SIMIND Monte Carlo validation for Lutetium-177 imaging

Objectives: Due to the development of new peptides Lu-177 has gained popularity as one of the isotopes of choice for peptide receptor radionuclide therapy using the theranostics approach. The choice of; collimator, gamma ray energy to image and energy window settings for Lu-177 is well investigated using Monte Carlo (MC) simulations. MC offers the advantage of evaluating phenomena involved in nuclear medicine imaging which may be cumbersome to measure physically. Gamma cameras are subject to a variety of detector and electronic challenges therefore it is deemed useful to validate the capability of MC programs to model physical gamma cameras. The MC program SIMIND has been validated for modelling various gamma cameras for a number of different radionuclides. Validation of SIMIND for a Siemens Symbia T16 dual head SPECT/CT gamma camera with Lu-177 has not been reported. The aim of this study was to validate the MC simulation setup of the Siemens Symbia T16 dual head SPECT/CT gamma camera for Lu-177 using SIMIND.

Methods: The validation was done by comparing experimental and simulated gamma camera performance criteria tests for the Lu-177208 keV photopeak with a medium-energy low-penetration collimator. These tests included four planar NEMA stipulated tests namely; intrinsic and extrinsic energy resolution, extrinsic spatial resolution and extrinsic sensitivity. In addition, two SPECT tests termed air and water calibration factor (CF) were validated. The SPECT tests used an activity-filled sphere suspended in air and in awater-filled cylindrical phantom to generate the air and water CF respectively. Sequential SPECT and CT images were acquired following a clinical imaging protocol. These images were reconstructed using FLASH 3D with CT-based attenuation correction, TEW scatter correction and full 3D collimator-detector response. Experimental measurements were repeated three times, while simulations were only performed once for each test. The data was processed and the results compared.

Results: The intrinsic experimental and simulated energy resolution values were 9.6 ± 0.06 % and 9.9 % respectively. The % and absolute difference of -3.1 % and 0.3 % between these values indicated good agreement. The corresponding extrinsic energy resolution values were 9.4 ± 0.09 % and 9.9 % for the experimental and simulated data, with a % and absolute difference of -5.3 % and 0.5 %. The calculated FWHM and FWTM for the extrinsic spatial resolution were 11.5 ± 0.35 mm and 20.8 ± 0.82 mm for the experimental and 11.0 mm and 20.1 mm for the simulated data. The system sensitivity values of 10.0 ± 0.03 cps/MBq vs. 10.3 cps/MBq showed good agreement between the experimental and simulated values. The SPECT experimental and simulated air CF values were 8.4 ± 1.04 cps/MBq and 8.3 cps/MBq with % and absolute differences of 1.2 % and 0.1 cps/MBq. The water CF values of 7.8 ± 0.94 cps/MBq vs. 7.2 cps/MBq showed slightly higher % and absolute differences of 7.7 % and 1.6 cps/MBq between the experimental and simulated studies. These larger differences may be attributed to the introduction of attenuation and scatter by the water-filled phantom in both the experimental and simulated setup. Conclusion: The results obtained in this study were found to be in good agreement for both the planar and SPECT tests. From this study, it is evident that SIMIND can emulate the Symbia T16 successfully and therefore can be used with confidence in future studies to model planar and SPECT Lu-177 images.