[OA104] Creation of specific normal databases for perfusion quantification of low-dose myocardial spect studies. (August 2018)

• Record Type: Journal Article
• Authors: Lecchi, Michela; Scabbio, Camilla; Capozza, Antonella; Malaspina, Simona; Selvaggi, Claudio; Lucignani, Giovanni; Sole, Angelo Del
• Is Part Of: Physica medica. Volume 52(2018)Supplement 1
• Page Start: 40
• Page End: 41
• Publication Date: 2018-08
• 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.2018.06.176
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Purpose: We have recently demonstrated that iterative reconstruction algorithms with resolution recovery (IRR algorithms) requires the adoption of the own specific normal databases (NDBs). The present work was aimed to generate three home-made NDBs (HM-NDSs) with different count-statistics (from 100% to 25% of the traditional reference) in order to reduce the patient doses to very low values ( ∼ 3 mSv) and at the same time, to perform a correct myocardial perfusion quantification. Methods: Low-dose imaging was simulated using different acquisition times (different count statistics). Three simultaneous scans were acquired with three different acquisition-time/projections (30, 15 and 8 s) and reconstructed using the IRR algorithm Astonish TM with default parameters ( γ -camera BrightView, Philips). Fifty-two patients with low likelihood of coronary artery disease (23 male and 29 female) were enrolled. Three HM-NDBs with different count-statistics were then generated: 100%-HM-NDBs, 50%-HM-NDBs and 25%-HM-NDBs. Using these new NDBs, percent summed rest (SR%) and stress (SS%) scores were calculated for another group of 25 consecutive patients (5 normal and 20 abnormal). The SR% and SS% values found for 100%-HM-NDBs were compared with those obtained with the NDBs available on the workstation, 100%-WS-NDBs (Bland–Altman plots). Moreover, the impact of the study count-statistics on percent scores was evaluated using the own specific NDBs (ANOVA analysis and Tukey test as post hoc, p < 0.05). Results: Significantly higher SS% values were found for the 100%-HM-NDBs respect to the 100%-WS-NDBs (mean diff.: 1.72% with a 95% confidence interval of 0.45–2.99%), while no significant difference was found for SR% (mean diff.: 0.72% with a 95%-CI of −0.08–1.52%). Post-hoc test showed significantly higher SS% values (p = 0.0012) for the 100%-HM-NDBs respect to the 25%-HM-NDBs (6.64 ± 1.33 vs. 5.16 ± 1.03). Conclusions: The present work confirm that SS% depend significantly on both NDBs and study count-statistics. A 50% reduction in patient dose ( ∼ 6 mSv) is finally the limit for Astonish TM (with the default parameters) in order to prevent significant variation in perfusion quantification of stress scans.