The influence of measurement frequencies on the accuracy of tissue impedance parametric modeling

The study was performed as a part of research to develop an electrical impedance spectroscopic system for muscle tissue ischemia estimation. Results of simulations done to determine the measurement frequencies that provide optimal extraction of tissue electrical impedance model parameters from electrical impedance spectra are presented. Measurement frequency sets that are logarithmic, quasilinear, and frequency sets that produce angularly equidistant points on the Nyquist loci, were tested with a tissue impedance model extraction algorithm which calculates R/sub 0/, R/sub /spl infin//, /spl alpha/, and /spl tau/ tissue parameters from complex impedance spectra. The results show that the measurement frequency set which best estimates: R/sub 0/ should include the decade from 100 Hz-1 kHz; R/sub /spl infin// should not include frequencies under 1 kHz; /spl alpha/ and /spl tau/ should be equidistantly spaced on the Nyquist locus. Using two sets of measurement frequencies, one to estimate R/sub 0/, and the other for R/sub /spl infin//, /spl alpha/, and /spl tau/, provides the best overall performance.