practices to achieve a long-term reduction of S. aureus IMI prevalence. Finally, manageable risk factors associated with coagulase-negative staphylococci (CNS) IMI incidence, elimination, and prevalence were investigated in a similar manner. Because of the greater limitations of milk bacteriological culture to identify IMI caused by this group of pathogens, a semi-Bayesian latent class model approach was used. Sensitivity and specificity estimates from an internal validation study were used to link the observed IMI milk culture result to the latent true quarter IMI status. Non-adjusted estimates of IMI incidence, elimination, prevalence, and of associations with expositions appeared to be considerably biased by the diagnostic procedure’s imperfections. Most often, estimates were biased toward the null value. Again, IMI incidence was the main determinant of the herd IMI prevalence. Sand and wood products bedding were associated with lower CNS incidence and prevalence. Sending cows to pasture was also associated with lower CNS IMI incidence and prevalence.

Animal traits differ not only in mean, but also in variation around the mean. For instance, one sire’s daughter group may be very homogeneous, while another sire’s daughters are much more heterogeneous in performance. The difference in residual variance can partially be explained by genetic differences. Models for such genetic heterogeneity of environmental variance include genetic effects for the mean and residual variance, and a correlation between the genetic effects for the mean and residual variance to measure how the residual variance might vary with the mean. The aim of this thesis was to develop a method based on double hierarchical generalized linear models for estimating genetic heteroscedasticity, and to apply it on four traits in two domestic animal species; teat count and litter size in pigs, and milk production and somatic cell count in dairy cows. The method developed is fast and has been implemented in software that is widely used in animal breeding, which makes it convenient to use. It is based on an approximation of double hierarchical generalized linear models by normal distributions. When having repeated observations on individuals or genetic groups, the estimates were found to be unbiased. For the traits studied, the estimated heritability values for the mean and the residual variance, and the genetic coefficients of variation, were found in the usual ranges reported. The genetic correlation between mean and residual variance was estimated for the pig traits only, and was found to be favorable for litter size, but unfavorable for teat count.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The study aims to evaluate the effect of somatic cell count (SCC) and total bacterial count (TBC) on the constituents of milk, checking the influence that the climatic variables have on these hygienic-sanitary indicators. Data were obtained from 1,541 dairy farms located in 15 municipalities in the dairy region of Vale do Taquari, Rio Grande do Sul. It was tabulated the data from somatic cell count (SCC), total bacterial count (TBC) and milk composition, from June 2008 to December 2011, totaling 44,089 samples. The ambient temperature showed positive and significant correlation to score of somatic cell, while rainfall and relative humidity air showed no correlation. Moreover, the climatic variables have no significant correlation to the TBC. The fat, protein, minerals and total solids were directly correlated with the SCC and TBC, while solids-not-fat and lactose showed an opposite behavior. The principal component analysis (PCA) followed by agglomerative hierarchical clustering method, have showed that milk from SCC 401,000 to 750,000 cell mL-1 and TBC from 100,000 to 1,000,000 CFU mL-1, have the same quality, not justifying stratification in these intervals. In addition, it was observed that only limits regulatory standards are not sufficient to improve the quality of milk, suggesting parameters for payment system based on the bonus and penalty relative to the TBC and SCC milk. / Este estudo teve por objetivos avaliar o efeito da contagem de células somáticas (CCS) e da contagem bacteriana total (CBT) sobre os constituintes do leite, verificando a influência que as variáveis climáticas exercem sobre estes indicadores higiênico-sanitários. Os dados utilizados foram de 1.541 unidades produtoras de leite referentes a 15 municípios da bacia leiteira do Vale do Taquari, Rio Grande do Sul. Foram tabulados os dados de contagem de células somáticas (CCS), contagem bacteriana total (CBT) e composição centesimal do leite, referentes ao período de junho de 2008 a dezembro de 2011, totalizando 44.089 amostras analisadas. A temperatura ambiente apresentou correlação positiva e significativa com o escore de células somáticas, enquanto a precipitação pluviométrica e umidade relativa do ar ausência de correlação. Por outro lado, as variáveis climáticas não apresentaram correlação significativa com a CBT. Os teores de gordura, proteína, minerais e sólidos totais foram diretamente correlacionados com a CCS e a CBT, enquanto que os sólidos não gordurosos e a lactose apresentaram comportamento inverso. A análise de componentes principais (ACP) seguida pelo método hierárquico aglomerativo de agrupamento, permitiu constatar que leites com CCS maiores que 400.000 até 750.000 céls mL-1 e CBT superiores a 100.000 até 1.000.000 UFC mL-1, apresentam a mesma qualidade, não justificando a estratificação de intervalos nesta amplitude de variação. Paralelamente, observou-se que somente limites de normas regulatórias não são suficientes para melhoraria da qualidade do leite, sugerindo-se, parâmetros para sistemas de pagamento baseado na bonificação e penalização em relação à CCS e CBT do leite.

