F. La Foresta

Automatic Artifact Rejection From Multichannel Scalp EEG by Wavelet ICA

Electroencephalographic (EEG) recordings are often contaminated by artifacts, i.e., signals with noncerebral origin that might mimic some cognitive or pathologic activity, this way affecting the clinical interpretation of traces. Artifact rejection is, thus, a key analysis for both visual inspection and digital processing of EEG. Automatic artifact rejection is needed for effective real time inspection because manual rejection is time consuming. In this paper, a novel technique (Automatic Wavelet Independent Component Analysis, AWICA) for automatic EEG artifact removal is presented. Through AWICA we claim to improve the performance and fully automate the process of artifact removal from scalp EEG. AWICA is based on the joint use of the Wavelet Transform and of ICA: it consists of a two-step procedure relying on the concepts of kurtosis and Renyis entropy. Both synthesized and real EEG data are processed by AWICA and the results achieved were compared to the ones obtained by applying to the same data the “wavelet enhanced” ICA method recently proposed by other authors. Simulations illustrate that AWICA compares favorably to the other technique. The method here proposed is shown to yield improved success in terms of suppression of artifact components while reducing the loss of residual informative data, since the components related to relevant EEG activity are mostly preserved.

Neural-ICA and wavelet transform for artifacts removal in surface EMG

Recent works have shown that artifacts removal in biomedical signals, like electromyographic (EMG) or electroencephalographic (EEG) recordings, can be performed by using discrete wavelet transform (DWT) or independent component analysis (ICA). Often, the removal of some artifacts is very hard because they are superimposed on the recordings and they corrupt biomedical signals also in frequency domain. In these cases DWT and ICA methods cannot perform artifacts cancellation. We present a method based on the joint use of wavelet transform and independent component analysis. We show the obtained results and the comparisons among the proposed method, DWT and ICA techniques. In this preliminary study, a user interface is needed to identify the artifact.

Wavelet-ICA methodology for efficient artifact removal from Electroencephalographic recordings

Electroencephalographic (EEG) recordings are often contaminated by the artifacts, signals that have non-cerebral origin and that might mimic cognitive or pathologic activity and therefore distort the analysis of EEG. In this paper the issue of artifact extraction from Electroencephalographic data is addressed and a new technique for EEG artifact removal, based on the joint use of Wavelet transform and Independent Component Analysis (WICA), is presented and compared to two other techniques based on ICA and wavelet denoising. An artificial artifact-laden EEG dataset was created mixing a real EEG with a set of synthesized artifacts. This dataset was processed by WICA and the two other methods. The proposed technique had the best artifact separation performance for every kind of artifact also allowing for the minimum information loss.

Brain Activity Investigation by EEG Processing: Wavelet Analysis, Kurtosis and Renyi's Entropy for Artifact Detection

Electroencephalographic (EEG) recordings are employed in order to investigate the brain activity in neuro pathological subjects. Unfortunately EEG are often contaminated by the artifacts, signals that have non-cerebral origin and that might mimic cognitive or pathologic activity and therefore distort the analysis of EEG. In this paper we propose a multiresolution analysis, based on EEG wavelet processing, to extract the cerebral EEG rhythms. We also present a method based on Renyis entropy and kurtosis to automatically identify the Wavelet components affected by artifacts. Finally, we discuss as the joint use of wavelet analysis, kurtosis and Renyis entropy allows for a deeper investigation of the brain activity and we discuss the capability of this technique to become an efficient preprocessing step to optimize artifact rejection from EEG. This is the first technique that exploits the peculiarities of EEG to optimize EEG artifact detection.

Remarks about Preisach function approximation using Lorentzian function and its identification for nonoriented steels

In this paper, we will discuss the use of the Lorentzian function as a possible candidate for the accurate approximation of the Preisach function in the modeling of scalar hysteresis for nonoriented grain steels. In particular we discuss here the identification procedure of the parameters of the function from measured data.

