Roberto Gil-Pita

Cole equation and parameter estimation from electrical bioimpedance spectroscopy measurements - A comparative study

Since there are several applications of Electrical Bioimpedance (EBI) that use the Cole parameters as base of the analysis, to fit EBI measured data onto the Cole equation is a very common practice within Multifrequency-EBI and spectroscopy. The aim of this paper is to compare different fitting methods for EBI data in order to evaluate their suitability to fit the Cole equation and estimate the Cole parameters. Three of the studied fittings are based on the use of Non-Linear Least Squares on the Cole model, one using the real part only, a second using the imaginary part and the third using the complex impedance. Furthermore, a novel fitting method done on the Impedance plane, without using any frequency information has been implemented and included in the comparison. Results show that the four methods perform relatively well but the best fitting in terms of Standard Error of Estimate is the fitting obtained from the resistance only. The results support the possibility of measuring only the resistive part of the bioimpedance to accurately fit Cole equation and estimate the Cole parameters, with entailed advantages.

EVOLVING EDITED k-NEAREST NEIGHBOR CLASSIFIERS

The k-nearest neighbor method is a classifier based on the evaluation of the distances to each pattern in the training set. The edited version of this method consists of the application of this classifier with a subset of the complete training set in which some of the training patterns are excluded, in order to reduce the classification error rate. In recent works, genetic algorithms have been successfully applied to determine which patterns must be included in the edited subset. In this paper we propose a novel implementation of a genetic algorithm for designing edited k-nearest neighbor classifiers. It includes the definition of a novel mean square error based fitness function, a novel clustered crossover technique, and the proposal of a fast smart mutation scheme. In order to evaluate the performance of the proposed method, results using the breast cancer database, the diabetes database and the letter recognition database from the UCI machine learning benchmark repository have been included. Both error rate and computational cost have been considered in the analysis. Obtained results show the improvement achieved by the proposed editing method.

Cole Parameter Estimation from the Modulus of the Electrical Bioimpeadance for Assessment of Body Composition. A Full Spectroscopy Approach.

Abstract Activities around applications of Electrical Bioimpedance Spectroscopy (EBIS) have proliferated in the past decade significantly. Most of these activities have been focused in the analysis of the EBIS measurements, which eventually might enable novel applications. In Body Composition Assessment (BCA), the most common analysis approach currently used in EBIS is based on the Cole function, which most often requires curve fitting. One of the most implemented approaches for obtaining the Cole parameters is performed in the impedance plane through the geometrical properties that the Cole function exhibit in such domain as depressed semi-circle. To fit the measured impedance data to a semi-circle in the impedance plane, obtaining the Cole parameters in an indirect and sequential manner has several drawbacks. Applying a Non-Linear Least Square (NLLS) iterative fitting on the spectroscopy measurement, obtains the Cole parameters considering the frequency information contained in the measurement. In this work, from experimental total right side EBIS measurements, the BCA parameters have been obtained to assess the amount and distribution of whole body fluids. The values for the BCA parameters have been obtained using values for the Cole parameters estimated with both approaches: circular fitting on the impedance plane and NLLS impedance-only fitting. The comparison of the values obtained for the BCA parameters with both methods confirms that the NLLS impedance-only is an effective alternative as Cole parameter estimation method in BCA from EBIS measurements. Using the modulus of the Cole function as the model for the fitting would eliminate the need for performing phase detection in the acquisition process, simplifying the hardware specifications of the measurement instrumentation when implementing a bioimpedance spectrometer.

