Extracting Parasite Effects of Electrical Bioimpedance Measurements

Abstract

The objective of this work is to develop a technique for filtering parasitic effects from the impedance spectra (IS) measured in biological material phantoms. IS data are contaminated with unexpected capacitive and inductive effects from cable, input/output amplifiers capacitances, electrode polarization, temperature and contact pressure when collecting data. It is proposed a model which contains an RLC-network in series with the Cole model (RSC), then called RLC-Cole. It was built four circuits composed by resistors, capacitors and inductors. An impedance analyzer (HF2IS) was used to perform the measurements in the frequency range of 1 to 3000 kHz. Data were fitted into the model and comparisons to the nominal values were made. In order to validate the proposed model, a gelatin phantom and a chicken breast muscle impedance spectra were also collected and analyzed. After filtering, Cole fitting was performed. Results showed a maximum root-mean-square error of 1% for the circuits, 2.63% for the gelatin phantom, whereas 2.01% for the chicken breast. The RLC-Cole model could significantly remove parasitic effects out of a tissue impedance spectrum measured by a 4-point electrode probe. This may be highly important in EIS systems whose objective is to discriminate a normal tissue from a cancerous one.

Keywords: Bioimpedance; Cole model; filtering algorithm; parasitic impedance.

Fig.1

Schematic diagram of the model and measuring system. (a) Measuring system. (b) Parasitic impedances. (c) First proposed RLC-Cole-RLC model. (d) Final proposed RLC-Cole model. (e) Phantom measurements.

Fig.2

a) Mean impedance spectra of one of the electrical phantom; (b) gelatin phantom; (c) chicken breast muscle; (d) nominal and fitted values of the electrical phantom 3 (see table 1); e) fitted values for gelatin phantom; f) fitted values for chicken breast muscle.