Title :
Parameter optimization and effective imaging volume determination of helical scan for a pinhole animal SPECT
Author :
Rutao Yao ; Xiao Deng ; Tiantian Dai ; Qingyang Wei ; Beaudoin, Jean-Francois ; Naaman, Charles ; Tianyu Ma ; Cadorette, Jules ; Lecomte, Roger
Author_Institution :
Dept. of Nucl. Med., State Univ. of New York at Buffalo, Buffalo, NY, USA
fDate :
Oct. 27 2012-Nov. 3 2012
Abstract :
We developed SPECT imaging capability on a LabPETTM animal scanner. The add-on SPECT uses a multiplepinhole collimator and a step-and-shoot helical scan scheme. The objective of this work was to determine I) the effective imaging volume of a given scan protocol, and 2) the helical scan parameters, that is, the angular and axial step sizes, for achieving uniform high resolution images throughout the imaging volume. Three rotational step sizes, {1°, 2°, 3°}, and three axial step sizes, {0.3 mm, 0.6 mm, 1.2 mm}, were combined to form 9 helical scan protocols. These step sizes were derived based on the reference values analytically calculated using the spatial resolution of the SPECT system and the Nyquist sampling theorem. Two numerical indices, the sampling completeness percentage (SCP), and the directional resolution (DR), were designed to quantify the projection sampling and resolution performance of the helical protocols. Scanning of a digital sphere-grid phantom centered in the SPECT system´s FOV was simulated for each of the 9 helical protocols. Aggregated RMS resolution values derived from the image of distributed spheres over the FOV were used to assess the resolution performance of the scan protocols, and for validating the predictive value of the SCP and DR indices, which were calculated over the matching grid. The RMS results showed that the effective image volume of a helical scan where sampling is adequate was defined by the start and end pinhole planes. The results also showed that using a small axial step size (about half the value derived from Nyquist theorem) achieved best imaging resolution, while the angular step size was less critical. The RMS images correlated well with that of SCP and DR. We conclude the following for designing a helical scan protocol with a multiple pinhole collimator. First, the effective imaging volume should be confined between the start and end planes of the pinholes´ movement range. Second, using a- axial step size that is half of what is calculated from the Nyquist theorem is desirable for optimal resolution performance. In addition, SCP and DR are useful indices for assessing scan protocols analytically.
Keywords :
Nyquist criterion; biomedical measurement; collimators; image resolution; image sampling; medical image processing; parameter estimation; phantoms; single photon emission computed tomography; volume measurement; LabPETTM; Nyquist sampling; digital sphere-grid phantom; effective image volume; image resolution; imaging volume determination; multiple pinhole collimator; parameter optimization; pinhole animal SPECT; rotational step; step-and-shoot helical scan scheme; SPECT; multiple pinhole; optimal helical pitch;
Conference_Titel :
Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC), 2012 IEEE
Conference_Location :
Anaheim, CA
Print_ISBN :
978-1-4673-2028-3
DOI :
10.1109/NSSMIC.2012.6551872