Fig. 5

Comparison of absorbed doses. Different dose calculation algorithms were used to compare average absorbed doses, namely (I) LD algorithm (red), (II) MIRD analytical approach (violet), (II) Kernel convolution using Philips Stratos (blue), (IV) full MC using Raydose (green stripe pattern), (V) MCNP4c (orange crosshatch pattern). The vertical bar chart displays the mean absorbed dose and its standard deviation. For methods I-II-II, the standard deviation was assessed from dose-volume-histograms obtained after calculation. For methods IV and V, the activity of ⁹⁰Y contained in the cylindrical insert was recovered from ⁹⁰Y–PET images using the dAVH approach described by Eq. 1. Then, the mean activity value, along with its standard deviation, has been used to assess doses with methods IV and V, thus providing a mean dose ± standard deviation. MC-based dose evaluations are reported with a different pattern to differentiate MC calculations with other algorithms that do not directly account for the presence of the acrylic wall. Average absorbed doses are reported along with their standard deviations. All dose algorithms provided comparable dose estimates with deviations below 9%, except for dose calculations performed on day 1, where the standard deviation over the average absorbed dose values is 20%. Insert: comparison between average absorbed dose per unit activity into the cylindrical insert. When the activity concentration decreases down to 0.22 MBq/mL (day 12) the standard deviation is approximately 100% the dose value