Abstract

Abstract
This work presents a neural network used to diagnosis patients with benign or malignant breast cancer. The study is carried out using the Breast Cancer Wisconsin dataset. To solve the problem a feedforward neural network (NN) with multilayers was used. In the work, the implementation was made in Python, using two different libraries (sklearn and keras). Experimental results were obtained by performing simulations in both developed applications, and the performance of the neural classifier was evaluated through the performance measures of the classification systems and the ROC curve. The results were promising, since the NN was able to discriminate with high accuracy the two separable sets discriminating the benign or malignant tumor patients.

Abstract

Abstract

Abstract
This paper compares optimal path planning algorithms based on a Genetic Algorithm and a Particle Swarm Optimization algorithm applied to multiple bioinspired robots in a 2D environment simulation. The planning objectives are related to the harvesting of an apple plantation in which three swarm of butterflies were run, counting the fruits on the ground to optimize the harvest in a cooperative way. Robotic swarms must travel through points on the map to count the fruits. The time for each swarm was also counted for the comparison results.

Abstract
In this paper we describe a comparative classification of Pneumonia using Convolution Neural Network. The database used was the dataset Labeled Optical Coherence Tomography (OCT) and Chest X-Ray Images for Classification made available by (Kermany, 2018) with a total of 5863 images, with 2 classes: normal and pneumonia. To evaluate the generalization capacity of the models, cross-validation of k-fold was used. The classification models proved to be efficient compared to the work of (Kermany et al., 2018) which obtained 92.8 % and the present work had an average accuracy of 95.30 %.