Thiago L. T. da Silveira

Bio: Thiago L. T. da Silveira is an academic researcher from Universidade Federal do Rio Grande do Sul. The author has contributed to research in topics: Discrete cosine transform & Computer science. The author has an hindex of 7, co-authored 24 publications receiving 240 citations. Previous affiliations of Thiago L. T. da Silveira include Universidade Federal de Santa Maria & University of Rio Grande.

Single-channel EEG sleep stage classification based on a streamlined set of statistical features in wavelet domain.

Abstract: The main objective of this study was to enhance the performance of sleep stage classification using single-channel electroencephalograms (EEGs), which are highly desirable for many emerging technologies, such as telemedicine and home care. The proposed method consists of decomposing EEGs by a discrete wavelet transform and computing the kurtosis, skewness and variance of its coefficients at selected levels. A random forest predictor is trained to classify each epoch into one of the Rechtschaffen and Kales’ stages. By performing a comprehensive set of tests on 106,376 epochs available from the Physionet public database, it is demonstrated that the use of these three statistical moments has enhanced performance when compared to their application in the time domain. Furthermore, the chosen set of features has the advantage of exhibiting a stable classification performance for all scoring systems, i.e., from 2- to 6-state sleep stages. The stability of the feature set is confirmed with ReliefF tests which show a performance reduction when any individual feature is removed, suggesting that this group of feature cannot be further reduced. The accuracies and kappa coefficients yield higher than 90 % and 0.8, respectively, for all of the 2- to 6-state sleep stage classification cases.

Automated drowsiness detection through wavelet packet analysis of a single EEG channel

Abstract: Ratio indices computed from a single EEG channel used as drowsiness indicators.Delta and gamma brain rhythms successfully used for drowsiness detection.Wavelet packet transform as the main tool to localize specific brain frequency ranges.Transition from alert to drowsy state is taken as main event of interest.Wilcoxon signed rank test analysis pointed out the contribution of proposed indices. Advances in materials engineering, electronic circuits, sensors, signal processing and classification techniques have allowed computational systems to interpret biological quantities, recognizing physiological conditions. The next scientific challenge is to turn those technologies portable, wearable or even implantable, above all, being energy efficient. A prospective application for the next generation of portable electroencephalogram (EEG) signal processing systems is hazard prevention in attention-demanding activities. EEG keeps closest connection to the preoptic area where sleep is originated. In this paper, a methodology for assessing alertness level based on a single EEG channel (Pz-Oz) is proposed, allowing the reduction of the required hardware and the computational time of the algorithms, besides being more portable than multi-channel based ones. Two new spectral power-based indices (i) γ/? and (ii) ( γ + β )/( ? + α ) are computed from EEG rhythms through the normalized Haar discrete wavelet packet transform (WPT). The Haar WPT allows precisely resolving the brain rhythms into packets whilst demanding a relatively low computational cost. The effectiveness of the proposed indices in drowsiness detection is evaluated by comparison with five indices originally proposed for multi-channel processing. Statistical Wilcoxon signed rank test is applied to evaluate the performance of the entire set of indices, evidencing the significant changes in the alert-drowsy transitions of 20 subjects of a public database. The proposed indices (ii) and (i) presented the most and second more significant p-Values (p < 0.001 and p?=?0.001), respectively.

Superpixel Image Classification with Graph Attention Networks

Abstract: This paper presents a methodology for image classification using Graph Neural Network (GNN) models. We transform the input images into region adjacency graphs (RAGs), in which regions are superpixels and edges connect neighboring superpixels. Our experiments suggest that Graph Attention Networks (GATs), which combine graph convolutions with self-attention mechanisms, outperforms other GNN models. Although raw image classifiers perform better than GATs due to in-formation loss during the RAG generation, our methodology opens an interesting avenue of research on deep learning beyond rectangular-gridded images, such as 360-degree field of view panoramas. Traditional convolutional kernels of current state-of-the-art methods cannot handle panoramas, whereas the adapted superpixel algorithms and the resulting region adjacency graphs can naturally feed a GNN, without topology issues.

Low-complexity 8-point DCT approximation based on angle similarity for image and video coding

Abstract: The principal component analysis (PCA) is widely used for data decorrelation and dimensionality reduction. However, the use of PCA may be impractical in real-time applications, or in situations were energy and computing constraints are severe. In this context, the discrete cosine transform (DCT) becomes a low-cost alternative to data decorrelation. This paper presents a method to derive computationally efficient approximations to the DCT. The proposed method aims at the minimization of the angle between the rows of the exact DCT matrix and the rows of the approximated transformation matrix. The resulting transformations matrices are orthogonal and have extremely low arithmetic complexity. Considering popular performance measures, one of the proposed transformation matrices outperforms the best competitors in both matrix error and coding capabilities. Practical applications in image and video coding demonstrate the relevance of the proposed transformation. In fact, we show that the proposed approximate DCT can outperform the exact DCT for image encoding under certain compression ratios. The proposed transform and its direct competitors are also physically realized as digital prototype circuits using FPGA technology.

