Optimization of Block Sequential Regularized Expectation Maximization Reconstruction Parameters for Striatal SPECT Using a Ring-Type Semiconductor SPECT/CT System

Le résumé fourni par la source

This phantom study aimed to optimize the noise-suppression parameter (β) and the edge-preservation parameter (γ) values in the block sequential regularized expectation maximization (BSREM) reconstruction method for 123I-2-β-carbomethoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl)nortropane striatal SPECT using a ring-type cadmium–zinc–telluride camera. Methods: We used a striatum phantom filled with an 123I solution, with adjusted concentrations in the left and right caudate and left and right putamen. The phantom was scanned for 120 min using the StarGuide system, and 20-min SPECT data were extracted from the total acquisition. Using the BSREM method, we generated 42 reconstructed images by varying β and γ. We calculated the percent coefficient of variation and percent contrast for each image. Visual evaluations were conducted by a nuclear medicine physician and 2 technologists. Additionally, we calculated the striatal uptake ratio (SUR) and the ratio of the left caudate SUR to putamen SUR in the left striatum (left striatal ratio). Results: The percent coefficient of variation decreased as the β value increased, whereas percent contrast increased as the γ value increased. We conducted a visual evaluation of 15 reconstructed images selected by physical assessment. Images reconstructed with a β value of 0.03 with a γ value of 5 and those with a β value of 0.05 with a γ value of 10 were deemed suitable for diagnostic purposes because of their accurate depiction of tracer accumulation in each striatal region. SUR and left striatal ratio were comparable for these 2 image sets. Conclusion: Evaluation of the optimal conditions for striatal SPECT imaging with BSREM reconstruction revealed that the optimal β value and γ value combinations were a β value of 0.03 and γ value of 5 and a β value of 0.05 and γ value of 10.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.