Submitted by Cláudia Bueno (claudiamoura18@gmail.com) on 2016-01-08T15:32:37Z No. of bitstreams: 2 Dissertação - Thamara Venâncio de Almeida - 2015.pdf: 1390852 bytes, checksum: 5f0b1c9700873dd65b4e3d743efec82f (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2016-01-11T06:39:56Z (GMT) No. of bitstreams: 2 Dissertação - Thamara Venâncio de Almeida - 2015.pdf: 1390852 bytes, checksum: 5f0b1c9700873dd65b4e3d743efec82f (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2016-01-11T06:39:56Z (GMT). No. of bitstreams: 2 Dissertação - Thamara Venâncio de Almeida - 2015.pdf: 1390852 bytes, checksum: 5f0b1c9700873dd65b4e3d743efec82f (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2015-08-21 / For assessing compliance with the quality requirements established by the Normative Instruction No. 62, a milk sample each property or community tank should be analyzed on a monthly basis by one of the laboratories accredited by the Brazilian Laboratories Network for Milk Quality Control . There are still doubts about what would be the maximum temperature and the time of milk sample storage without jeopardizing the results of the electronic analysis. The objective of this study was to evaluate the effects of temperature and storage time of cooled samples of raw milk on the results of milk quality electronic analysis. Refrigerated raw milk samples were collected from expansion tanks for individual use of farms located in the middle region of the state of Goiás, and stored at four different temperatures (3 °C, 11 °C, 17 °C and 25 °C) for 16 days. Total bacterial count (TBC), somatic cell count (SCC) and chemical composition were performed daily. The results were submitted to analysis of variance in split plot design in randomized blocks, and the means were compared by Duncan test. We concluded that, when azidiol is added as preservative to samples for TBC, they can be analyzed up to 16 days after collection when stored at temperatures of 3 °C and 11 °C, and up to 10 days when stored at 17 °C. Moreover, when bronopol is added as preservative, samples for SCC and chemical composition analysis can be stored for up to 16 days after collection when stored at temperatures of 3 °C and 11 °C, and for up to seven days when stored at 17 °C. / Para a avaliação do cumprimento dos requisitos de qualidade estabelecidos pela Instrução Normativa nº 62, uma amostra de leite de cada propriedade ou tanque comunitário deve ser analisada mensalmente por um dos laboratórios credenciados pela Rede Brasileira de Laboratórios de Controle da Qualidade do Leite. Ainda existem dúvidas sobre quais seriam a temperatura e o tempo máximos de armazenamento das amostras de leite sem que houvesse comprometimento dos resultados das análises eletrônicas. Objetivou-se com este trabalho avaliar os efeitos da temperatura e do tempo de armazenamento de amostras de leite cru refrigerado nos resultados das análises eletrônicas da qualidade do leite. Amostras de leite cru refrigerado foram coletadas de tanques de expansão de uso individual de fazendas localizadas na mesorregião Centro Goiano no Estado de Goiás e armazenadas em quatro temperaturas diferentes (3°C, 11°C, 17°C e 25°C) durante 16 dias. Foram realizadas diariamente análises de contagem bacteriana total (CBT), contagem de células somáticas (CCS) e composição química. Os resultados obtidos foram submetidos à análise de variância em esquema de parcelas subdivididas no tempo em delineamento de blocos ao acaso e as médias comparadas pelo teste de Duncan. Concluiu-se que as amostras destinadas à análise de CBT, adicionadas do conservante azidiol, podem ser analisadas por até 16 dias após a coleta quando armazenadas em temperaturas de 3°C e 11°C e por até 10 dias quando armazenadas a 17°C. E as amostras destinadas às análises de CCS e composição química, adicionadas do conservante bronopol, podem ser analisadas por até 16 dias após a coleta quando armazenadas em temperaturas de 3°C e 11°C e por até sete dias quando armazenadas a 17°C.

