Computational models based on Artificial Intelligence to assess the thermal stress of dairy cattle through non-invasive measurements

• Grant number: 18/05989-4
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): August 01, 2018
• Effective date (End): January 31, 2020
• Field of knowledge: Agronomical Sciences - Agricultural Engineering
• Principal researcher: Rafael Vieira de Sousa
• Grantee: Alex Vinicius da Silva Rodrigues
• Home Institution: Faculdade de Zootecnia e Engenharia de Alimentos (FZEA). Universidade de São Paulo (USP). Pirassununga , SP, Brazil

Abstract

Recent research and innovation works seek the development of decision support tools for animal production systems. Among the main research topics are the development of technologies and computational models to estimate the performance and thermal stress of the animal through non-invasive measurements. Therefore, the objective of the work is the construction and evaluation of models, based on artificial intelligence, to estimate the physiological variables associated with the thermal stress of milk cattle. For the generation of the models will be considered environmental data such as temperature and humidity, and physiological data of rectal temperature, respiratory rate and temperature of body surfaces will be considered as obtained by infrared. The models for the evaluation of thermal stress will be based on the estimation of the rectal temperature or respiratory rate of cattle using data collected in the free stall. Different models based on artificial neural networks will be generated, varying the topology and the input sets by cross-validation method. The models will be analyzed and compared with regression conventional methods through the parameters of the dispersion diagram between the measured and estimated values (rectal temperature and respiratory rate). In addition, the prediction values generated by the best models based on artificial neural networks will be used to classify the individual stress level of each animal (normal, alert, danger and emergency), and will be compared with the classification generated by the Temperature Humidity Index (THI). (AU)