TY - JOUR T1 - Ultrafast Brain Perfusion SPECT Using a CZT Semiconductor Gamma Camera JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1656 LP - 1656 VL - 59 IS - supplement 1 AU - Iichiro Osawa AU - Takashi Ushimi AU - Nanami Okano AU - Kaiji Inoue AU - Eito Kozawa AU - Mamoru Niitsu AU - Ichiro Matsunari Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/1656.abstract N2 - 1656Objectives: Prior myocardial perfusion SPECT studies have shown that ultrafast imaging is feasible using CZT-based SPECT owing to improved image quality compared with conventional SPECT. We evaluated the feasibility of reducing acquisition time for brain perfusion SPECT imaging using a novel CZT-based whole-body SPECT system (Discovery NM/CT 670 CZT). Methods: We enrolled a total of 30 consecutive patients (52-88 y, mean 71 y, Male 17, Female 13) who underwent 99mTc-ECD brain perfusion SPECT for evaluation of dementia. SPECT data were acquired in a list mode in 6 degree increments for 36 s/projection, resulting in a total acquisition time of 18 minutes with the 2-head camera. The acquisition data were reconstructed using 6.25%, 12.5%, 25%, 50%, and 100% of each projection data, which were equivalent to 1.1, 2.3, 4.5, 9.0, and 18 minutes acquisition, respectively. These images were automatically standardized using Q.Brain, with pixel values normalized to the individual global brain. Subsequently, we compared regional uptake ratios between 100% and each time-reduced image, and calculated Pearson correlation coefficients. Furthermore, we compared the correlation coefficients of the 100% vs each time-reduced image by Meng-Rhosenthal-Rubin Methods: Bonferroni correction was applied to minimize type I errors with multiple comparisons. Results: Injected dose of 99mTc-ECD was 824.9 ± 57.6 MBq (average ± standard deviation). Correlation coefficients of 100% vs 50%, 100% vs 25%, 100% vs 12.5%, and 100% vs 6.25% were 0.973, 0.959, 0.930, and 0.765, respectively. There was significant difference between two correlation coefficients for each comparison. Conclusion: The 50% image was most strongly correlated with the 100% image, while the 25% as well as the 12.5% images also showed very strong correlations with the 100% image. Thus, reduction in acquisition time for brain perfusion SPECT seems to be feasible using the CZT-based whole-body SPECT camera. ER -