Multiplierless 16-point DCT approximation for low-complexity image and video coding

Abstract: An orthogonal 16-point approximate discrete cosine transform (DCT) is introduced. The proposed transform requires neither multiplications nor bit-shifting operations. A fast algorithm based on matrix factorization is introduced, requiring only 44 additions—the lowest arithmetic cost in literature. To assess the introduced transform, computational complexity, similarity with the exact DCT, and coding performance measures are computed. Classical and state-of-the-art 16-point low-complexity transforms were used in a comparative analysis. In the context of image compression, the proposed approximation was evaluated via PSNR and SSIM measurements, attaining the best cost–benefit ratio among the competitors. For video encoding, the proposed approximation was embedded into a HEVC reference software for direct comparison with the original HEVC standard. Physically realized and tested using FPGA hardware, the proposed transform showed 35 and 37 % improvements of area-time and area-time-squared VLSI metrics when compared to the best competing transform in the literature.

Multi-kernel extreme learning machine for EEG classification in brain-computer interfaces

Abstract: Multi-kernel extreme learning machine based method is proposed for EEG classification.Supplementary information from different kernels are integrated for better accuracy.Extensive experimental comparison confirms superiority of the proposed method. One of the most important issues for the development of a motor-imagery based brain-computer interface (BCI) is how to design a powerful classifier with strong generalization capability. Extreme learning machine (ELM) has recently proven to be comparable or more efficient than support vector machine for many pattern recognition problems. In this study, we propose a multi-kernel ELM (MKELM)-based method for motor imagery electroencephalogram (EEG) classification. The kernel extension of ELM provides an elegant way to circumvent calculation of the hidden layer outputs and inherently encode it in a kernel matrix. We investigate effects of two different kernel functions (i.e., Gaussian kernel and polynomial kernel) on the performance of kernel ELM. The MKELM method is subsequently developed by integrating these two types of kernels with a multi-kernel learning strategy, which can effectively explore the supplementary information from multiple nonlinear feature spaces for more robust classification of EEG. An extensive experimental comparison with two public EEG datasets indicates that the MKELM method gives higher classification accuracy than those of the other competing algorithms. The experimental results confirm that superiority of the proposed MKELM-based method for accurate classification of EEG associated with motor imagery in BCI applications. Our method also provides a promising and generalized solution to investigate the complex and nonlinear information for various applications in the fields of expert and intelligent systems.

A Deep Learning Model for Automated Sleep Stages Classification Using PSG Signals.

Abstract: Sleep disorder is a symptom of many neurological diseases that may significantly affect the quality of daily life. Traditional methods are time-consuming and involve the manual scoring of polysomnogram (PSG) signals obtained in a laboratory environment. However, the automated monitoring of sleep stages can help detect neurological disorders accurately as well. In this study, a flexible deep learning model is proposed using raw PSG signals. A one-dimensional convolutional neural network (1D-CNN) is developed using electroencephalogram (EEG) and electrooculogram (EOG) signals for the classification of sleep stages. The performance of the system is evaluated using two public databases (sleep-edf and sleep-edfx). The developed model yielded the highest accuracies of 98.06%, 94.64%, 92.36%, 91.22%, and 91.00% for two to six sleep classes, respectively, using the sleep-edf database. Further, the proposed model obtained the highest accuracies of 97.62%, 94.34%, 92.33%, 90.98%, and 89.54%, respectively for the same two to six sleep classes using the sleep-edfx dataset. The developed deep learning model is ready for clinical usage, and can be tested with big PSG data.

Learning sleep stages from radio signals: a conditional adversarial architecture

Abstract: We focus on predicting sleep stages from radio measurements without any attached sensors on subjects. We introduce a new predictive model that combines convolutional and recurrent neural networks to extract sleep-specific subject-invariant features from RF signals and capture the temporal progression of sleep. A key innovation underlying our approach is a modified adversarial training regime that discards extraneous information specific to individuals or measurement conditions, while retaining all information relevant to the predictive task. We analyze our game theoretic setup and empirically demonstrate that our model achieves significant improvements over state-of-the-art solutions.

Neural network analysis of sleep stages enables efficient diagnosis of narcolepsy

Abstract: Analysis of sleep for the diagnosis of sleep disorders such as Type-1 Narcolepsy (T1N) currently requires visual inspection of polysomnography records by trained scoring technicians. Here, we used neural networks in approximately 3,000 normal and abnormal sleep recordings to automate sleep stage scoring, producing a hypnodensity graph—a probability distribution conveying more information than classical hypnograms. Accuracy of sleep stage scoring was validated in 70 subjects assessed by six scorers. The best model performed better than any individual scorer (87% versus consensus). It also reliably scores sleep down to 5 s instead of 30 s scoring epochs. A T1N marker based on unusual sleep stage overlaps achieved a specificity of 96% and a sensitivity of 91%, validated in independent datasets. Addition of HLA-DQB1*06:02 typing increased specificity to 99%. Our method can reduce time spent in sleep clinics and automates T1N diagnosis. It also opens the possibility of diagnosing T1N using home sleep studies.