Low-noise measurement for electrical impedance tomography

Hugh Mccann, Mandana Rafiei-Naeini, P. Wright, H. McCann

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    Abstract

    EIT has unique potential to provide fast imaging of impedance changes in the deep brain. Both EIT sensitivity and spatial resolution can be greatly improved by increased measurement SNR. In this paper we present a prototype EIT channel with better than 70 dB SNR at excitation frequencies from 10 kHz to 100 kHz. The system is divided into three major sub-systems: digital waveform generation and demodulation, current excitation and voltage measurement. The digital sub-system, presently implemented on a Xilinx Spartan 3-1000 FPGA, provides 16-bit excitation waveform data to a DAC and subsequent reconstruction filter, which in turn provides the input to a Howland current source. The SNR of the resulting excitation current is better than 70dB. Voltage measurement is accomplished using a buffered differential amplifier followed by an anti-aliasing filter and ADC. Phase-sensitive demodulation is performed within the FPGA and the resulting phase and amplitude information returned to a PC via a USB interface. © Springer-Verlag 2007.
    Original languageEnglish
    Title of host publicationIFMBE Proceedings|IFMBE Proc.
    Pages324-327
    Number of pages3
    Volume17
    Publication statusPublished - 2007
    Event13th International Conference on Electrical Bioimpedance and the 8th Conference on Electrical Impedance Tomography 2007, ICEBI 2007 - Graz
    Duration: 1 Jul 2007 → …

    Conference

    Conference13th International Conference on Electrical Bioimpedance and the 8th Conference on Electrical Impedance Tomography 2007, ICEBI 2007
    CityGraz
    Period1/07/07 → …

    Keywords

    • Current source
    • DDS
    • EIT
    • Phase-sensitive demodulation

    Fingerprint

    Dive into the research topics of 'Low-noise measurement for electrical impedance tomography'. Together they form a unique fingerprint.

    Cite this