Limits...
Investigation of different sparsity transforms for the PICCS algorithm in small-animal respiratory gated CT.

Abascal JF, Abella M, Sisniega A, Vaquero JJ, Desco M - PLoS ONE (2015)

Bottom Line: We compared variants of the PICCS algorithm with different transforms in the prior penalty function: gradient, unitary, and wavelet transform.The resulting data were used to simulate scenarios with different dose levels and numbers of projections.Nevertheless, the wavelet domain was found to be less prone to patchy cartoon-like artifacts than the commonly used gradient domain.

View Article: PubMed Central - PubMed

Affiliation: Departamento de Bioingeniería e Ingeniería Aeroespacial, Universidad Carlos III de Madrid, Madrid, Spain; Instituto de Investigación Sanitaria Gregorio Marañón (IiSGM), Madrid, Spain.

ABSTRACT
Respiratory gating helps to overcome the problem of breathing motion in cardiothoracic small-animal imaging by acquiring multiple images for each projection angle and then assigning projections to different phases. When this approach is used with a dose similar to that of a static acquisition, a low number of noisy projections are available for the reconstruction of each respiratory phase, thus leading to streak artifacts in the reconstructed images. This problem can be alleviated using a prior image constrained compressed sensing (PICCS) algorithm, which enables accurate reconstruction of highly undersampled data when a prior image is available. We compared variants of the PICCS algorithm with different transforms in the prior penalty function: gradient, unitary, and wavelet transform. In all cases the problem was solved using the Split Bregman approach, which is efficient for convex constrained optimization. The algorithms were evaluated using simulations generated from data previously acquired on a micro-CT scanner following a high-dose protocol (four times the dose of a standard static protocol). The resulting data were used to simulate scenarios with different dose levels and numbers of projections. All compressed sensing methods performed very similarly in terms of noise, spatiotemporal resolution, and streak reduction, and filtered back-projection was greatly improved. Nevertheless, the wavelet domain was found to be less prone to patchy cartoon-like artifacts than the commonly used gradient domain.

No MeSH data available.


Related in: MedlinePlus

Influence of the number of projections and X-ray flux on image profiles.Top: Profile (yellow line) drawn over heart and lung areas (left figure) and profile drawn over a bone area (right figure), overimposed on the high-dose protocol image. Middle and bottom: Normalized profiles for reference high-dose FDK (target) and for the low-dose protocol reconstructed with TV-PICCS (TV) and WT-PICCS (WT) for different number of projections (middle) and different X-ray flux values (bottom).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4383608&req=5

pone.0120140.g006: Influence of the number of projections and X-ray flux on image profiles.Top: Profile (yellow line) drawn over heart and lung areas (left figure) and profile drawn over a bone area (right figure), overimposed on the high-dose protocol image. Middle and bottom: Normalized profiles for reference high-dose FDK (target) and for the low-dose protocol reconstructed with TV-PICCS (TV) and WT-PICCS (WT) for different number of projections (middle) and different X-ray flux values (bottom).

Mentions: Fig 5 shows zooms of Fig 4 to better depict differences between TV-PICCS and WT-PICCS when varying the source flux and number of projections. When decreasing the X-ray flux (third column in Fig 5), both algorithms converged in fewer iterations towards more blurred images (the loss of resolution can also be seen in the profile plotted in Fig 6). As the iteration number increases, the sparsity is imposed in the different domains. In the case of TV-PICCS, the algorithm leads to patchy artifacts for all noise levels tested, although with high noise the patchy artifact is less evident. WT- PICCS is more robust against noise level and maintains a more natural texture for all three scenarios. Varying the number of projections has a similar effect on texture: TV-PICCS presents patchy artifacts while WT-PICCS maintains a more natural texture. As the number of projections decreases, the missing data produce streaks that are not removed by any of the algorithms due to the coherent nature of this artifact, which cannot be removed by the TV term common to both algorithms.


