Susana Hornillo-Mellado

Fast Technique for Noninvasive Fetal ECG Extraction

This letter describes a fast and very simple algorithm for estimating the fetal electrocardiogram (FECG). It is based on independent component analysis, but we substitute its computationally demanding calculations for a much simpler procedure. The resulting method consists of two steps: 1) a dimensionality reduction step and 2) a computationally light postprocessing stage used to enhance the FECG signal.

The Maternal Abdominal ECG as Input to MICA in the Fetal ECG Extraction Problem

This letter presents a successful system for recovering the fetal electrocardiogram using multidimensional ICA (MICA). MICA requires as many observations as sources. To increase the number of observations, MICA is often applied to data sets that include measurements taken at the mothers thoracic region. However, experiments suggest that the propagation from the maternal heart to the mothers abdomen is not a simple delay, and that approach may fail. Alternatively, our method first estimates the maternal ECG directly from the mothers abdomen. Then, inputs this estimated ECG to MICA. Experiments show superior performance as compared with the traditional approach.

Application of Independent Component Analysis to Edge Detection and Watermarking

It has been shown that a typical Independent Component Analysis (ICA) of an image provides a good estimation for the responses of the neurons of the visual cortex and their patterns of excitation. In this paper we apply ICA to obtain a decomposition of the image into a set of basis or features. These particular features can be used to develop a new algorithm for detecting the edges of the image. The performance of the proposed approach is demonstrated experimentally.

Independent component analysis based on first-order statistics

This communication puts forward a novel method for independent source extraction in instantaneous linear mixtures. The method is based on the conditional mean of the whitened observations and requires some prior knowledge of the positive support of the desired source. A theoretical performance analysis yields the closed-form expression of the asymptotic interference-to-signal ratio achieved by this technique. The analysis includes the effects of inaccuracies in the estimation of the positive support of the desired source in single-step and iterative implementations of the algorithm. Numerical experiments validate the fitness of the asymptotic approximations. As it is based on first-order statistics, the method is extremely cost-effective, which makes it an attractive alternative to second- and higher-order statistical techniques in power-limited scenarios.

Connections between ICA and sparse coding revisited

Recently, the application of Independent Component Analysis (ICA) to natural images has raised a great interest. Some outstanding features have been observed, like the sparse distribution of the independent components and the special appearance of the ICA bases (most of them look like edges). This paper provides a new insight on this behaviour, being supported by experimental results. In particular, a mathematical proof of the relation between ICA and sparse coding is given.

Theoretical Method for Solving BSS-ICA Using SVM

In this work we propose a new method for solving the blind source separation (BSS) problem using a support vector machine (SVM) workbench. Thus, we provide an introduction to SVM-ICA, a theoretical approach to unsupervised learning based on learning machines, which has frequently been proposed for classification and regression tasks. The key idea is to construct a Lagrange function from both the objective function and the corresponding constraints, by introducing a dual set of variables and solving the optimization problem. For this purpose we define a specific cost function and its derivative in terms of independence, i.e. inner products between the output and the objective function, transforming an unsupervised learning problem into a supervised learning machine task where optimization theory can be applied to develop effective algorithms.

Characterization of the Sources in Convolutive Mixtures: A Cumulant-Based Approach

This paper addresses the characterization of independent and non-Gaussian sources in a linear mixture. We present an eigensystem based approach to determine the number of independent components in the signal received by a single sensor. The temporal structure of the sources is also characterized using fourth-order statistics.

Independent component analysis: a low-complexity technique

This paper presents a new algorithm to solve the Independent Component Analysis (ICA) problem that has a very low computational complexity. The most remarkable feature of the proposed algorithm is that it does not need to compute higher-order statistics (HOS). In fact, the algorithm is based on trying to guess the sign of the independent components, after which it approximates the rest of the values.

Fast independent component analysis using a new property

In this paper we present a new theoretical characterization of the solutions to the Independent Component Analysis (ICA) problem. As an application, we also propose an algorithm that is directly based on that theoretical characterization. The algorithm has a very low computational complexity, and the experiments show that it is fast and reliable.

Effective blind separation of skewed sources

This article proposes a new algorithm for estimating skewed real signals in blind source separation. It is based on the eigendecomposition of a third-order statistics matrix. This matrix is updated iteratively by the algorithm in order to increase the robustness of the estimated source signals against numerical errors and perturbations. Simulations show the effectiveness of the approach.