The concern about food safety, animal welfare, and environmental impacts caused by intensive production systems has led to an increasing interest in foods derived from a sustainable production system. In this context, organic milk production in Brazil has grown considerably in recent years, but little is known about the quality of the food produced in the country. In order to evaluate the quality of organic milk, the hygienic quality, chemical composition, and safety concerning pesticide residues was assessed. Five brands of certified organic pasteurized milk and five brands of conventional pasteurized milk commercialized in three Brazilian states were collected between May 2011 and March 2012, every two months, totaling 29 samples of organic milk (OM) and 27 conventional milk (CM). The collection of raw milk samples was made every two months, from July 2011 to May 2012 in 20 organic farms and 20 conventional farms in western Santa Catarina, totaling 120 OM and 115CM samples. In pasteurized milk, physicochemical analyzes were performed and the fatty acid profile was determined. In raw milk, chemical composition analyzes, somatic cell count (SCC), and total bacterial count (TBC) were carried out. Both types of milk were subjected to the determination of pesticides and veterinary drugs. When the results of raw milk production in the two systems were compared, no difference for the mean value of fat, protein and total solids, or SCC was found. Organic milk showed better quality in relation to microbial contamination and higher content of lactose. For pasteurized milk with the exception of non fat solids, all average values of physical and chemical parameters of milk met the limits of the Brazilian legislation. The fatty acid profile was affected by geographical origin, and to a lesser extent, the effect of season, probably due to differences on the feeding regime. The milk obtained in Rio Grande do Sul and São Paulo were characterized by having higher values of health-promoting fatty acids, including conjugated linoleic acid (CLA). However, the milk collected in the Federal District was characterized by higher concentrations of saturated fatty acids (SFA). The results for raw and pasteurized milk indicate that the organic management alone had no effect on milk quality regarding chemical composition, which are determined by several factors that are independent of the production system. No veterinary drugs was found in any of the samples analyzed. However, for contamination with pesticides no difference was observed when considering the number of contaminated samples in either production system. Results indicate the need to monitor milk for the presence of pesticides in order to improve food safety, especially for organic milk, which should be free of such contaminants. / A preocupação com segurança alimentar, bem-estar animal e impactos no meio ambiente causados pelos sistemas intensivos de produção, tem levado a um maior interesse por alimentos que sejam oriundos de um sistema de produção sustentável. Neste contexto, a produção de leite orgânico no Brasil cresceu consideravelmente nos últimos anos, porém pouco se sabe sobre a qualidade deste alimento produzido no país. Para avaliar a qualidade do leite orgânico foram realizadas análises que contemplaram aspectos da qualidade higiênica, composição química e a segurança quanto à presença de resíduos químicos. Cinco marcas de leite orgânico pasteurizado certificado e cinco de leite pasteurizado convencional comercializados em três Estados brasileiros foram coletadas entre maio de 2011 e março de 2012, com intervalo de dois meses, totalizando 29 amostras de leite orgânico (LO) e 27 de leite convencional (LC). As coletas de leite cru foram feitas de julho de 2011 a maio de 2012, também bimestrais, em 20 propriedades orgânicas e 20 propriedades convencionais na região Oeste de Santa Catarina, totalizando 120 amostras de LO e 115 de LC. No leite