Automatic Analytical Modeling of EIS Data by Evolutive Programming Based on Cultural Algorithms

Abstract

Efficiency and accuracy problems in state-of-the-art analytical modeling of electrochemical phenomena through impedance spectroscopy are faced by a Cultural Hybrid Evolutionary Modeling Algorithm (CHEMA). Automatic model definition is improved by an evolutionary program exploiting a solution-search strategy based on a cultural mechanism: information on search advance is transmitted to all potential solutions, rather than only to a small inheriting subset, such as in traditional genetic approach. Experimental results of the proposed approach application to electrochemical impedance spectroscopy for biomedical purposes are presented.

A cultural evolutionary programming approach to automatic analytical modeling of electrochemical phenomena through impedance spectroscopy

“…In the following, the two blocks (i) GEPCAMI, and (ii) CuPBeS are described. model creation, and (ii) CAMI, cultural algorithm for model identification [30].…”

“…The second section of GEPCAMI, CAMI (figure 1), is a CA aimed at identifying each circuit model created by GEPMoC on a set of given experimental EIS data [30]. With this aim, for each circuit model, a population of corresponding candidate numerical values is created (circuit parameter space of figure 1).…”

In-vivo Test Procedure and Instrument Characterization for EIS-Based Diagnosis of Prosthesis Osseointegration

“…Information about electrical characteristics of the tissue are gathered by injecting low-level sinusoidal currents in a given frequency bandwidth by external electrodes, such as in Electrochemical Impedance Spectroscopy (EIS) [6]- [9]. In previous works, the authors proposed a low-invasive measurement method based on EIS, and showed its effectiveness for in-vitro [10]- [11] and in-vivo [12] experiments for the evaluation of osseointegration of metallic implants.…”

Editors’ Choice—AutoEIS: Automated Bayesian Model Selection and Analysis for Electrochemical Impedance Spectroscopy

“…Alternatively, the application of machine learning offers an approach to exploring the ECM solution space that can be more transparent and reproducible. [31][32][33][34][35][36][37][38][39][40] Evolutionary algorithms have been used since 2004 to explore ECM structure and have shown great promise in ECMs-based EIS analysis. 31,39,40 By combining genetic algorithm (GA) 41 and gene expression programming (GEP), 42 the evolutionary algorithms can automatically optimize both the circuit structure and fitting parameters of circuit components by sequentially searching through the solution space.…”