Reducing the uncertainties of lung counting when employing an array of four HPGe detectors

The basic assumption made when calibrating a Lung Counter is that the deposition of the radioactive aerosols in lungs is homogenous, and therefore the calibration is made with realistic phantoms containing lung sets with homogeneous contamination. Unfortunately, the deposition of the radioactive aerosols is rarely homogenous, being a function of the aerosol size and breathing rate(1) . In some cases, the contamination can even be present in the form of single ''hot' particles. It is expected that the count rate will be strongly dependent on the position of the ''hot particle's' because of the geometrical factor (which is determined by the distance) and the absorption factor (which is dependent on the path in the lung and in the chest). In a work dealing with the subject(2), the lung was divided into 16 volume cells, and the uncertainty due to inhomogeneous distribution was estimated by Monte Carlo simulation. The deviations from homogeneity were evaluated to be by factors of 26, 3 and 1.8 for photon energies of 17, 60 and 1000 keV respectively. In the present work, an estimation of the possible deviations is presented for the system employed at NRCN, and a procedure is proposed to diminish the errors of the activity determined due to the non-homogeneous source distribution in the lungs