Session Item

With the increasing use of magnetic resonance imaging (MRI) in radiotherapy (RT), there is a need for anthropomorphic multimodality phantoms to be used for various tasks including end-to-end tests of MRI-guided RT workflows. This project aims to find materials that mimic human tissue for MRI and computed tomography (CT) imaging and assess their stability over time and after radiation exposure.

Samples of the materials listed in Table 1 were produced according to the manufacturer's instructions and placed in a test tube. To measure stability over time, the CT number and T1 and T2 relaxation times of the samples have been measured monthly over 7 months. CT scans were acquired using Philips Brilliance BigBore (Philips Medical Systems, Best, Netherlands) with the following settings: 120 kVp, 104 mAs, 2 mm slice thickness, and in-plane resolution of 1.17 mm.

MRI relaxation times were measured using a 3T Siemens Prisma MRI scanner (Siemens Healthineers, Erlangen, Germany). Inversion recovery and Spin-Echo-Multi-Contrast sequences were used to create T1 and T2 maps, respectively. On CT images, T1 and T2 maps, volumes within the materials were drawn, and the mean and standard deviation were measured within these volumes.

To measure stability after radiation exposure, other samples of the same materials were exposed to radiation using an Elekta Versa HD linear accelerator, delivering 10, 100, 250, 500, and 1000 Gy. CT number and MRI relaxation times after radiation exposure were measured.

Only materials 1-4 have CT number, T1 and T2 relaxation times that fell within the range of the brain’s grey and white matter. Figure 1 shows the mean of CT number, T1 and T2 relaxation times, and the standard deviation for each material over time and after radiation exposure.

Regarding the stability of materials over time, all the observed change for all four materials was within two standard deviations (of the pixel values) which indicates the materials are stable over this timeframe. As for the stability after radiation exposure, all the observed changes were within two standard deviations (of the pixel values) except for the T2 relaxation time for material No. 1, and the change that occurred after exposure to 10 Gy in material No. 3. Further investigation will be done to investigate the reasons for these changes.

This project demonstrated the possibility of producing materials that mimic brain tissue in both MRI and CT. Four candidate materials were found; from these, all appear stable over time, and two are stable after radiation exposure.