The dimensions of the phantoms were based on measurements made on healthy adult rats and mice in our laboratory. For both phantoms, an elongated hollow cylinder with a round end was manufactured. The mould consisted of an outer cylinder (a test tube) within which was centrally placed a Perspex rod. For the rat phantom, the outer diameter and wall thickness were nominally 10 mm and 2 mm, respectively, and for the mouse phantom, they were 5 mm and 1.5 mm, respectively. The central rod was raised above the bottom of the outer tube by an amount equal to the required wall thickness. A 15% MAPK inhibitor concentration of PVA (PVA Gels, Kingston, NY, USA) in water
was used. The PVA gel was heated to 80°C–95°C in a water bath, drawn into a 10-ml syringe and then injected into the mould to a depth of 2 cm for the rat phantom and 8 mm for the mouse phantom. The gel was allowed to settle overnight to allow any air bubbles to dissipate. The moulds were
subject to two, four or six freeze–thaw cycles. Each cycle consisted of cooling at 0.5°C per minute to − 20°C, maintaining the temperature for 8 h and then allowing a rise to room temperature (22°C) at a rate selleck compound of 0.5°C/min. The mould was maintained at room temperature for at least 8 h prior to separation of the PVA from the mould. The finished phantoms were stored in deionized water to prevent dehydration. Relaxation time constants T1 and T2 have been reported for PVA at field strengths between 1 T and 3 T ,  and , but there are no reported values taken at higher magnetic field strengths. Phantoms
were moulded from Ibrutinib mouse PVA; subjected to two, four and six freeze–thaw cycles; and then imaged in a 7-T MRI scanner [Agilent Technologies (formerly Varian, Inc.), Santa Clara, CA, USA]. Values of T1 were measured in the “short-axis” view using a fast spin echo sequence with inversion preparation and inversion times TI ranging from 10 ms to 3000 ms. The resulting image intensities were fitted to an exponential recovery curve using software on the scanner. Values of T2 were measured using fast spin echo sequences with echo times TE ranging from 10 ms to 60 ms, and the image intensities were fitted to an exponential decay curve using scanner software. The cardiac phantoms were mounted as shown in Fig. 1 within a sealed unit that could be filled with water and including an overflow as a precautionary measure in case of leakage during MRI scanning. The phantom was connected via stiff ¼-in. PTFE tubing (Cole-Parmer, Vernon Hills, IL, USA) to a gear pump (Michael Smith Engineering, Woking, UK). The phantom, tubing and gear pump were primed with water. The pump flow rate was controlled using a waveform generator. An offset sinusoidal waveform was applied in order to generate sinusoidal flow and hence cyclic distension of the phantom. Pumping frequencies up to 5 Hz [i.e., 300 beats per min (bpm)] were used for the rat phantom and up to 8 Hz (480 bpm) for the mouse phantom.