Electrode–Tissue Impedance Measurement CMOS ASIC for Functional Electrical Stimulation Neuroprostheses

The design of a complementary metal-oxide-semiconductor (CMOS) integrated circuit that has the capability of evaluating the electrode-tissue contact and the lead functionality of implanted electrodes in neuroprostheses for functional electrical stimulation applications is presented. The system allows in a quite simple way the verification of the state of the electrode by the evaluation of its impedance. By using a 1-bit analog-to-digital converter, the voltage drop between the anode and cathode of the stimulating electrode is measured when a controllable amplitude current is delivered. The impedance that is associated with the electrode will be obtained from the ratio between the anode-to-cathode voltage drop and the delivered current. Final resolution for the impedance measurement system will depend on the range of available current amplitudes. Based on the same impedance measurement method, two different systems for two different CMOS integrated neuroprostheses have been implemented. Both measurement systems allow an impedance characterization range from a few ohms to several kilo-ohms. This resistance range is enough to evaluate if the electrode is working inside normal conditions or not; however, some damage has taken place, making the electrical stimulation using that electrode impossible. In vitro results of the impedance measurement systems are provided.

I. Introduction

Several papers reported in the literature deal with the measurement and evaluation of impedance for biomedical applications [1]–[4]. In these papers, two- and four-electrode tissue impedance measurement systems are explained. There are a lot of applications in which the impedance of the tissue can give some insights into the state of the organ or the state of some illness (e.g., ischemia [5], hypoxia [6], and tumours [7]). In all these cases, a clear impedance measurement (information on both phase and magnitude versus frequency) is necessary. These impedance measurement systems are basically based on the injection of a sinusoidal current (with a variable frequency) through the electrodes and the measurement of the resulting voltage. The output signal is an amplitude-modulated voltage, and to obtain the resistance, signal processing is needed. Usually, an analog demodulator is implemented using a mixer, a low-pass filter, a sampler, and a final analog-to-digital converter (ADC). Other solutions imply the use of synchronous sampling to demodulate bioelectric impedance signals [3]. Opposite to these frequency-domain techniques, time-domain measurement techniques have also been reported as a new method to measure bioimpedance. The use of single stimulation square wave and measurement of the resulting current intensity at three given times [8] or a direct analysis of the electrode voltage drop are some examples [9]. The main drawback of both solutions is the need for fast ADCs and powerful data processors. Some of the measurement systems above [9] have been also applied to implantable medical devices.