Estimation of body fluids with bioimpedance spectroscopy: state of the art methods and proposal of novel methods.

Determination of body fluids is a useful common practice in determination of disease mechanisms and treatments. Bioimpedance spectroscopy (BIS) methods are non-invasive, inexpensive and rapid alternatives to reference methods such as tracer dilution. However, they are indirect and their robustness and validity are unclear. In this article, state of the art methods are reviewed, their drawbacks identified and new methods are proposed. All methods were tested on a clinical database of patients receiving growth hormone replacement therapy. Results indicated that most BIS methods are similarly accurate (e.g. < 0.5 ± 3.0% mean percentage difference for total body water) for estimation of body fluids. A new model for calculation is proposed that performs equally well for all fluid compartments (total body water, extra- and intracellular water). It is suggested that the main source of error in extracellular water estimation is due to anisotropy, in total body water estimation to the uncertainty associated with intracellular resistivity and in determination of intracellular water a combination of both.

Robustness study of the different immittance spectra and frequency ranges in bioimpedance spectroscopy analysis for assessment of total body composition.

The estimation of body fluids is a useful and common practice for assessment of disease status and therapy outcomes. Electrical bioimpedance spectroscopy (EBIS) methods are noninvasive, inexpensive and efficient alternatives for determination of body fluids. One of the main source of errors in EBIS measurements in the estimation of body fluids is capacitive coupling. In this paper an analysis of capacitive coupling in EBIS measurements was performed and the robustness of the different immittance spectra against it tested. On simulations the conductance (G) spectrum presented the smallest overall error, among all immittance spectra, in the estimation of the impedance parameters used to estimate body fluids. Afterwards the frequency range of 10-500 kHz showed to be the most robust band of the G spectrum. The accuracy of body fluid estimations from the resulting parameters that utilized G spectrum and parameters provided by the measuring device were tested on EBIS clinical measurements from growth hormone replacement therapy patients against estimations performed with dilution methods. Regarding extracellular fluid, the correlation between each EBIS method and dilution was 0.93 with limits of agreement of 1.06 ± 2.95 l for the device, 1.10 ± 2.94 l for G [10-500 kHz] and 1.04 ± 2.94 l for G [5-1000 kHz]. Regarding intracellular fluid, the correlation between dilution and the device was 0.91, same as for G [10-500 kHz] and 0.92 for G [5-1000 kHz]. Limits of agreement were 0.12 ± 4.46 l for the device, 0.09 ± 4.45 for G [10-500 kHz] and 0.04 ± 4.58 for G [5-1000 kHz]. Such close results between the EBIS methods validate the proposed approach of using G spectrum for initial Cole characterization and posterior clinical estimation of body fluids status.

Ultrasonic flaw detection using radial basis function networks (RBFNs).

Ultrasonic flaw detection has been studied many times in the literature. Schemes based on thresholding after a previous matched filter use to be the best solution, but results obtained with this method are only satisfactory when scattering and attenuation are not considered. In this paper, we propose an alternative solution to thresholding detection method. We deal with the usage of different flaw detection methods comparing them with the proposed one. The experiment tries to determinate whether a given ultrasonic signal contains a flaw echo or not. Starting with a set of 24,000 patterns with 750 samples each one, two subsets are defined for the experiments. The first one, the training set, is used to obtain the detection parameters of the different methods, and the second one is used to test the performance of them. The proposed method is based on radial basis functions networks, one of the most powerful neural network techniques. This signal processing technique tries to find the optimal decision criterion. Comparing this method with thresholding based ones, an improvement over 25-30% is obtained, depending on the probability of false alarm. So our new method is a good alternative to flaw detection problem.