R
Romis Attux
Researcher at State University of Campinas
Publications - 172
Citations - 1451
Romis Attux is an academic researcher from State University of Campinas. The author has contributed to research in topics: Blind signal separation & Source separation. The author has an hindex of 18, co-authored 168 publications receiving 1167 citations.
Papers
More filters
Book
Unsupervised Signal Processing: Channel Equalization and Source Separation
TL;DR: Channel Equalization Source Separation and the Unsupervised Deconvolution Problem Fundamental Theorems Bussgang Algorithms The Shalvi-Weinstein Algorithm.
Journal ArticleDOI
A deep convolutional neural network for COVID-19 detection using chest X-rays
Pedro R. A. S. Bassi,Romis Attux +1 more
TL;DR: In this article, Xu et al. used layer-wise relevance propagation (LRP) to generate heatmaps of chest X-ray images to improve the interpretability of deep neural networks.
Journal ArticleDOI
Comparative analysis of strategies for feature extraction and classification in SSVEP BCIs
Sarah N. Carvalho,Sarah N. Carvalho,Thiago Bulhões da Silva Costa,Luisa Fernanda Suarez Uribe,Diogo C. Soriano,Glauco Ferreira Gazel Yared,Luis Coradine,Romis Attux +7 more
TL;DR: A comparative analysis of different signal processing techniques for each BCI system stage concerning steady state visually evoked potentials (SSVEP), which includes feature extraction performed by different spectral methods, leads to a representative and helpful comparative overview of robustness and efficiency of classical strategies.
Journal ArticleDOI
A Deep Convolutional Neural Network for COVID-19 Detection Using Chest X-Rays
Pedro R. A. S. Bassi,Romis Attux +1 more
TL;DR: The state-of-the-art performances achieved show that chest X-rays can become a cheap and accurate auxiliary method for COVID-19 diagnosis and improve the interpretability of the deep neural networks and indicate an analytical path for future research on diagnosis.
Journal ArticleDOI
2012 Special Issue: An extended echo state network using Volterra filtering and principal component analysis
TL;DR: A novel architecture for an ESN in which the linear combiner is replaced by a Volterra filter structure is presented and the principal component analysis technique is used to reduce the number of effective signals transmitted to the output layer.