RESUMO
This study aims to use advanced machine learning techniques supported by Principal Component Analysis (PCA) to estimate body weight (BW) in buffalos raised in southeastern Mexico and compare their performance. The first stage of the current study consists of body measurements and the process of determining the most informative variables using PCA, a dimension reduction method. This process reduces the data size by eliminating the complex structure of the model and provides a faster and more effective learning process. As a second stage, two separate prediction models were developed with Gradient Boosting and Random Forest algorithms, using the principal components obtained from the data set reduced by PCA. The performances of both models were compared using R2, RMSE and MAE metrics, and showed that the Gradient Boosting model achieved a better prediction performance with a higher R2 value and lower error rates than the Random Forest model. In conclusion, PCA-supported modeling applications can provide more reliable results, and the Gradient Boosting algorithm is superior to Random Forest in this context. The current study demonstrates the potential use of machine learning approaches in estimating body weight in water buffalos, and will support sustainable animal husbandry by contributing to decision making processes in the field of animal science.
RESUMO
This study aimed to predict Blackbelly sheep carcass tissue composition using ultrasound measurements and machine learning models. The models evaluated were decision trees, random forests, support vector machines, and multi-layer perceptrons and were used to predict the total carcass bone (TCB), total carcass fat (TCF), and total carcass muscle (TCM). The best model for predicting the three parameters, TCB, TCF, and TCM was random forests, with mean squared error (MSE) of 0.31, 0.33, and 0.53; mean absolute error (MAE) of 0.26, 0.29, and 0.53; and the coefficient of determination (R2) of 0.67, 0.69, and 0.76, respectively. The results showed that machine learning methods from in vivo ultrasound measurements can be used as determinants of carcass tissue composition, resulting in reliable results.
Assuntos
Aprendizado de Máquina , Músculos , Animais , Ovinos , Ultrassonografia/veterinária , Redes Neurais de Computação , Algoritmo Florestas AleatóriasRESUMO
This study was carried out to evaluate the residual feed intake (RFI), volatile fatty acid (VFA) production and enteric methane (CH4) from growing Pelibuey sheep. In this case, 12 non-castrated Pelibuey with an initial average live weight (LW) of 21.17 ± 3.87 kg and an age of 3 months, were housed in individual pens and fed a basal diet with 16% of crude protein and 11 MJ ME for 45 days. Dry matter intake (DMI) was measured and the daily weight gain (DWG) was calculated using a linear regression between the LW and experimental period. Mean metabolic live weight (LW0.75) was calculated. RFI was determined by linear regression with DWG and LW0.75 as independent variables. Lambs were classified as low, medium, and high RFI. Feed efficiency was determined as DWG/DMI. For determining rumen pH, ammonia nitrogen concentration NH3-N), and VFA, ruminal fluid was obtained using an esophageal probe on day 40. Feed intake of low RFI lambs was approximately 16% lower (p < 0.05) while growth rate was not significantly different. Their average energy loss, expressed as CH4 production per kilogram of metabolic weight, was 17% lower (p < 0.05).
RESUMO
The objective of the present study was to determine the needs of milk intake (MI) and their components (i.e., lactose, fat, protein, and energy) to increase the average daily gain (ADG) of Pelibuey lambs during the pre-weaning period. Data of 62 lambs were used (35 males and 27 females; 32 of single lambing and 30 twins). All lambs were housed in individual pens with their mothers during the first 45 days post-lambing, without access to mother's food. Every 10 d, MI was estimated by the double lamb weighing method, and then, the intake of each milk component was calculated, considering both milk chemical composition and its estimated intake. The ADG was associated with MI and its components through a linear regression analysis. All milk variables were highly (P < 0.05; 0.69 ≤ r ≤ 0.85) correlated with ADG. The ADG explained (P < 0.0001) between 48 and 72% of the variation observed for MI and its components, being lower for fat intake and higher for MI. Given the positive slope parameter (ß1), an increase of one kilogram of body weight required intakes of 3.1 kg of milk, 117 g of fat, 203 g of protein, 187 g of lactose, or 16.8 MJ of energy. In conclusion, Pelibuey lambs to gain one kilogram of live weight during the pre-weaning period need to consume 3.1 kg of milk having an adequate amount of protein, fat, lactose, and energy.
