Artificial neural networks to predict the presence of Neosporosis in cattle.
Journal:
Mathematical biosciences and engineering : MBE
PMID:
40296806
Abstract
The prediction of bovine infectious diseases is a constant challenge as generally, only laboratory data is available not allowing the study of their relationship with each disease's risk factors. The diseases neosporosis and bovine viral diarrhea, which are present in Colombia, the United States, Mexico, Brazil, and Argentina, cause reproductive problems in cattle and generate economic losses for ranchers. Although there are mathematical models that can evaluate which cattle are susceptible to these diseases, these provide limited information, maintaining the need for a model that provides information on both transmission and mechanisms for controlling the disease. In this article, a machine learning model is presented that combines laboratory data with risk factors in a neural network to predict the presence of bovine neosporosis. The proposed model was implemented with data from previous studies conducted in the municipality of Sotaquirá, Boyacá, Colombia, and obtained an accuracy of 94% in predicting the presence of the disease. It can be concluded that incorporating laboratory data into machine learning algorithms improves the prediction of the presence of these diseases. Furthermore, the proposed system not only predicts but also provides useful information for clinical decision-making, making it a valuable tool in the veterinary field.