Spatio-temporal analysis of surface electromyography signals by independent component and time-scale analysis

Recent work has demonstrated that the Independent Components (ICs) of simultaneously-recorded surface Electromyography (sEMG) recordings are more reliable in monitoring repetitive movements and better correspond with ongoing brain-wave activity than raw sEMG recordings. With this implementation, static spatial filters are used to create linear combinations of sEMGs that are maximally independent. We propose extending this formulation by applying time-scale analysis to the computed ICs to assess the critical assumption of stationarity of the spatial filters. We applied combined ICA/time-scale analyses to sEMGs recorded from the arm and shoulder from a single normal subject making repetitive pointing movements. Several of the ICs were clearly modulated in phase with the pointing movements, but still varied from trial-to-trial. Time-scale analysis allowed the extraction of that portion of each IC that was consistent across trials. Our results suggest that the ICs derived by static filters used in the ICA-only implementation may be affected by non-stationarities in the relation between sEMG signals, and that the combined ICA/wavelet approach is a practical means to extract highly reproducible features in sEMG recordings.

SVM classification of epileptic EEG recordings through multiscale permutation entropy

Electroencephalogram (EEG) is a non-invasive diagnostic tool in clinical neurophysiology, especially with respect to epilepsy. The epileptic status is characterized by reduced complexity. New markers, based on nonlinear dynamics, like Permutation Entropy (PE) have been developed to measure EEG complexity. In this paper, Multiscale Permutation Entropy (MPE) complexity measure is proposed as a potentially useful framework for detecting epileptic events in EEG data and to distinguish healthy controls from patients. The achieved results show that: 1) MPE is able to discriminate between the two categories; 2) the use of multiple scales may substantially improve the specificity of the diagnosis. This is shown through an SVM-based classification network with three different kernels. The use of the SVM approach is also useful to infer clues about the extracted features.

Discovering Network Phenomena in the Epileptic Electroencephalography through Permutation Entropy Mapping

The genesis of epileptic seizures is nowadays still mostly unknown. The hypothesis that most of scientist share is that an abnormal synchronization of different groups of neurons seems to trigger a recruitment mechanism that leads the brain to the seizure in order to reset this abnormal condition. If we aim to understand the genesis of epileptic seizures so that we can be able to control them, we need a system to be able to detect, follow and interfere with these dynamics. The aim of this paper is to introduce a technique to detect these phenomena and, to this purpose, a powerful brain mapping based on the Permutation Entropy (PE) is proposed. PE topography is constructed and then modelled in order to come up with a spatiotemporal clustering of the areas of the brain. This technique was tested over 24 EEG dataset from patients affected by absence seizures and on 40 EEG from healthy subjects. The results show an abnormal coupling among the electrodes that will be involved in seizure development. In particular, the frontal/temporal area (critical during the ictal stages in these patients) appears steadily associated to higher PE levels, compared to the rest of the brain, even during the interictal stages.

PCA and ICA for the extraction of EEG components in cerebral death assessment

The electroencephalogram (EEG) analysis provides a functional tool to verify a qualitative clinical check. In this paper some techniques, i.e. principal component analysis (PCA) and independent component analysis (ICA), are implemented in order to extrapolate in EEG signals very few dominant components that contain almost all the information necessary to have an adequate knowledge of the brain activity. To obtain that, the compression ability of PCA is mixed with the statistical independence property of the ICA. The achieved results show that in most cases of cerebral death diagnosis, in which the EEG analysis is performed when the brain activity is very reduced, even few components are enough to depict the complete brain activity.

Intracranial pressure signals forecasting with wavelet transform and neuro-fuzzy network

In this paper, we propose a novel approach to forecast the time evolution of the intracranial pressure (ICP) signal acquired by means of optical fiber catheters in patients with neurological pathologies. The proposed processing system uses wavelet transform and neuro-fuzzy network (NFN) to make the forecast. It is noteworthy to underline that a forecasting approach based only on NFN would provide information limited to a few samples. The advantage of this method is that it turns to account both the wavelet ability to focus information in few parameters and the NFN performance to expand the forecasting to a larger set, improving the agreement between the experimental data and the predicted ones.