Assuntos
Ingestão de Alimentos , Leite/química , Carneiro Doméstico/crescimento & desenvolvimento , Aumento de Peso , Animais , Feminino , Modelos Lineares , Masculino , DesmameRESUMO
In this research communication we describe the creation of an equation for the prediction of milk yield (MY) from udder volume (UV). A total of 280 measurements were collected between 5 and 15 d postpartum (pp) from 36 multiparous Pelibuey ewes. Study variables were measured between 2 and 9 weeks pp and MY was measured by manual milking, UV prior to and following milking was measured using the technique of making moulds from aluminium foil. The MY ranged from 0.09 to 0.83 kg/d, meanwhile UV prior and following milking ranged from 155 to 1940 and 90 to 1520 cm3, respectively. Measurements of UV had a moderate to high (P < 0.01; 0.58 ≤ r ≤ 0.78) correlation with MY. The UV prior to milking was the best prediction model for MY, which explained 62% of the variation in MY. This equation presented moderate precision (r2 = 0.61) and high accuracy (bias correction factor = 0.94), confirming a good reproducibility index (concordance correlation coefficient = 0.73). Modelling efficiency (MEF = 0.59) showed moderate concordance between observed and predicted values. In conclusion, MY in lactating Pelibuey ewes could be predicted in a moderate way using the predictor variable UV measured with the technique of moulds made with aluminium foil.
Assuntos
Lactação/fisiologia , Glândulas Mamárias Animais/anatomia & histologia , Leite , Ovinos/fisiologia , Animais , Feminino , Glândulas Mamárias Animais/fisiologia , Ovinos/anatomia & histologiaRESUMO
The objective of the present study was to determine the associations between the FAMACHA© score, body condition score (BCS) and several haematological parameters of Pelibuey ewes. In total, 52 clinically healthy Pelibuey ewes were evaluated in this study. A blood sample was collected from each animal to determine haematological variables. The BCS was evaluated on a scale of 1 to 5 and subsequently grouped into four categories: (1) < 1.5, (2) 2.0-2.5, (3) 3.0-3.5 and (4) > 4.0. The coloration of conjunctival mucosa was evaluated using the FAMACHA© system. The analysed haematological parameters were red blood cell (RBC) count, haemoglobin (HGB), haematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), red blood cell distribution width standard deviation (RDW-SD) and red blood cell distribution width coefficient of variation (RDW-CV). A descriptive statistical analysis was performed, and the Pearson correlation coefficients between variables were estimated. Haematological parameters were analysed with one-way analyses of variance, and the Tukey test was applied when significant differences were detected. The FAMACHA© score was only associated with MCHC (r = - 0.30, P < 0.05). The haematological parameters of RBC count, HCT, RDW-SD and RDW-CV statistically differed among BCS groups. The BCS was associated with RBC count, HCT, RDW-SD and RDW-CV, with r values ranging from 0.35 to 0.48. The haematological values were similar to those reported in ewes from Iran and Iraq. Hence, the optimization of the BCS calculation of Pelibuey ewes can generate association tables with haematological parameters to aid in the sanitary and feeding management of sheep.
