TY - JOUR T1 - Possibility of Shortening Examination Time in Amyloid PET using <sup>18</sup>F-Flutemetamol JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 3082 LP - 3082 VL - 61 IS - supplement 1 AU - Tamaki Otani AU - Hideki Otsuka AU - Kou Matsushita AU - Yoichi Otomi AU - Yamato Kunikane AU - Shota Azane AU - Hirokazu Miyoshi Y1 - 2020/05/01 UR - http://jnm.snmjournals.org/content/61/supplement_1/3082.abstract N2 - 3082Introduction: Amyloid PET examination using 18F-flutemetamol is regularly performed over a 30-min acquisition time at Tokushima University Hospital. However, we have experienced many cases of obvious body movements of the patients within 20 min of starting the acquisition. Thus, we optimized the acquisition time and image reconstruction conditions, and examined whether the acquisition time could be shortened without degrading image quality. Methods: A Hoffman brain phantom was filled with 20 MBq of 18F, PET was collected for 30 min, and images were reconstructed into 10-, 20-, and 30-min time periods. The OS-EM algorithm was used for image reconstruction, which used a 2-mm or 4-mm Gaussian filter, and a combination of iteration and subset number. The percentage contrast and the coefficient of variation as a uniformity were used as physical evaluation indices for the reconstructed image. The reference value was &gt;55% in the percentage contrast and &lt;15% in the coefficient of variation. Five patterns with sufficient convergence of the contrast and excellent uniformity were selected. Seven radiographers performed visual evaluation using the five selected patterns, and ranked them in terms of contrast, uniformity, and overall evaluation. The imaging data of 15 clinical patients (positive: n=7; negative: n=8) were reconstructed under the five pattern conditions selected in the phantom stud. We then calculated the standardized uptake value ratio by removing the pons accumulation from the accumulation of each brain region, and compared between each condition. Results: By basic examination using the Hoffman brain phantom, five patterns were selected: (a) 30 min-4 mm-4/18, (b) 30 min-2 mm-4/18, (c) 20 min-4 mm-4/24, (d) 20 min-2 mm-2/32, and (e) 10 min-4 mm-4/12 (acquisition time-filter thickness-iteration/subset). The contrast did not change with the acquisition time, and was not ranked in visual evaluation. For uniformity, the longer the acquisition time the thicker the filter and the better the uniformity. When contrast and uniformity were comprehensively evaluated, the order was (c), (a), (b), (d), and (e). Thus, an acquisition &gt;20 min is desirable, with a recommended filter thickness of 4 mm. For the clinical images, there was no significant change in the acquisition time or the reconstruction conditions for the standardized uptake value ratio regardless of whether it was positive or negative. Therefore, shortening of the acquisition time did not affect the diagnostic results. Conclusions: Our data suggest that the acquisition time can be shortened to 20 min without degrading the conventional image quality, which may reduce patient burden. ER -