Browsing by Subject "Partial Volume Effect"

Brain Positron emission tomography (PET) has been widely employed for the clinic diagnosis of Alzheimer's disease (AD). Studies have shown that PET imaging is helpful in differentiating healthy elderly individuals, mild cognitive impairment (MCI) individuals, and AD individuals (Nordberg, Rinne, Kadir, & Långström, 2010). However, PET image quality and quantitative accuracy is degraded from partial volume effects (PVEs), which are due to the poor spatial resolution of PET. As a result, the compensation of PVEs in PET may be of great significance in the improvement of early diagnosis of AD. There are many different approaches available to address PVEs including region-based methods and voxel-based methods. In this study, a voxel-based PVE compensation technique using high-resolution anatomical images was investigated. The high-resolution anatomical images could be computed tomography (CT) or magnetic resonance imaging (MRI) images. Such methods have been proposed and investigated in many studies (Vunckx et al., 2012). However, relatively little research has been done on comparing the effects of different MRI images on voxel-based PVE correction methods. In this study, we compare the effect of 6 different MRI image protocols on PVE compensation in PET images. The MRI protocols compared in this study are T1-, T2-, proton-density (PD)-weighted and 3 different inversion recovery MRI protocols.

Results: OSEM and MAP/ICD images with isotropic prior are blurry and/or noisy. Compared with the OSEM and MAP/ICD images obtained by using an isotropic prior, the PET image reconstructed using anatomical information show better contrast and less noise. Visually, the PET image reconstructed with the ZeroCSF prior gave the PET image that visually appears to match best with the PET phantom. PET images reconstructed with T2, PD and ZeroWM image are similar to one another in image quality, but relative to the PET phantom and the ZeroCSF PET image, these images have poor contrast between CSF pockets and surrounding GM tissue, and they have less contrast between GM and WM. PET image reconstructed with T1 image had a better GM and CSF contrast, some of the CSF pockets in GM were reconstructed, but the WM region was very noisy. PET images reconstructed with ZeroGM image had noticeably worse performance on the GM reconstruction. Analysis suggest that these effects are caused by differences in tissue contrast with different MRI protocols

Keywords: PET, MRI, partial volume effect, image reconstruction, SPECT, Alzheimer's disease.

AbstractIntroduction: SPECT and PET are long established methods for functional imaging of bone lesions, including lesions in bone marrow and bone metastasis. These imaging modalities are however limited by poor spatial resolution which degrades quantitative accuracy and precise localization. This limitation in quantitative accuracy corresponds to the partial volume effect (PVE), in which a portion of the radiotracer activity truly in one structure appears, in the image, to be in nearby image voxels. To some extent PVEs can be corrected by iterative image reconstruction algorithms, such as ordered-subsets expectation maximization (OSEM), that model spatial resolution. This approach is however limited by noise, which is amplified as spatial resolution is recovered and PVEs are reduced. SPECT and PET imaging often involves CT as well. CT provides very high-resolution anatomical information which can be used to correct PVEs in SPECT and PET. One approach to PVE correction is using Markov Random Fields (MRFs) that incorporate anatomical information. However, there has been relatively little investigation on MRF-based PVE correction for SPECT/PET bone imaging using CT information. In this work, two types of CT anatomical information are considered: (i) partial anatomical information (pAI) which distinguishes, for example, compact bone from bone marrow but does not otherwise distinguish the tumor from surrounding tissue and (ii) full anatomical information (fAI), which fully distinguishes tumor from surrounding tissue. Image reconstructions involving pAI and fAI are referred to as RpAI and RfAI, respectively. RfAI is expected to provide the best correction of tumor PVEs, but RpAI may be more often available from CT images. The objective of the work is to assess the effectiveness of RpAI as compared to RfAI and OSEM.

Methods: Radiotracer (SPECT/PET) and attenuation coefficient (CT) phantoms were generated using XCAT software. Tumor lesions with high activity were added to the bone marrow in the radiotracer phantom. Two CT phantoms, pAI and fAI, were generated, with the fAI CT phantom including reduced CT number in the tumor-lesion locations. Projection data were simulated, and images were reconstructed using the computer code SPECT-MAP, with modeled spatial resolutions of 12mm (SPECT-like data) and 6mm (PET-like data). The RpAI and RfAI image reconstructions were performed using the iterative coordinate descent (ICD) algorithm and the Bowsher prior. The reconstructions were performed with projection data at 4 noise levels: 5M-, 50M-, and 100M-counts and noise-free. Reconstructed images were evaluated by visual inspection and by root-mean-square (RMS) error across the entire image and in 2 small ROIs (ROI-1 and ROI-2) surrounding the tumor lesions.

Results: The estimated rmsemin calculated from ROI-1 and ROI-2 reconstructed images of noisy (5M counts) projection data with res-12mm using OSEM, RpAI and RfAI were 0.92E-5 & 0.82E-5 (both at iteration 5, subset 9), 0.76E-5 & 0.65E-5 (both at OPS of 1.0E+4), and 0.44E-5 & 0.44E-5 (both at OPS of 1.0E+4), respectively; while for res-6mm, the rmsemin were 0.85E-5 & 0.81E-5 (both at iteration 10, subset 9), 0.57E-5 & 0.53E-5 (both at OPS of 1.0E+4), 0.37E-5 & 0.37E-5 (both at OPS of 1.0E+4), respectively. At both spatial resolutions, the RpAI reconstructions, using partial anatomical information only, provided reduced RMS errors compared to OSEM. Conclusions: At spatial resolutions characteristic of SPECT and PET, the partial anatomical information available from normal bone structures such as marrow and compact bone can improve estimation of hot-spot lesions, as measured by visual inspection and RMS error.