Phys. Med. Biol. 43 (1998) 695–702. Printed in the UK PII: S0031-9155(98)83025-X Experimental procedure for the manufacture and calibration of polyacrylamide gel (PAG) for magnetic resonance imaging (MRI) radiation dosimetry C Baldock† ⇤ , R P Burford‡, N Billingham§, G S Wagnerk, S Patval¶, R D Badawi + and S F Keevil ⇤ † Centre for Medical and Health Physics, Queensland University of Technology, Brisbane, Australia ‡ Department of Polymer Science, University of New South Wales, Sydney, Australia § School of Chemistry, Physics and Environmental Science, University of Sussex, Brighton, UK k Department of Medical Physics, Royal Sussex County Hospital, Brighton, UK ¶ Joint Department of Physics, Royal Marsden Hospital, London, UK + Clinical PET Centre, St Thomas’ Hospital, London, UK ⇤ Division of Radiological Sciences, United Medical and Dental Schools, Guy’s Hospital, London, UK Received 2 April 1997, in final form 9 September 1997 Abstract. A simple methodology for the manufacture and calibration of polyacrylamide gel (PAG) for magnetic resonance imaging (MRI) radiation dosimetry is presented to enable individuals to undertake such work in a routine clinical environment. Samples of PAG were irradiated using a linear accelerator and imaged using a 0.5 T (22 MHz) Philips Gyroscan MRI scanner. The mean spin–lattice relaxation rate was measured using a ‘turbo-mixed’ sequence, consisting of a series of 90 ◦ pulses, each followed by acquisition of a train of spin echoes. The mean sensitivity for five different batches of PAG in the range up to 10 Gy was calculated to be 0.0285 s -1 Gy -1 for the mean spin–lattice relaxation rate with a percentage standard deviation of 1.25%. The overall reproducibility between batches was calculated to be 2.69%. This methodology, which introduces the novel use of pre-filled nitrogen vials for calibration, has been used to develop techniques for filling anatomically shaped anthropomorphic phantoms. 1. Introduction In 1984, it was proposed that nuclear magnetic resonance (NMR) relaxation measurements of particular irradiated materials may be used to determine absorbed dose of ionizing radiation (Gore et al 1984). The first dosimeter to be investigated was the well known Fricke dosimeter solution consisting of ferrous (Fe 2+ ) ions. When irradiated, ferrous ions in the dosimeter are converted to ferric (Fe 3+ ) ions. The magnetic resonance spin–lattice and spin–spin relaxation rates (1/T 1 and 1/T 2 respectively) of the dosimeter are related to the concentrations of Fe 2+ and Fe 3+ produced and the corresponding absorbed dose. Magnetic resonance imaging (MRI) may be used to image the increased paramagnetic enhancement of the water proton relaxation rates in the ferrous/ferric ion system. Following the initial studies, a number of authors reported work on MRI radiation dosimetry where ferrous sulphate solutions were incorporated into a gel matrix in order to ⇤ Addressee for correspondence: Clive Baldock, Centre for Medical and Health Physics, Queensland University of Technology, GPO Box 2434, Brisbane, Q4001, Australia. 0031-9155/98/030695+08$19.50 c  1998 IOP Publishing Ltd 695