pasteurizado foram realizadas análises físico-químicas e o perfil de ácidos graxos. No leite cru foram realizadas as análises de composição química, contagem de células somáticas (CCS) e contagem bacteriana total (CBT). Ambos os tipos de leite foram submetidos à determinação de agrotóxicos e medicamentos veterinários. Comparando-se os resultados do leite cru, entre os dois sistemas de produção, não foi encontrado diferença para o valor médio de gordura, proteína e sólidos totais, assim como para CCS. O leite orgânico apresentou melhor qualidade em relação a contaminação microbiana e teor superior de lactose. Para o leite pasteurizado com exceção do extrato seco desengordurado (ESD), todos os valores médios dos parâmetros físicos e químicos atenderam aos limites da legislação brasileira. O perfil de ácidos graxos sofreu efeito da origem geográfica, e em menor escala o efeito da sazonalidade, provavelmente por diferenças no regime alimentar. Os leites obtidos no Rio Grande do Sul e em São Paulo se caracterizaram por possuir valores superiores de ácidos graxos promotores da saúde, entre eles o ácido linoléico conjugado (CLA), enquanto os leites coletados no Distrito Federal se caracterizaram por concentrações superiores de ácidos graxos saturados. Os resultados para o leite cru e pasteurizado indicam que o manejo orgânico por si só, não influenciou na qualidade do leite quanto aos aspectos de composição química, que são determinados por vários fatores que independem do sistema de produção. Não foram encontrados medicamentos veterinários em nenhuma das amostras analisadas. Quanto à contaminação por agrotóxicos não houve diferença quando considerado o número de amostras contaminadas em ambos os sistemas de produção. Os resultados indicam a necessidade de monitorar o leite quanto a presença de agrotóxicos a fim de melhorar a segurança alimentar, principalmente para leite orgânico, que deve estar livre desse tipo de contaminantes.

In der vorliegenden Arbeit wurde die Bedeutung des Einflusses von Laktationsnummer, Laktationsstadium, Körperkondition und Absetzverfahren auf Merkmale der Milchleistung und Eutergesundheit einer Herde Ostfriesischer Milchschafe untersucht. Die Tiere wurden nach dem Ablammen zufällig einer Früh- bzw. Spätabsetzergruppe zugeordnet (Absetzen der Lämmer 3 bzw. 42 Tage post partum). An insgesamt 40 Terminen wurden über eine vollständige Laktation Vorgemelks-, Hälftenanfangsgemelks- und Einzeltiergemelksproben gewonnen. Das Spektrum der untersuchten Merkmale umfasste die Milchmenge, die Milchinhaltsstoffe (Fett, Eiweiß, Laktose), die klinische Untersuchung des Euters, die somatische Zellzahl, die elektrische Leitfähigkeit und bakteriologische Untersuchungen. Zusätzlich wurden das Körpergewicht und die Körperkondition der Mutterschafe erfasst. Das durchschnittliche Leistungsniveau der Herde betrug 301±101,3 kg Milch, bei mittleren Milchfett-, Milcheiweiß-, und Laktosegehalten von 5,00, 5,14 bzw. 5,00 % (150-Tageleistung). Die Laktationsnummer hatte keinen signifikanten Effekt auf die Milchleistung. Schafe der Spätabsetzer-Gruppe produzierten im Anschluss an die Säugephase eine um 300 g signifikant höhere Testtagsmilchmenge als Tiere der Frühabsetzer-Gruppe. Das Absetzverfahren hatte keinen nachweisbaren Effekt auf die Milchinhaltsstoffe. Die Herde wies eine gute Eutergesundheit auf. Der Anteil bakteriologisch positiver Schafmilchproben betrug 28,5 %. Als dominante Erreger wurden in 96,6 % der bakteriologisch positiven Schafmilchproben Koagulasenegative Staphylokokken nachgewiesen. Der Anteil bakteriologisch positiver Euterhälftenbefunde stieg signifikant mit dem Fortschreiten der Laktation und zunehmender Laktationsnummer. Eine Euterinfektion mit Koagulasenegativen Staphylokokken sowie das Verfahren des Spätabsetzens beeinflussten alle drei untersuchten Eutergesundheitsparameter (logarithmierte Zellzahl, elektrische Leitfähigkeit und Laktosegehalt) negativ (p < 0,001).

