A further assessment of an electromagnetic method to measure body composition.

Two experimental investigations of the electromagnetic method to measure body composition are presented. In the first experiment, the impedance of a single-loop transmitting antenna, located at the centre of a 150 m helical receiving coil, was observed to have a sharp minimum at 2.40 MHz. The impedance of the antenna at this frequency with respect to a series resistor was observed to go through a maximum when a small volume of 1% saline, of cross-section 0.5 cm2, was introduced coaxially into the coil. The frequency shift for larger volumes of saline (> 0.8 l) was highly correlated with the surface area of the liquid (r = 0.998, 0.997) and less well correlated with volume (r = 0.966, 0.967). In the second experiment, a five-turn solenoid, 2 m long and 1.5 m in diameter, was used to induce voltages in two coaxial receiving antennae placed 1 m apart inside the solenoid. The differential induced voltage was observed to vary as coaxial saline phantoms were traversed through the solenoid, with the variation being well correlated with the surface area of the liquid (r = 0.993, 0.996) and less well correlated with the volume of liquid in the phantom (r = 0.987, 0.958) for volumes above 100 ml. These observations suggest that the (resonant or non-resonant) electromagnetic field only responds to surface eddy currents in a coaxial saline phantom and therefore the TOBEC method may not be valid.