Detecção de mastite subclínica bovina utilizando a termografia por infravermelho

• Autor(a) principal: SILVA, Rodes Angelo Batista da
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFRPE
• Texto Completo: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/8285
• topic: Mastite; Diagnóstico; Termografia infravermelha; CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA

This research was conducted with the objective of evaluating the use of infrared thermography as a diagnostic tool for detecting subclinical mastitis in dairy cattle. Besides characterizing the spatial variability of the surface temperature of the udder of the animals and developing computational methodology for classification of mastitis tables, as a tool to support decision making. Data collection was carried out in a milk production unit, Fazenda Roçadinho, located in the municipality of Capoeiras, Mesoregion Agreste, Microregion of Vale do Ipojuca, State of Pernambuco. The quantity of samples was determined according to the selection criteria and totalled 24 animals, under different clinical conditions (healthy, with clinical and subclinical mastitis). The physiological data of the udder surface temperature (TS, °C) Were recorded; Eyeball temperature (TO, °C); Rectal temperature (TR, °C); Respiratory rate (RR, mov. min-1) and environment variables, air temperature (Tar, °C) and relative air humidity (RH, %). The thermal images of the udder of the animals were obtained from a thermographic camera, in the left anterolateral, anterolateral right, posterior and inferior, four images per animal, totaling 96 images for analysis of their respective Mamary rooms. Thermography allowed To Identify the temperature difference of the surface of the mamary rooms. The animals with positive classification for subclinical mastitis had values between 33.2 ± 0.67 ºC and 34.64 ± 1.07 ºC; For the negative rooms, values between 29.3 ± 1.78 ºC and 32.24 ± 0.62 ºC. The individuals with clinical mastitis had a temperature between 34.0 and 37.5 ºC. The geostatistical analyses successfully identified the spatial dependence and the main component technique allowed to verify the correlation of the variables TO, TR, FR, Tar and RH with the clinical picture of the animals and the degree of dependence between the variables Studied. The software developed proved to be efficient in classifying thermal images for the detection of clinical conditions of mastitis, with accuracy of 90.9%, specificity of 57.14% and sensitivity of 85.71% of the algorithm implemented for the analyses.