Assuntos
Índices de Eritrócitos , Hemoglobinas , Carneiro Doméstico/fisiologia , Animais , Contagem de Eritrócitos/veterinária , Índices de Eritrócitos/veterinária , Feminino , Hematócrito/veterinária , Hemoglobinas/análise , Irã (Geográfico) , Carneiro Doméstico/sangueRESUMO
The overall goal of this study was to estimate the prevalence of gastrointestinal (GI) parasites in dogs in the city of Villahermosa in Tabasco, Mexico. The study population consisted of 302 owned dogs that had limited access to public areas. A fecal sample was collected from each animal and examined for GI parasites by conventional macroscopic analysis and centrifugal flotation. Fecal samples from 80 (26.5%) dogs contained GI parasites. Of these, 58 (19.2%) were positive for helminths and 22 (7.3%) were positive for protozoan parasites. At least seven parasitic species were identified. The most common parasite was Ancylostoma caninum which was detected in 48 (15.9%) dogs. Other parasites detected on multiple occasions were Cystoisospora spp. (n = 19), Toxocara canis (n = 7) and Giardia spp. (n = 3). Three additional parasites, Dipylidium caninum, Trichuris vulpis and Uncinaria spp., were each detected in a single dog. No mixed parasitic infections were identified. In summary, we report a moderately high prevalence of GI parasites in owned dogs in Villahermosa, Tabasco. Several parasitic species identified in this study are recognized zoonotic pathogens which illustrates the important need to routinely monitor and treat dogs that live in close proximity to humans for parasitic infections.
Assuntos
Doenças do Cão/epidemiologia , Enteropatias Parasitárias/veterinária , Animais , Doenças do Cão/parasitologia , Cães , Fezes/parasitologia , Enteropatias Parasitárias/epidemiologia , Enteropatias Parasitárias/parasitologia , México/epidemiologia , PrevalênciaRESUMO
Abstract The overall goal of this study was to estimate the prevalence of gastrointestinal (GI) parasites in dogs in the city of Villahermosa in Tabasco, Mexico. The study population consisted of 302 owned dogs that had limited access to public areas. A fecal sample was collected from each animal and examined for GI parasites by conventional macroscopic analysis and centrifugal flotation. Fecal samples from 80 (26.5%) dogs contained GI parasites. Of these, 58 (19.2%) were positive for helminths and 22 (7.3%) were positive for protozoan parasites. At least seven parasitic species were identified. The most common parasite was Ancylostoma caninum which was detected in 48 (15.9%) dogs. Other parasites detected on multiple occasions were Cystoisospora spp. (n = 19), Toxocara canis (n = 7) and Giardia spp. (n = 3). Three additional parasites, Dipylidium caninum, Trichuris vulpis and Uncinaria spp., were each detected in a single dog. No mixed parasitic infections were identified. In summary, we report a moderately high prevalence of GI parasites in owned dogs in Villahermosa, Tabasco. Several parasitic species identified in this study are recognized zoonotic pathogens which illustrates the important need to routinely monitor and treat dogs that live in close proximity to humans for parasitic infections.
Resumo O objetivo geral deste estudo foi estimar a prevalência de parasitas gastrointestinais (GI) em cães na cidade de Vilhahermosa, em Tabasco, México. A população estudada consistiu de 302 cães com donos, com acesso limitado a áreas públicas. Uma amostra fecal de cada animal foi coletada e examinada para parasitas GI por análise macroscópica convencional e centrífugo-flutuação. Amostras fecais de 80 (26,5%) cães apresentaram parasitas GI. Destes, 58 (19,2%) foram positivos para helmintos e 22 (7,3%) foram positivos para protozoários. Pelo menos 7 espécies parasitas foram identificadas. O parasita mais comum foi Ancylostoma caninum, detectado em 48 (15,9%) cães. Outros parasitas detectados em diversas ocasiões foram Cystoisospora spp. (n = 19), Toxocara canis (n = 7) e Giardia spp. (n=3). Adicionalmente, três parasitas foram detectados em apenas um cão – Dipylidium caninum, Trichuris vulpis e Uncinaria spp. Nenhuma infecção mista foi observada. Em resumo, nós identificamos neste trabalho uma prevalência moderadamente alta de parasitas GI em cães com donos, em Villahermosa, Tabasco. Várias espécies de parasitas identificados são reconhecidamente patógenos zoonóticos, o que indica a necessidade de monitorar rotineiramente e tratar infecções parasitárias em cães que vivem em proximidade a populações humanas.