Abstract
Bioelectrical impedance in vivo measurements suffer from many potential sources of error due to the patient-instrument interface. The total common-mode rejection ratio (CMRRT) was investigated experimentally for three measurement channel circuit versions, including electrode-skin impedance imbalance. The first version was of the `classical' type. The second one makes use of a differential filter at the input of the instrumentation amplifier. The third circuit was a frequency-converting structure, where the signal was demodulated before being amplified. The differential demodulator was based on synchronous sampling using floating capacitors. The experiments were accomplished with simulated imbalance of the real and imaginary parts of electrode-skin impedances. To reduce unwanted common-mode voltage, a differential accurately balanced current source was used. Considering an application in impedance cardiography, the experiments were carried out at a single frequency of 40 kHz. The results showed the advantage of the circuits using frequency conversion and differential input filter, rendering at least 15 dB higher CMRRT. The most significant reduction of CMRRT resulted from imbalance of the capacitance component of voltage-sensing electrode impedances. The third circuit showed an unexpected behaviour of CMRRT improvement with higher imbalance of the electrode-skin impedance resistance component.