A novel approach for removing the hook effect artefact from Electrical Bioimpedance spectroscopy measurements

Very often in Electrical Bioimpedance (EBI) spectroscopy measurements the presence of stray capacitances creates a measurement artefact commonly known as Hook Effect. Such an artefact creates a hook-alike deviation of the EBI data noticeable when representing the measurement on the impedance plane. Such Hook Effect is noticeable at high frequencies but it also causes a data deviation at lower measurement frequencies. In order to perform any accurate analysis of the EBI spectroscopy data, the influence of the Hook Effect must be removed. An established method to compensate the hook effect is the well known Td compensation, which consists on multiplying the obtained spectrum, Zmeas(ω) by a complex exponential in the form of exp[jωTd]. Such a method cannot correct entirely the Hook Effect since the hook-alike deviation occurs a broad frequency range in both magnitude and phase of the measured impedance, and by using a scalar value for Td. First a scalar only modifies the phase of the measured impedance and second, a single value can truly corrects the Hook Effect only at a single frequency. In addition, the process to select a value for the scalar Td by an iterative process with the aim to obtain the best Cole fitting lacks solid scientific grounds. In this work the Td compensation method is revisited and a modified approach for correcting the Hook Effect including a novel method for selecting the correcting values is proposed. The initial validation results confirm that the proposed method entirely corrects the Hook Effect at all frequencies.

Experimental validation of a method for removing the capacitive leakage artifact from electrical bioimpedance spectroscopy measurements

Often when performing electrical bioimpedance (EBI) spectroscopy measurements, the obtained EBI data present a hook-like deviation, which is most noticeable at high frequencies in the impedance plane. The deviation is due to a capacitive leakage effect caused by the presence of stray capacitances. In addition to the data deviation being remarkably noticeable at high frequencies in the phase and the reactance spectra, the measured EBI is also altered in the resistance and the modulus. If this EBI data deviation is not properly removed, it interferes with subsequent data analysis processes, especially with Cole model-based analyses. In other words, to perform any accurate analysis of the EBI spectroscopy data, the hook deviation must be properly removed. Td compensation is a method used to compensate the hook deviation present in EBI data; it consists of multiplying the obtained spectrum, Zmeas(ω), by a complex exponential in the form of exp(–jωTd). Although the method is well known and accepted, Td compensation cannot entirely correct the hook-like deviation; moreover, it lacks solid scientific grounds. In this work, the Td compensation method is revisited, and it is shown that it should not be used to correct the effect of a capacitive leakage; furthermore, a more developed approach for correcting the hook deviation caused by the capacitive leakage is proposed. The method includes a novel correcting expression and a process for selecting the proper values of expressions that are complex and frequency dependent. The correctness of the novel method is validated with the experimental data obtained from measurements from three different EBI applications. The obtained results confirm the sufficiency and feasibility of the correcting method.

Multilayer perceptrons applied to traffic sign recognition tasks

The work presented in this paper suggests a Traffic Sign Recognition (TSR) system whose core is based on a Multilayer Perceptron (MLP). A pre-processing of the traffic sign image (blob) is applied before the core. This operation is made to reduce the redundancy contained in the blob, to reduce the computational cost of the core and to improve its performance. For comparison purposes, the performance of the a statistical method like the k-Nearest Neighbour (k-NN) is included. The number of hidden neurons of the MLP is studied to obtain the value that minimizes the total classification error rate. Once obtained the best network size, the results of the experiments with this parameter show that the MLP achieves a total error probability of 3.85%, which is almost the half of the best obtained with the k-NN.

Study of Two Error Functions to Approximate the Neyman–Pearson Detector Using Supervised Learning Machines

A study of the possibility of approximating the Neyman-Pearson detector using supervised learning machines is presented. Two error functions are considered for training: the sum-of-squares error and the Minkowski error with R = 1. The study is based on the calculation of the function the learning machine approximates to during training, and the application of a sufficient condition previously formulated. Some experiments about signal detection using neural networks are also presented to test the validity of the study. Theoretical and experimental results demonstrate, on one hand, that only the sum-of-squares error is suitable to approximate the Neyman-Pearson detector and, on the other hand, that the Minkowski error with R = 1 is suitable to approximate the minimum probability of error classifier.

