ICA with Reference

Exploring the variability of single trials in somatosensory evoked responses using constrained source extraction and RMT

This paper describes the theoretical background of a new data-driven approach to encephalographic single-trial (ST) data analysis. Temporal constrained source extraction using sparse decomposition identifies signal topographies that closely match the shape characteristics of a reference signal, one response for each ST. The correlations between these ST topographies are computed for formal Correlation Matrix Analysis (CMA) based on Random Matrix Theory (RMT). The RMT-CMA provides clusters of similar ST topologies in a completely unsupervised manner. These patterns are then classified into deterministic set and noise using well established RMT results. The efficacy of the method is applied to EEG and MEG data of somatosensory evoked responses (SERs). The results demonstrate that the method can recover brain signals with time course resembling the reference signal and follow changes in strength and/or topography in time by simply stepping the reference signal through time.

A Novel Extraction Method of Fetal Electrocardiogram From the Composite Abdominal Signal

In contrast to the ultrasonic measurement of fetal heart motion, the fetal electrocardiogram (ECG) provides clinically significant information concerning the electrophysiological state of a fetus. In this paper, a novel method for extracting the fetal ECG from abdominal composite signals is proposed. This method consists of the cancellation of the mother's ECG and blind source separation with the reference signal (BSSR). The cancellation of the mother's ECG component was performed by subtracting the linear combination of mutually orthogonal projections of the heart vector. The BSSR is a fixed-point algorithm, the Lagrange function of which includes the higher order cross-correlation between the extracted signal and the reference signal as the cost term rather than a constraint. This realizes the convexity of the Lagrange function in a simple form, which guarantees the convergence of the algorithm. By practical application, the proposed method has been shown to be able to extract the P and T waves in addition to the R wave. The reliability and accuracy of the proposed method was confirmed by comparing the extracted signals with the directly recorded ECG at the second stage of labor. The gestational age-dependency of the physiological parameters of the extracted fetal ECG also coincided well with that of the magnetocardiogram, which proves the clinical applicability of the proposed method.

Epileptic seizure predictability from scalp EEG incorporating constrained blind source separation

Most of the methods for prediction of epilepsy recently reported in the literature are based on the evaluation of chaotic behavior of intracranial electroencephalographic (EEG) recordings. These recordings require intensive surgical operations to implant the electrodes within the brain which are hazardous to the patient. Here, we have developed a novel approach to quantify the dynamical changes of the brain using the scalp EEG. The scalp signals are preprocessed by means of an effective block-based blind source separation (BSS) technique to separate the underlying sources within the brain. The algorithm significantly removes the effect of eye blinking artifacts. An overlap window procedure has been incorporated in order to mitigate the inherent permutation problem of BSS and maintain the continuity of the estimated sources. Chaotic behavior of the underlying sources has then been evaluated by measuring the largest Lyapunov exponent. For our experiments, we provided twenty sets of simultaneous intracranial and scalp EEG recordings from twenty patients. The above recordings have been compared. Similar results were obtained when the intracranial electrodes recorded the electrical activity of the epileptic focus. Our preliminary results show a great improvement when the epileptic focus is not captured by the intracranial electrodes.

Temporally constrained ICA: an application to artifact rejection in electromagnetic brain signal analysis

Independent component analysis (ICA) is a technique which extracts statistically independent components from a set of measured signals. The technique enjoys numerous applications in biomedical signal analysis in the literature, especially in the analysis of electromagnetic (EM) brain signals. Standard implementations of ICA are restrictive mainly due to the square mixing assumption-for signal recordings which have large numbers of channels, the large number of resulting extracted sources makes the subsequent analysis laborious and highly subjective. There are many instances in neurophysiological analysis where there is strong a priori information about the signals being sought; temporally constrained ICA (cICA) can extract signals that are statistically independent, yet which are constrained to be similar to some reference signal which can incorporate such a priori information. We demonstrate this method on a synthetic dataset and on a number of artifactual waveforms identified in multichannel recordings of EEG and MEG. cICA repeatedly converges to the desired component within a few iterations and subjective analysis shows the waveforms to be of the expected morphologies and with realistic spatial distributions. This paper shows that cICA can be applied with great success to EM brain signal analysis, with an initial application in automating artifact extraction in EEG and MEG.