M. Ipsale

A New Approach to Detection of Muscle Activation by Independent Component Analysis and Wavelet Transform

Recent works have 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. In this paper we propose to detect single muscle activation, when the arms reach a target, by means of ICs time-scale decomposition. Our analysis starts with acquisition of sEMG (surface EMG) signals; source separation is performed by a neural net-work that implements on Independent Component Analysis algorithm. In this way we obtain a signal set each representing single muscle activity. The wave-let transform, lastly, is utilised to detect muscle activation intervals.

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.

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.

Fuzzy approach to modeling scalar hysteresis

The aim of this paper was to discuss a new approach based on fuzzy logic able to model hysteresis loop of soft materials. In the fuzzy approach the five regions of magnetic field H and of magnetization M are described by membership functions.

A Comparison of ICA Algorithms in Biomedical Signal Processing

In the last years Independent Component Analysis (ICA) has been applied with success in signal processing and many algorithms have been developed in order to perform ICA. In this paper we review some algorithms, like INFOMAX (Bell and Sejnowski 1995), extended-INFOMAX (Lee, Girolami and Sejniowski 1997), FastICA (OjA, and Hyvarinen 1999), that solve the ICA problem under the assumption of the linear mixture model. We also show an overview of the nonlinear ICA algorithms and we discuss the MISEP (Almeida 2003). In order to test the performances of the reviewed algorithms, we present some applications of ICA in biomedical signal processing. In particular the application of ICA to the electroencephalographic (EEG) and surface electromyographic (sEMG) recordings are shown.

Independent component analysis for meaningful peaks detection in EEG signals

The electroencephalogram (EEG) signal is a useful and well known tool able to represent the brain electrical activity. In the EEG interpretation some peak activities could be related to some particular pathologic conditions, such as epilepsy or any other neurological damage. However, in most cases, high values of EEG signal represent an artifact activity. By means of an appropriate algorithm, we can understand if an high signal value is an artifact activity or it represents an irregular neural activity caused by some cerebral damage. In this paper some techniques, i.e. Independent Component Analysis (ICA) and Principal Component Analysis (PCA), are implemented in order to identify in EEG signals only those components that contain almost all the artifact contribution and that it is necessary to remove in order 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 EEG mapping is used to have a visual measure of the algorithm goodness.

Detecting Bioelectric Muscle Activity Corrupted by Superimposed Magnetic Resonance Field

Most clinical measurements rely on bioelectromagnetic phenomena. These events allow us to record electric or magnetic signals during the activity of living tissues. In this paper, we put our attention on the bioelectric fields that occur in the muscle activity. In fact during the body movements, the muscle contractions produce a bioelectric potential distribution that can be measured by putting the electrodes on the skin. In clinical applications, the monitoring of muscle activity can be performed in a noninvasive way, by placing a fixed number of electrodes on the skin surface; this technique is called surface electromyography (sEMG), and it is able to reveal the electric field generated by each muscle activity. Unfortunately, the sEMG suffers from the external electromagnetic fields. In recent years, many authors investigated the correlation between muscle and brain activity by performing the sEMG during the functional magnetic resonance imaging (fMRI). The fMRI is the most reliable technique to evaluate the brain activity because it allows us to obtain some images of the human body with very high resolutions. Unfortunately, joint measurement is a very difficult task because of the high electromagnetic interference between the resonance coils (very high magnetic fields) and the sEMG electrodes. In this paper, we present a method based on wavelet analysis to reveal sEMG in voluntary contractions when the measurement is made in a fMRI environmentMost of clinical measurements base their functionality on bioelectromagnetism, in particular in the measurements of the electric or magnetic signals produced by the activity of living tissues. The surface electromyography (sEMG) analysis is able to reveal the electric field generated by each muscle activity, whereas the magnetic resonance imaging (MRI) allows to obtain some images of the human body with very high resolutions, interpreting a magnetic field emitted after stimulation. The joint measurement is a very difficult task because of the high electromagnetic interference between the resonance coils (very high magnetic fields) and the EMG electrodes, that makes unrevealed the sEMG signal. In this paper we present a method to reveal sEMG signal in voluntary contractions when the measurement is made in a MRI environment

TIME-FREQUENCY ANALYSIS FOR CHARACTERIZING EMG SIGNALS DURING FMRI ACQUISITIONS

Research on human sensorimotor functions has hugely increased after electromyogram (EMG) analysis was replaced by functional magnetic resonance imaging (fMRI), that allows to obtain a direct visualization of the brain areas involved in motor control. Very meaningful results could be obtained if the two analysis could be correlated. Our goal is to acquire the EMG data during an fMRI task. The main problems in doing this are related to the electromagnetic compatibility between the resonance coils (very high magnetic fields) and the EMG electrodes. In this study we developed a system that can characterize the entire EMG signal corrupted by the magnetic fields generated by the magnetic resonance gradients. The entire system consists in a hardware equipment (shielded cables and wires) and a software analysis (effective mean analysis and wavelet analysis). The results show that a motor task was correctly delivered by our post processing analysis of the signal.

Intracranial Pressure Signal Processing by Adaptive Fuzzy Network

The aim of this work is the analysis of an intracranial pressure (ICP) signal, measured by means of an optical fiber catheter. We want to propose an alternative method to valuate the pressure inside the skull, without any knowledge of compliance curve, that can be valuated directly only by means of invasive and dangerous methods. First, we propose a classic Fourier processing in order to filter the ICP signal by its spectral components due at cardiac and respiratory activity. Then we perform the same analysis by wavelet transform, in order to implement a multiresolution analysis. The wavelet tool can perform also a very reliable data compression. We can demonstrate the advantages in using a neuro-fuzzy network on wavelet coefficients in order to obtain an optimal prediction of ICP signal. Various network structures are presented, in order to obtain several trade-off between computational time and prediction mean square error. Such analysis was performed by changing the fuzzy rule numbers, modifying the cluster size of the data. A real-time implementation was also proposed in order to allows the clinical applications.