Skip to main content
Figure 1 | Malaria Journal

Figure 1

From: Towards a precise test for malaria diagnosis in the Brazilian Amazon: comparison among field microscopy, a rapid diagnostic test, nested PCR, and a computational expert system based on artificial neural networks

Figure 1

General design of the Artificial Neural Network used by the MalDANN software. (A) The neural network used by the MalDANN software was based on the Multilayer Perceptron, which consists of: (i) one input layer, where the standards and data are presented to the neural network; (ii) intermediate (or hidden) layers, where all the processing of the neural network is performed; and (iii) one output layer, in which the result of the network is presented to the observer. (B) Two software versions were created using different neural network structures to perform the diagnosis of asymptomatic Plasmodium infections. One version used epidemiological variables, and plasma levels of IL-10 and IFN-gamma were added to the epidemiological variables in the second version. (C) The intuitive interface of the MalDANN software was developed in order to facilitate the input of the data into the artificial network. * First, 31 epidemiological variables were added to the system for data mining. Of these, five variables presented very strong association with the asymptomatic malaria. The same five variables were added to the MalDANN version that used cytokine data.

Back to article page