Investigation of different sparsity transforms for the PICCS algorithm in small-animal respiratory gated CT.

Abascal JF, Abella M, Sisniega A, Vaquero JJ, Desco M - PLoS ONE (2015)

Influence of the number of projections and X-ray flux on image profiles.Top: Profile (yellow line) drawn over heart and lung areas (left figure) and profile drawn over a bone area (right figure), overimposed on the high-dose protocol image. Middle and bottom: Normalized profiles for reference high-dose FDK (target) and for the low-dose protocol reconstructed with TV-PICCS (TV) and WT-PICCS (WT) for different number of projections (middle) and different X-ray flux values (bottom).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4383608&req=5

pone.0120140.g006: Influence of the number of projections and X-ray flux on image profiles.Top: Profile (yellow line) drawn over heart and lung areas (left figure) and profile drawn over a bone area (right figure), overimposed on the high-dose protocol image. Middle and bottom: Normalized profiles for reference high-dose FDK (target) and for the low-dose protocol reconstructed with TV-PICCS (TV) and WT-PICCS (WT) for different number of projections (middle) and different X-ray flux values (bottom).
Mentions: Fig 5 shows zooms of Fig 4 to better depict differences between TV-PICCS and WT-PICCS when varying the source flux and number of projections. When decreasing the X-ray flux (third column in Fig 5), both algorithms converged in fewer iterations towards more blurred images (the loss of resolution can also be seen in the profile plotted in Fig 6). As the iteration number increases, the sparsity is imposed in the different domains. In the case of TV-PICCS, the algorithm leads to patchy artifacts for all noise levels tested, although with high noise the patchy artifact is less evident. WT- PICCS is more robust against noise level and maintains a more natural texture for all three scenarios. Varying the number of projections has a similar effect on texture: TV-PICCS presents patchy artifacts while WT-PICCS maintains a more natural texture. As the number of projections decreases, the missing data produce streaks that are not removed by any of the algorithms due to the coherent nature of this artifact, which cannot be removed by the TV term common to both algorithms.

Bottom Line: We compared variants of the PICCS algorithm with different transforms in the prior penalty function: gradient, unitary, and wavelet transform.The resulting data were used to simulate scenarios with different dose levels and numbers of projections.Nevertheless, the wavelet domain was found to be less prone to patchy cartoon-like artifacts than the commonly used gradient domain.

View Article: PubMed Central - PubMed

Affiliation: Departamento de Bioingeniería e Ingeniería Aeroespacial, Universidad Carlos III de Madrid, Madrid, Spain; Instituto de Investigación Sanitaria Gregorio Marañón (IiSGM), Madrid, Spain.

ABSTRACT
Respiratory gating helps to overcome the problem of breathing motion in cardiothoracic small-animal imaging by acquiring multiple images for each projection angle and then assigning projections to different phases. When this approach is used with a dose similar to that of a static acquisition, a low number of noisy projections are available for the reconstruction of each respiratory phase, thus leading to streak artifacts in the reconstructed images. This problem can be alleviated using a prior image constrained compressed sensing (PICCS) algorithm, which enables accurate reconstruction of highly undersampled data when a prior image is available. We compared variants of the PICCS algorithm with different transforms in the prior penalty function: gradient, unitary, and wavelet transform. In all cases the problem was solved using the Split Bregman approach, which is efficient for convex constrained optimization. The algorithms were evaluated using simulations generated from data previously acquired on a micro-CT scanner following a high-dose protocol (four times the dose of a standard static protocol). The resulting data were used to simulate scenarios with different dose levels and numbers of projections. All compressed sensing methods performed very similarly in terms of noise, spatiotemporal resolution, and streak reduction, and filtered back-projection was greatly improved. Nevertheless, the wavelet domain was found to be less prone to patchy cartoon-like artifacts than the commonly used gradient domain.

No MeSH data available.


Related in: MedlinePlus