Sensorized Garments and Textrode-Enabled Measurement Instrumentation for Ambulatory Assessment of the Autonomic Nervous System Response in the ATREC Project

Advances in textile materials, technology and miniaturization of electronics for measurement instrumentation has boosted the development of wearable measurement systems. In several projects sensorized garments and non-invasive instrumentation have been integrated to assess on emotional, cognitive responses as well as physical arousal and status of mental stress through the study of the autonomous nervous system. Assessing the mental state of workers under stressful conditions is critical to identify which workers are in the proper state of mind and which are not ready to undertake a mission, which might consequently risk their own life and the lives of others. The project Assessment in Real Time of the Stress in Combatants (ATREC) aims to enable real time assessment of mental stress of the Spanish Armed Forces during military activities using a wearable measurement system containing sensorized garments and textile-enabled non-invasive instrumentation. This work describes the multiparametric sensorized garments and measurement instrumentation implemented in the first phase of the project required to evaluate physiological indicators and recording candidates that can be useful for detection of mental stress. For such purpose different sensorized garments have been constructed: a textrode chest-strap system with six repositionable textrodes, a sensorized glove and an upper-arm strap. The implemented textile-enabled instrumentation contains one skin galvanometer, two temperature sensors for skin and environmental temperature and an impedance pneumographer containing a 1-channel ECG amplifier to record cardiogenic biopotentials. With such combinations of garments and non-invasive measurement devices, a multiparametric wearable measurement system has been implemented able to record the following physiological parameters: heart and respiration rate, skin galvanic response, environmental and peripheral temperature. To ensure the proper functioning of the implemented garments and devices the full series of 12 sets have been functionally tested recording cardiogenic biopotential, thoracic impedance, galvanic skin response and temperature values. The experimental results indicate that the implemented wearable measurement systems operate according to the specifications and are ready to be used for mental stress experiments, which will be executed in the coming phases of the project with dozens of healthy volunteers.

Modified LMS-Based Feedback-Reduction Subsystems in Digital Hearing Aids Based on WOLA Filter Bank

Digital hearing aids usually suffer from acoustic feedback. This feedback corrupts the speech signal, causes instability, and damages the speech intelligibility. To solve these problems, an acoustic feedback reduction (AFR) subsystem using adaptive algorithms such as the least mean square (LMS) algorithm is needed. Although this algorithm has a reduced computational cost, it is very unstable. To avoid this situation, other AFR subsystems based on modifications of the LMS algorithm are used. Such algorithms are given as follows: 1) normalized LMS (NLMS); 2) filtered-X LMS (FXLMS); and 3) normalized FXLMS (NFXLMS). These algorithms are tested in three digital hearing aid categories: 1) in the ear (ITE); 2) in the canal (ITC); and behind the ear (BTE). The first and second categories under study suffer from great feedback effects due to the short distance between the loudspeaker and the microphone, whereas the third category suffers from these effects due to the high signal level at the hearing aid output; thus, robust AFR subsystems are needed. The added stable gains (ASGs) over the limit gain when AFR subsystems are working in the digital hearing aids are studied for all the categories. The ASG is determined as a tradeoff between two measurements: 1) segmented signal-to-noise ratio (objective measurement) and 2) speech quality (subjective measurement). The results show how the digital hearing aids that work with AFR subsystems adapted with the NLMS or the NFXLMS algorithms can achieve up to 18 dB of increase over the limit gain. After analyzing the results, it is observed that the subjective measurement always limits the achieved ASG, but when the NLMS algorithm is used, it is appreciated that the objective measurement is a good approximation for estimating the maximum achieved ASG. Finally, taking into consideration the hearing aid performances and the computational cost of each AFR subsystem implementation, an AFR subsystem based on the NLMS algorithm to adapt feedback-reduction filters that are 128 coefficients long is proposed.

A hearing aid simulator to test adaptive signal processing algorithms

This paper deals with the description of a hearing aid simulation tool. This tool simulates the real behavior of digital DSP-based hearing aids with the aim of getting a very promising performance, which can be used for further design and research, and for a better fitting of the hearing impaired patient. The main parameters to program are the noise reduction techniques and the compression and feedback reduction algorithms. Also any other configuration is possible due to the access to the simulated signals in the hearing aid. So we can get a very promising performance which can be used for further design and research and for a better fitting of the hearing impaired patient. Results using a multilevel multifrequency hearing aid with real data collected from 18 patients show how the multifrequency compression techniques adapt the normal perceptible sounds to the hearing impaired patient perceiving area.