Error evaluation of the D-shuttle dosimeter technique in positron emission tomography study

• Published in: Radiological physics and technology 2019-12, Vol.12 (4), p.363-373
• Main Authors: Islam, Md. Shahidul, Watanuki, Shoichi, Tashiro, Manabu, Watabe, Hiroshi
• Format: Article
• Language: English
• Subjects: Bias; Dosimeters; Dosimetry; Emission analysis; Imaging; Medical and Radiation Physics; Medicine; Medicine & Public Health; Motion; Nuclear Medicine; Performance evaluation; Phantoms, Imaging; Positron emission; Positron-Emission Tomography - instrumentation; Radiation Dosimeters; Radiology; Radiotherapy; Research Design; Tomography; Torso
• ISSN: 1865-0333; 1865-0341
• DOI: 10.1007/s12194-019-00530-w

The D-shuttle dosimeter technique is a convenient approach for estimating the radiation dosimetry in a positron emission tomography (PET) study that employs multiple D-shuttle dosimeters attached to the body surface of a patient. To bring this technique into clinical usage, it is very important to evaluate its performance by investigating the bias associated with D-shuttle dosimeter positioning and by comparing the estimates with those of the whole-body dynamic PET imaging technique. The torso cavity and six spheres of the NEMA body phantom were filled with 18 F-FDG solution, and then, the phantom was imaged for 1 h. We assumed the mislocated positioning of the D-shuttle dosimeters by shifting them in the z -direction (upper) in a range of 1–5 cm from the original positions. The cumulative radioactivities, absorbed doses, and effective dose were estimated using accurate and mislocated positions of the D-shuttle dosimeters. For comparison, the cumulative radioactivities were also estimated from the PET images, and then, the absorbed doses and effective dose were computed. The maximum bias of the average estimated cumulated radioactivities and the effective doses was − 15.0% and − 19.7% for the 1 cm shifted positions, respectively. The ratios of absorbed doses obtained from D-shuttle and PET measurement against the actual values were between 0.9 and 1.3, and 0.7 and 1.0, respectively. The bias associated with the D-shuttle dosimeter positions was significant and probably consistent, and both dosimetric techniques exhibited good performance in this phantom study.