We have developed an impedance imaging system to reconstruct cross-sectional images of the body's electrical characteristics based on static tissue impedance. The hardware system consists of a data collection subsystem and the Intel 380 host microcomputer system with an Intel 80286 microprocessor, an Intel 80287 numeric data processor, and an Intel 80186 microprocessor-based display board. The system is capable of initiating a data collection from an array of current-sensing electrodes and reconstructing impedance images based on these data measurements. We have tested the data collection subsystem with physical phantom models, and we have found that the prototype system is capable of discriminating high resistivity regions in contrast with the low resistivity background. Our system is flexible in that each electrode's function (sensing currents, applying voltages, grounding body surfaces, and disconnected from the body) can be programmed individually so that a variety of electrode configurations for different projection techniques can be tested for optimal system performance. Various reconstruction algorithms have been developed and tested particularly for this imaging modality. Since a computer body model is needed for some impedance reconstruction algorithms, we have created two- and three-dimensional computer body models based on the finite element method approach, and verified our finite element modelling technique by building physical phantoms and comparing measured experimental results with simulation results predicted by the computer model. We have found that the sensitivity is a function of position, pixel size (image resolution) and background resistivity. We have also tried to compensate the low sensitivity of impedance changes in the central region.(ABSTRACT TRUNCATED AT 250 WORDS)
Read full abstract