Experimental and theoretical study of the Swansea in vivo neutron activation analysis clinical facility for measurement of total body nitrogen

• Main Author: Al-Agel, F. A. M.
• Published: Swansea University 1997
• Subjects: 612.015
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.635869

This is work concerned with modification, assessment, optimisation, and calibration of a prompt-capture neutron activation analysis facility based on a <SUP>252</SUP>Cf source for the measurement of total body nitrogen <I>in vivo</I> via the <SUP>14</SUP>N(n,γ)<SUP>15</SUP>N reaction. To achieve these goals, the Swansea facility was studied by both theoretical simulation and experimental measurement. The geometry of the facility was entered into a radiation transport modelling MCNP code based on a Monte Carlo process. A number of configurations were tested to pursue fulfilment of the above goals. Regarding the neutron beam, the results of calculations performed with MCNP code together with experimental measurements of thermal flux indicated that the facility can deliver a suitable thermal flux throughout a phantom when irradiated using a 5 cm bismuth collimator (the 'short' collimator) and 50 μg <SUP>252</SUP>Cf source. Use of a lower activity neutron source to that originally utilized with this facility (200 μg) permits a considerable cost saving. For the 'short' collimator, gamma-ray detector positions were investigated theoretically and were complemented by practical measurements; optimum positions at which the maximum values of signal-to-noise ratio in the nitrogen and hydrogen region of interest and the uniformity of irradiation and detection of nitrogen were determined. The MCNP simulations have shown that the contribution of the hydrogen counts (background) arising from the facility can be a considerable proportion (up to 19%) of the calibration phantom counts and can have a significant effect on the calculation of total body nitrogen. A correction for the hydrogen background contribution has been implemented. MCNP calculations were also used to assess the influence on calibration of overlying fat around a body phantom. The results indicate that errors of typically 35% for a 6 cm fat layer may arise if calibration is undertaken using a homogenous tissue equivalent solution phantom but where in practices the subject of the same external dimensions has a layer of fat.