Abstract
1778
Objectives Image quality of the whole body FDG-PET degrades as the body size increases, if acquired with the same injected activity per body weight and the same scanning period. Lack of phantoms with various sizes makes it difficult to determine the optimum scanning condition for large as well small sized subjects. We have devised a useful technique to create phantoms of variable size.
Methods A body phantom of 40cm width was created in a design similar to the standard 30cm wide NEMA IEC body phantom. Layers of polyolefin foam plates were placed around on the lining of the 40cm phantom to make the actual width of the phantom 40, 39, 38, 36, 35, 33, 32, 31, 30cm, and also 30, 28, 26cm similarly using the standard 30cm body phantom. The phantom of each actual size was filled with 18F solution of 5.3kBq/ml in the background area and with solution of 4 times the activity concentration within the 10mm sphere, and was scanned for 1,2,3, …10 min with Discovery ST-EP PET/CT. Noise equivalent count (NEC), NEC density (=NEC/volume), hot sphere contrast (Q10), back ground variability (N10) and random fraction were measured.
Results As the size increased with the same scanning period, N10 and random fraction increased, NEC density and Q10 decreased, and NEC remained roughly constant. NEC density was associated with N10 better than NEC, with root square of NEC density inversely proportional to N10. To obtain the same NEC density on phantoms of different size containing the same activity concentration, the scanning time had to be increased proportionally to the cross sectional area. This corresponded to increasing the scanning time roughly proportionally to the body weight, based on the relationship between body weight and abdominal cross sectional area in the Japanese population.
Conclusions This phantom technique proved to be an easy and useful way of evaluating image quality and obtaining optimum scanning condition for different body thickness