The authors have developed a trimodal PET/SPECT/CT scanner for small animal imaging. The gamma ray subsystems are based on monolithic crystals coupled to multianode photomultiplier tubes (MA-PMTs), while computed tomography (CT) comprises a commercially available microfocus x-ray tube and a CsI scintillator 2D pixelated flat panel x-ray detector. In this study the authors will report on the design and performance evaluation of the multimodal system.
X-ray transmission measurements are performed based on cone-beam geometry. Individual projections were acquired by rotating the x-ray tube and the 2D flat panel detector, thus making possible a transaxial field of view (FOV) of roughly 80 mm in diameter and an axial FOV of 65 mm for the CT system. The single photon emission computed tomography (SPECT) component has a dual head detector geometry mounted on a rotating gantry. The distance between the SPECT module detectors can be varied in order to optimize specific user requirements, including variable FOV. The positron emission tomography (PET) system is made up of eight compact modules forming an octagon with an axial FOV of 40 mm and a transaxial FOV of 80 mm in diameter. The main CT image quality parameters (spatial resolution and uniformity) have been determined. In the case of the SPECT, the tomographic spatial resolution and system sensitivity have been evaluated with a99mTc solution using single-pinhole and multi-pinhole collimators. PET and SPECT images were reconstructed using three-dimensional (3D) maximum likelihood and ordered subset expectation maximization (MLEM and OSEM) algorithms developed by the authors, whereas the CT images were obtained using a 3D based FBP algorithm.
CT spatial resolution was 85μm while a uniformity of 2.7% was obtained for a water filled phantom at 45 kV. The SPECT spatial resolution was better than 0.8 mm measured with a Derenzo-like phantom for a FOV of 20 mm using a 1-mm pinhole aperture collimator. The full width at half-maximum PET radial spatial resolution at the center of the field of view was 1.55 mm. The SPECT system sensitivity for a FOV of 20 mm and 15% energy window was 700 cps/MBq (7.8 × 10−2%) using a multi-pinhole equipped with five apertures 1 mm in diameter, whereas the PET absolute sensitivity was 2% for a 350–650 keV energy window and a 5 ns timing window. Several animal images are also presented.
The new small animal PET/SPECT/CT proposed here exhibits high performance, producing high-quality images suitable for studies with small animals. Monolithic design for PET and SPECT scintillator crystals reduces cost and complexity without significant performance degradation.
This study was supported by the Spanish Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica (I+D+I) under Grant No. FIS2010-21216-CO2-01 and Valencian Local Government under Grant PROMETEO 2008/114. The authors also thank Brennan Holt for checking and correcting the text.
II. SYSTEM DESIGN
II.A. CT subsystem
II.B. SPECT subsystem
II.C. PET subsystem
II.D. PET/SPECT/CT integration
III.A. System calibration
III.A.1. CT subsystem geometric calibration
III.A.2. SPECT subsystem geometric calibration
III.B. CT subsystem performance
III.B.1. Spatial resolution
III.B.2. Image uniformity and noise characteristics
III.C. SPECT subsystem performance
III.C.2. Spatial resolution
III.D. Fused images: Animal studies
IV. RESULTS AND DISCUSSION
IV.A.1. Spatial resolution
IV.A.2. Image uniformity and noise characteristics
IV.B.2. Spatial resolution
IV.D. Fused images: Animal studies
- Single photon emission computed tomography
- Computed tomography
- Positron emission tomography
- Medical imaging
- Spatial resolution
Data & Media loading...
Article metrics loading...
Full text loading...