Bovine mastitis adversely affects the dairy industry around the world. This disease is caused by a diverse range of bacteria, broadly categorised as minor and major pathogens. In-line tools that help identify these bacterial groupings in the early stages of the disease are advantageous as timely decisions could be made before the cow develops any clinical symptoms. The first objective of this research was to identify the most informative milk parameters for the detection of minor and major bacterial pathogens. The second objective of this research was to evaluate the potential of supervised and unsupervised neural network learning paradigms for the detection of minor infected and major infected quarters in the early stages of the disease. The third objective was to evaluate the effects of different proportions of infected to non-infected cases in the training data set on the correct classification rate of the supervised neural network models as there are proportionately more non-infected cases in a herd than infected cases. A database developed at Lincoln University was used to achieve the research objectives. Starting at calving, quarter milk samples were collected weekly from 112 cows for a period of fourteen weeks, resulting in 4852 samples with complete records for somatic cell count (SCC), electrical resistance, protein percentage, fat percentage, and bacteriological status. To account for the effects of the stage of lactation on milk parameters with respect to days in milking, data was divided into three days in milk ranges. In addition, cow variation was accounted for by the sire family from which the cow originated and the lactation number of each cow. Data was pre-processed before the application of advanced analytical techniques. Somatic cell score (SCS) and electrical resistance index were derived from somatic cell count and electrical resistance, respectively. After pre-processing, the data was divided into training and validation sets for the unsupervised neural network modelling experiment and, for the supervised neural network modelling experiments, the data was divided into training, calibration and validation sets. Prior to any modelling experiments, the data was analysed using statistical and multivariate visualisation techniques. Correlations (p<0.05) were found between the infection status of a quarter and its somatic cell score (SCS, 0.86), electrical resistance index (ERI, -0.59) and protein percentage (PP, 0.33). The multivariate parallel visualisation analysis validated the correlation analysis. Due to significant multicolinearity [Correlations: SCS and ERI (-0.65: p<0.05); SCS and PP (0.32: p<0.05); ERI and PP (-0.35: p<0.05)], the original variables were decorrelated using principle component analysis. SCS and ERI were found to be the most informative variables for discriminating between non-infected, minor infected and major infected cases. Unsupervised neural network (USNN) model was trained using the training data set which was extracted from the database, containing approximately equal number of randomly selected records for each bacteriological status [not infected (NI), infected with a major pathogen (MJI) and infected with a minor pathogen (MNI)]. The USNN model was validated with the remaining data using the four principle components, days in milk (DIM), lactation number (LN), sire number, and bacteriological status (BS). The specificity of the USNN model in correctly identifying non infected cases was 97%. Sensitivities for correctly detecting minor and major infections were 89% and 80%, respectively. The supervised neural network (SNN) models were trained, calibrated and validated with several sets of training, calibration and validation data, which were randomly extracted from the database in such a way that each set has a different proportion of infected to non-infected cases ranging from 1:1 to 1:10. The overall accuracy of these models based on validation data sets gradually increased with increase in the number of non-infected cases in the data sets (80% for the 1:1, 84% for 1:2, 86% for 1:4 and 93% for 1:10). Specificities of the best models for correctly recognising non-infected cases for the four data sets were 82% for 1:1, 91% for 1:2, 94% for 1:4 and 98% for 1:10. Sensitivities for correctly recognising minor infected cases for the four data sets were 86% for 1:1, 76% for 1:2, 71% for 1:4 and 44% for 1:10. Sensitivities for correctly recognising major infected cases for the four data sets were 20% for 1:1, 20% for 1:2, 30% for 1:4 and 40% for 1:10. Overall, sensitivity for the minor infected cases decreased while that of major infected cases increased with increase in the number non-infected cases in the training data set. Due to the very low prevalence of MJI category in this particular herd, results for this category may be inconclusive. This research suggests that somatic cell score and electrical resistance index of milk were the most effective variables for detecting the infection status of a quarter followed by milk protein and fat percentage. The neural network models were able to differentiate milk containing minor and major bacterial pathogens based on milk parameters associated with mastitis. It is concluded that the neural network models can be developed and incorporated into milking machines to provide an efficient and effective method for the diagnosis of mastitis.

