Neonatal incubators are hospital medical equipment used in the treatment and monitoring of newborns. To deal with patients that are sensitive to temperature changes, it is necessary to have certified equipment, to ensure that it works properly. However, during certification, it is necessary to interrupt the operation of the incubator, directly affecting the number of newborns assisted with this equipment. Thus, this article aims to
identify the temperature and humidity dynamic data performed through artificial neural networks of Multilayer Perceptron (MLP) and Radial Basis Function (RBF) types. The results obtained through these neural models can be used in the design of controllers, or even as virtual sensors. As a goal of this work, it
is expected a better functioning and monitoring of the incubator, in addition to the optimization of the certification process, when associated with the instability of temperature and humidity, enabling a higher number of incubators in operation to exist, as well as an efficient reading of its data.