Hypsarrhythmia is a specific chaotic morphology, present in the interictal period of the electroencephalogram (EEG) signal in patients with West Syndrome (WS), a severe form of childhood epilepsy and that, recently, was also identified in the examinations of patients with Zika Virus Congenital Syndrome (ZVCS). This innovative work proposes the development of a computational methodology for analysis and differentiation, based on the time-frequency domain, between the chaotic pattern of WS and ZVCS hypsarrhythmia. The EEG signal time-frequency analysis is carried out from the Continuous Wavelet Transform (CWT). Four joint moments—joint mean—\(\mu _{(t,f)}\), joint variance—\(\sigma _{(t,f)}^2\), joint skewness—\(\lambda _{(t,f)}\), and joint kurtosis—\(\kappa _{(t,f)}\)—and four entropy measurements—Shannon, Log Energy, Norm, and Sure—are obtained from the CWT to compose the representative feature vector of the EEG hypsarrhythmic signals under analysis. The performance of eight classical types of machine learning algorithms are verified in classification using the k-fold cross validation and leave-one-patient-out cross validation methods. Discrimination results provided 78.08% accuracy, 85.55% sensitivity, 73.21% specificity, and AUC = 0.89 for the ANN classifier.

This chapter proposes to analyze two configurations of neural networks to compose the expert set in the development of a multilevel speech signal pattern recognition system of 30 commands in the Brazilian Portuguese language. Then, multilayer perceptron (MLP) and learning vector quantization (LVQ) networks have their performances verified during the training, validation and test stages in the speech signal recognition, whose patterns are given by two-dimensional time matrices, result from mel-cepstral coefficients coding by the discrete cosine transform (DCT). In order to avoid the pattern separability problem, the patterns are modified by a nonlinear transformation to a high-dimensional space through a suitable set of Gaussian radial base functions (GRBF). The performance of MLP and LVQ experts is improved and configurations are trained with few examples of each modified pattern. Several combinations were performed for the neural network topologies and algorithms previously established to determine the network structures with the best hit and generalization results.

The diagnosis of epilepsy is conducted through visual inspection of electroencephalogram (EEG) signal recordings. However, due to the variations in convulsive disorders, it can be challenging for clinicians to constantly monitor the patient for seizure type, especially because EEG records contain hours of signal. Nevertheless, these patterns present in EEG signals can also be identified through signal classification methods based on signal processing and machine learning approaches. In light of this, this study proposes the development of a methodology for epileptic seizure type classification based on analysis of time-frequency characteristics of EEG signals, using Continuous Wavelet Transform (CWT) and joint moments of time-frequency distribution. Epileptic seizure classification was performed using a convolutional neural network (CNN), employing k-fold cross-validation methods. Accuracy, sensitivity, specificity, and area under the curve (AUC) metrics were obtained to validate this algorithm. The achieved results for the CNN classifier were 96.54% accuracy, 96.54% sensitivity, 96.54% specificity, and AUC = 0.90%.