La litière de fumier recyclé (LFR) est utilisée dans les fermes laitières québécoises depuis quelques années, et ce, malgré le manque de connaissances scientifiques quant aux risques reliés à l’utilisation de ce produit pour la santé des animaux. Le premier objectif de cette thèse était de décrire les méthodes de production de LFR dans un contexte québécois et les pratiques de régie associées. Le deuxième objectif du projet consistait à déterminer le potentiel de recroissance de certaines espèces bactériennes dans la LFR lorsqu’elle est contaminée. Par la suite, le niveau de propreté et la prévalence de lésions aux jarrets chez les vaches laitières exposées ou non à ce produit ont été estimés. Finalement, l’association entre l’utilisation de LFR et l’incidence de mammite sous-clinique et clinique a été évaluée. Des études observationnelles cohorte ou transversale (selon l’objectif de recherche) ont été réalisées en 2018-2019 sur 27 fermes utilisant de la LFR et 61 fermes utilisant de la litière de paille, à titre comparatif. Les visites de fermes ont permis de constater que les méthodes utilisées pour produire de la LFR n’étaient pas standardisées et qu’elles ne permettaient pas un réel compostage de la fraction solide du fumier. Les essais de recroissance ont permis de démontrer que les différentes LFR (traitée en tas, contenant fermé ou cuve rotative) ne réagissaient pas de la même façon à une inoculation par des coliformes. La LFR de cuve rotative contenait une plus faible concentration initiale de Klebsiella spp. et a montré une croissance bactérienne significative dans les premières 24 h post-inoculation. Les LFR conditionnées dans des tas ou contenants avaient quant à elles une concentration initiale importante de Klebsiella spp. et n’ont pas démontré de croissance bactérienne significative suivant leur inoculation, ce qui suggère un effet de saturation de croissance bactérienne. Au cours de l’étude transversale, 30 vaches par troupeau, en moyenne, ont été notées à l’aide d’un score de propreté sur trois zones du corps ainsi que d’un score sur l’état des jarrets, afin d’estimer leur propreté et la présence de lésions aux jarrets. Le score de propreté attribué à chaque zone du corps des vaches allait de 1 à 4, 1 étant très propre et 4 très sale. Nous avons observé que les vaches logées sur LFR avaient, en général, une meilleure propreté du pis et du bas des pattes que celles logées sur paille. La LFR avait un effet protecteur pour le risque d’avoir un score de propreté du pis ≥ 3 (rapport de cotes (RC) : 0,43) ou d’avoir un score de 4 (RC : 0,29). Les vaches logées sur LFR avaient aussi le bas des membres plus propres que celles logées sur paille avec de plus faibles cotes d’avoir un score ≥ 2 (RC : 0,45), un score ≥ 3 (RC : 0,16) ou un score de 4 (RC : 0,07). Cependant, nous n’avons pas trouvé de différence entre les deux groupes d’animaux au niveau de la propreté du flanc et du haut des pattes. Lors de leur évaluation pour la propreté, les vaches ont aussi été notées à l’aide d’un score de lésions aux jarrets allant de 0 à 3. Un score de 0 représentait un jarret parfaitement sain alors qu’un score de 3 était attribué à un jarret présentant une enflure de plus de 2,5 cm. Les deux jarrets étaient évalués et le jarret ayant reçu le score le plus élevé était inclus dans les analyses. Nous n’avons pas identifié de différence quant aux scores de jarret entre les deux groupes d’animaux. Afin d’évaluer la santé de la glande mammaire, une étude cohorte d’une durée d’un an a été mise en place à partir de la visite de la ferme. Les dynamiques de comptages de cellules somatiques ont été suivies sur 11 031 vaches durant cette période afin d’analyser l’incidence de mammite sous-clinique. Nous n’avons pas été en mesure de détecter une différence d’incidence entre les deux groupes d’animaux. Au cours de l’étude cohorte, les producteurs laitiers ont identifié et fait parvenir au laboratoire 1 144 échantillons de lait provenant de vaches atteintes de mammites cliniques. L’incidence totale de mammite