Electrical properties of tissues in the human body can be imaged using
a technology known as Electrical Impedance Tomography, In this modali
ty, sinusoidal electrical currents are applied to the body using elect
rodes attached to the skin, and voltages that are developed on the ele
ctrodes are measured, Using these data, a reconstruction algorithm com
putes the conductivity and permittivity distributions within the body,
This paper describes the reconstruction algorithm, image display algo
rithm, and hardware of a real-time Electrical Impedance Tomograph know
n as the Real-Time Imaging System, The reconstruction algorithm, execu
ted by a commercially available coprocessor board that resides in a 38
6-based personal computer, is a modification of the Newton's One Step
Error Reconstructor (NOSER) that minimizes algorithm execution time by
precomputing many quantities. The image display algorithm, also execu
ted by the coprocessor board, maps the output of the reconstruction al
gorithm into an image which is displayed using a video graphics board,
The architecture of the system and execution times of algorithms impl
emented by the system are discussed, Using the continuous data acquisi
tion mode of the Real-Time Imaging System, data from the thorax of a n
ormal human subject were collected, Admittivity changes in the chest,
as a result of respiration and the cardiac cycle, are presented, Data
that were collected from the leg of a normal subject are shown which d
emonstrate capabilities of the triggered data acquisition mode of the
system, allowing data acquisition synchronization with an electrocardi
ogram.