Influence of the Pixel Sizes of Reference Computed Tomography on Single-photon Emission Computed Tomography Image Reconstruction Using Conjugate-gradient Algorithm.

PURPOSE: The frame-of-reference using computed-tomography (CT) coordinate system on single-photon emission computed tomography (SPECT) reconstruction is one of the advanced characteristics of the xSPECT reconstruction system. The aim of this study was to reveal the influence of the high-resolution frame-of-reference on the xSPECT reconstruction. METHODS: 99mTc line-source phantom and National Electrical Manufacturers Association (NEMA) image quality phantom were scanned using the SPECT/CT system. xSPECT reconstructions were performed with the reference CT images in different sizes of the display field-of-view (DFOV) and pixel. RESULTS: The pixel sizes of the reconstructed xSPECT images were close to 2.4 mm, which is acquired as originally projection data, even if the reference CT resolution was varied. The full width at half maximum (FWHM) of the line-source, absolute recovery coefficient, and background variability of image quality phantom were independent on the sizes of DFOV in the reference CT images. CONCLUSION: The results of this study revealed that the image quality of the reconstructed xSPECT images is not influenced by the resolution of frame-of-reference on SPECT reconstruction.

Impact of Novel Incorporation of CT-based Segment Mapping into a Conjugated Gradient Algorithm on Bone SPECT Imaging: Fundamental Characteristics of a Context-specific Reconstruction Method.

OBJECTIVES: The latest single-photon emission computed tomography (SPECT)/computed tomography (CT) reconstruction system, referred to as xSPECT Bone™, is a context-specific reconstruction system utilizing tissue segmentation information from CT data, which is called a zone map. The aim of this study was to evaluate the effects of zone-map enhancement incorporated into the ordered-subset conjugated gradient minimization (OSCGM) reconstruction method on SPECT images. METHODS: Image quality with zone-map enhanced OSCGM (OSCGMz) and non-enhanced OSCGM methods was compared using various reconstruction parameters. The compartment phantom had 3 radioactive sections with CT values of about 1000, 250, and 0 HU. SPECT data were acquired using a low-energy high-resolution (LEHR) collimator, with a 256×256 matrix and 2.4-mm pixel size. The performances of the 2 reconstruction methods (OSCGM vs. OSCGMz) were evaluated on the basis of %error, coefficient of variation (%CV), and normalized mean squared error (NMSE), and adequate iterative update numbers were determined. The relative CV representing the ratio of smoothed images to non-smoothed images was calculated to evaluate the effects of the Gaussian filter on each section set with different CT values. RESULTS: On comparing the OSCGM and OSCGMz methods, it was found that the %error of the OSCGMz method tended to show convergence with fewer updates, especially in the high CT value section mimicking bone absorption. In the water section, the %CV of the OSCGMz method was lower than that of the OSCGM method. The NMSE minimum values for the OSCGM and OSCGMz methods were obtained at 30 and 20 updates, respectively. The relative CV for the OSCGMz method in the water section decreased remarkably according to the size of the full width at half maximum (FWHM) of the Gaussian filter. CONCLUSION: The zone-map enhancement contributed to SPECT reconstruction for the reproduction of radioactive concentrations in bone tissues, using a low number of OSCGM updates. Our findings indicated that the incorporation of zone maps into SPECT reconstruction might improve image quality.