clinique n’était pas plus élevée dans les fermes LFR qu’au sein des fermes paille. Cependant, lorsque nous avons analysé l’incidence de mammite clinique par agent pathogène spécifique, nous avons pu constater que les vaches logées sur LFR étaient 7,0 fois plus à risque d’expérimenter une mammite clinique causée par Klebsiella pneumoniae que celles du groupe comparatif. / Recycled manure solids (RMS) bedding has been used on Quebec dairy farms for a number of years, despite the lack of scientific knowledge about the health risks associated with the use of this product for animals. The first objective of this thesis was to describe the RMS production methods in a Quebec context and the associated management practices. The second objective of the project was to determine the potential for regrowth of certain bacterial species in the RMS when it is contaminated. Subsequently, the level of cleanliness and prevalence of hock lesions in dairy cows exposed or not to this product were estimated. Finally, the association between RMS use and the incidence of subclinical and clinical mastitis was evaluated. Observational cohort or cross-sectional studies (depending on the research objective) were conducted in 2018-2019 on 27 farms using RMS and 61 farms using straw bedding for comparison. The farm visits highlighted that the methods used to produce RMS were not standardized and did not allow for true composting of the manure solid fraction. The regrowth trials showed that different RMS (treated in heap, closed container or rotating drum) did not react in the same way to coliform inoculation. The rotating drum RMS contained a lower initial concentration of Klebsiella spp. and experienced significant bacterial growth in the first 24 hours post-inoculation. The heap or closed container RMS had a high initial concentration of Klebsiella spp. and did not show significant bacterial growth following inoculation, suggesting a saturation effect on bacterial growth. In the cross-sectional study, an average of 30 cows were measured using a cleanliness score on three body areas as well as a hock lesion score to estimate their cleanliness and the presence of hock lesions. The cleanliness score assigned to each area of the cows' body ranged from 1 to 4, with 1 being very clean and 4 being very dirty. We found that cows housed on RMS generally had better udder and lower leg cleanliness than those housed on straw. Recycled manure solids bedding had a protective effect for the risk of having an udder cleanliness score ≥ 3 (odds ratio (OR): 0,43) or having a score of 4 (OR: 0,29). Cows housed on RMS also had cleaner lower legs than those housed on straw with lower odds of having a score ≥ 2 (OR: 0,45), a score ≥ 3 (OR: 0,16) or a score of 4 (OR: 0,07). However, we found no difference between the two groups of animals in flank and upper leg cleanliness. When they were evaluated for cleanliness, cows were also measured using a hock lesion score ranging from 0 to 3. A score of 0 represented a perfectly healthy hock, while a score of 3 was assigned to a hock with swelling greater than 2.5 cm. Both hocks were scored and the hock with the higher score was included in the analyses. We did not identify any difference regarding hock lesions between the two groups of animals. To assess mammary gland health, a one-year cohort study was set up from the farm visit. Somatic cell count dynamics were followed on 11 031 cows during this period to analyze the incidence of subclinical mastitis. We were not able to detect a difference in incidence between the two groups of animals. During the cohort study, dairy farmers identified and sent to the laboratory 1 144 milk samples from cows with clinical mastitis. The total incidence of clinical mastitis was not higher on RMS farms than on straw farms. However, when we analyzed the incidence of clinical mastitis by specific pathogen, we found that cows housed on RMS were 7.0 times more likely to experience clinical mastitis caused by Klebsiella pneumoniae than those in the comparison group.