CompanyName: Rensselaer Polytechnic Institute
Distingishability of Inhomogeneities Using Planar Electrode Arrays and
Different Patterns of Applied Current and Voltage in EIT
Tzu-Jen Kao1, J.C. Newell1, and G. J. Saulnier2 and D. Isaacson3
Rensselaer Polytechnic Institute Troy, NY 12180
1Department of Biomedical Engineering, 2Department of Electrical, Computer and Systems Engineering, 3Department of Mathematical Sciences
Three methods are presently used for breast cancer diagnosis by Electrical Impedance Tomography (EIT). One applies a constant voltage to a flat array of electrodes on the body and measures the resulting currents at each of several electrodes. Another method applies multiple patterns of current and measures the resulting voltages. The third uses voltage sources to apply specified current levels, and then analyzes the voltages that were required.
The purpose of the present study was to compare the distinguishability obtainable by these three methods. We used various current patterns with small and large conducting and insulating targets placed at several different depths and positions relative to a flat electrode array. We also compared the single current sources as used by Korjenevsky et al.  and the voltage sources used by the T-SCAN system  with these current patterns. We made these comparisons using several sizes and numbers of electrodes.
A rectangular saline filled tank containing an array of stainless steel electrodes was used as the conductive medium. The Adaptive Current Tomograph III system  was used as the impedance imaging device. In some studies, a second array was introduced at the opposite end of the tank. We studied the distinguishability of different inhomogeneities by applying optimal currents  and measuring voltages. We then compared these results with those obtained by applying a uniform voltage at all electrodes and measuring the currents at each electrode. A third case consisted of applying currents to one electrode at a time, and measuring voltages at all electrodes. The results are interpreted in terms of the distinguishability defined by Isaacson . The optimal patterns allowed one to detect inhomogeneities almost twice as far from the electrodes as was possible with the other patterns.
1. A Korjenevsky, V. Cherepenin and V. Kornienko "Electric Mammograph with 3D visualization" 1st EPSRC Engineering Network meeting. London, 1999.
2. www.transscan.co.il TransScan T-Scan 2000 - P970033, United States Food and Drug Administration, April 16, 1999.
3. Edic, P.M., G.J. Saulnier, J.C. Newell and D. Isaacson. "A real-time electrical impedance tomograph " IEEE Trans. Biomed. Eng. 42(9): 849-859, 1995.
4 Isaacson, D. Distinguishability of Conductivities by Electric Current Computed Tomography. IEEE Trans Med Imaging MI-5 (2): 92-95, 1986.