TY - JOUR T1 - Is movement an issue in ultrafast cardiac SPECT imaging systems? JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1365 LP - 1365 VL - 51 IS - supplement 2 AU - Catherine McGrath AU - Kenneth Yap AU - Bruce Van Every AU - Lana Volokh AU - Martin Cherk AU - Michael Kelly AU - Victor Kalff Y1 - 2010/05/01 UR - http://jnm.snmjournals.org/content/51/supplement_2/1365.abstract N2 - 1365 Objectives A fast imaging cardiac SPECT system, Discovery NM530C (GE Healthcare) employs novel solid state detectors and multi-pinhole collimation focusing on the heart. All system parts are static during acquisition. This study aimed to see if movement can be detected on this system and if so, its impact on image appearance. Methods An anthropomorphic cardiac phantom with 5MBq Tl-201 and 2mL air bubble within its ventricular activity, was imaged stationary and with 1.5cm & 2cm displacements in craniocaudal, lateral and anteroposterior axes. Motion of a point source was performed to identify the pinhole detectors most likely to detect displacement. Using GE proprietary list mode software acquisitions were reframed into 30sec/frame, in-house software reprocessed and extracted the anterior and left lateral pinhole detector images into cine format. Two patient volunteers undergoing clinical cardiac studies (0.8MBq/kg Tl-201, 8minutes), were imaged stationary and then displaced 2cm & 4cm craniocaudally midway through acquisition. Movement was then sought in 38 sequential clinical studies. Results Phantom defect visualization decreased with increasing displacement, especially craniocaudally. Point source motion analysis demonstrated that the anterior and left lateral pinhole detectors optimally displayed movement in all axes. The 2 volunteers demonstrated normal cardiac perfusion and their respective displacements were easily detected by the in-house software. Observable image degradation occurred with movement >2cm. In the clinical group, the software detected movement which did not appear to be >2cm and which occured at any phase of imaging in 8/38 patients, even though no movement was clinically observed. Conclusions Despite efficient detector technology patient movement can still occur undetected clinically which may result in image degradation and defect non-visualisation. An easy to interpret movement detection software has been developed to compliment this system and alert the need for reacquisition ER -