Previsional dosimetry in 99mTc-MAA SPECT-TC images for liver radioembolization with 90Y microspheres: Influence of iterative and FBP reconstruction algorithms on dose calculation. (September 2016)

• Record Type: Journal Article
• Authors: Richetta, E.; Radici, L.; Pasquino, M.; Cutaia, C.; Valzano, S.; Menga, M.; Pellerito, R.E.; Stasi, M.
• Is Part Of: Physica medica. Volume 32(2016)Supplement 3
• Page Start: 199
• Page End: 200
• Publication Date: 2016-09
• Subjects: Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2016.07.677
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Introduction: 90 Y radioembolization is preceded by a 99m Tc-MAA simulation to calculate tumour (T) and normal liver (NL) doses with MIRD, supposing a uniform distribution, or with voxel-dosimetry taking into account for the inhomogeneities. Purpose: To evaluate differences between iterative and FBP algorithms. Materials and methods: 18 patients SPECT-TC 99m Tc-MAA were acquired (SiemensIntevo) and reconstructed with iterative (Flash3D, attenuation-scatter corrected) and with FBP (no corrections). A physician countored VOIs on both series: mean doses to T and NL were obtained with MIRD. All series were processed with an home-made voxel code: mean doses and DVHs were evaluated (NL and T: D2% D20% D95% D100% and V30 Gy V40 Gy V50 Gy and V120 Gy for T). Wilcoxon paired-test compared volumes and doses. Acquisition on a phantom 1.25 MBq/ml (spheres 1 ÷ 11 ml) and cylinder (130 ml) was also performed. Results: FBP T volumes were smaller than IT ones in 15/18 patient images (median −18%), no statistical difference was found for NL. T mean doses were higher for FBP both with MIRD (17/18 patients, +10%) and voxel method (15/18 patients, +5%); NL mean doses were higher for FBP with MIRD only (14/18 patients, +25%). DVH analysis showed no statistical difference in Dx and Vx but with an extended range of variability. In phantom opposite results were found: mean doses and DVHs were lower in FBP than IT (spheres −22%, cylinder −53%). Conclusion: FBP and Iterative algorithms can affect volumes and mean-dose calculations probably due to not appropriate homogeneity hypothesis. At voxel level lower discrepancies were found on patients but not in phantom. Disclosure: The authors disclose any relationship that may bias their presentation.