Limits...
Experimental determination of the weighting factor for the energy window subtraction-based downscatter correction for I-123 in brain SPECT studies.

de Nijs R, Holm S, Thomsen G, Ziebell M, Svarer C - J Med Phys (2010)

Bottom Line: After correction, the measured specific binding ratio (image contrast) increased significantly for healthy subjects, typically by more than 20%, while the background counts outside of all subjects were effectively removed.A weighting factor of 1.1-1.2 can be applied in clinical practice.This correction effectively removes downscatter and significantly improves image contrast inside the brain.

View Article: PubMed Central - PubMed

Affiliation: Department of Clinical Physiology, Nuclear Medicine, PET and Cyclotron Unit, Copenhagen University Hospital, Copenhagen, Denmark.

ABSTRACT
Correction for downscatter in I-123 SPECT can be performed by the subtraction of a secondary energy window from the main window, as in the triple-energy window method. This is potentially noise sensitive. For studies with limited amount of counts (e.g. dynamic studies), a broad subtraction window with identical width is preferred. This secondary window needs to be weighted with a factor higher than one, due to a broad backscatter peak from high-energy photons appearing at 172 keV. Spatial dependency and the numerical value of this weighting factor and the image contrast improvement of this correction were investigated in this study. Energy windows with a width of 32 keV were centered at 159 keV and 200 keV. The weighting factor was measured both with an I-123 point source and in a dopamine transporter brain SPECT study in 10 human subjects (5 healthy subjects and 5 patients) by minimizing the background outside the head. Weighting factors ranged from 1.11 to 1.13 for the point source and from 1.16 to 1.18 for human subjects. Point source measurements revealed no position dependence. After correction, the measured specific binding ratio (image contrast) increased significantly for healthy subjects, typically by more than 20%, while the background counts outside of all subjects were effectively removed. A weighting factor of 1.1-1.2 can be applied in clinical practice. This correction effectively removes downscatter and significantly improves image contrast inside the brain.

No MeSH data available.


Related in: MedlinePlus

Illustration of the edge detection in a healthy subject (subject 3). The left panel (a) shows the so-called sinogram for a slice. Vertically the tangential position, and horizontally the 120 angles, are shown. The middle panel (b) shows the corresponding reconstructed slice. The edge is shown with maximum intensity (white). The right panel (c) shows the edge with a radial offset of 3 pixels.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2990116&req=5

Figure 0002: Illustration of the edge detection in a healthy subject (subject 3). The left panel (a) shows the so-called sinogram for a slice. Vertically the tangential position, and horizontally the 120 angles, are shown. The middle panel (b) shows the corresponding reconstructed slice. The edge is shown with maximum intensity (white). The right panel (c) shows the edge with a radial offset of 3 pixels.

Mentions: SPECT imaging was performed by recording projection data at 120 fixed angles, with an interval of 3° and a noncircular orbit. The mean radius of rotation was 13.9 cm. Reconstruction of projection data with standard filtered back-projection (FBP), both with and without downscatter correction, was performed in MATLAB 7.5 (Mathworks, USA). Matrix size was 128 × 128, with 2.33 mm pixels and identical slice thickness. A 3D low-pass 4th-order Butterworth post-filter with a cut-off frequency of 0.3 Nyquist (= 0.64 cm–1) was used. Attenuation correction with Chang’s first-order correction[13] was applied with an empirical linear attenuation factor of 0.10 cm–1 for I-123 imaging without Compton scatter correction of the primary 159 keV photons.[1114–16] The determination of the attenuation map was aided by an algorithm, which finds the most outward placed crossing of a manually set threshold (tuned at the edge) and the intensity for every projection angle in the sinogram [Figure 2a and b].


Experimental determination of the weighting factor for the energy window subtraction-based downscatter correction for I-123 in brain SPECT studies.

de Nijs R, Holm S, Thomsen G, Ziebell M, Svarer C - J Med Phys (2010)

Illustration of the edge detection in a healthy subject (subject 3). The left panel (a) shows the so-called sinogram for a slice. Vertically the tangential position, and horizontally the 120 angles, are shown. The middle panel (b) shows the corresponding reconstructed slice. The edge is shown with maximum intensity (white). The right panel (c) shows the edge with a radial offset of 3 pixels.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2990116&req=5

Figure 0002: Illustration of the edge detection in a healthy subject (subject 3). The left panel (a) shows the so-called sinogram for a slice. Vertically the tangential position, and horizontally the 120 angles, are shown. The middle panel (b) shows the corresponding reconstructed slice. The edge is shown with maximum intensity (white). The right panel (c) shows the edge with a radial offset of 3 pixels.
Mentions: SPECT imaging was performed by recording projection data at 120 fixed angles, with an interval of 3° and a noncircular orbit. The mean radius of rotation was 13.9 cm. Reconstruction of projection data with standard filtered back-projection (FBP), both with and without downscatter correction, was performed in MATLAB 7.5 (Mathworks, USA). Matrix size was 128 × 128, with 2.33 mm pixels and identical slice thickness. A 3D low-pass 4th-order Butterworth post-filter with a cut-off frequency of 0.3 Nyquist (= 0.64 cm–1) was used. Attenuation correction with Chang’s first-order correction[13] was applied with an empirical linear attenuation factor of 0.10 cm–1 for I-123 imaging without Compton scatter correction of the primary 159 keV photons.[1114–16] The determination of the attenuation map was aided by an algorithm, which finds the most outward placed crossing of a manually set threshold (tuned at the edge) and the intensity for every projection angle in the sinogram [Figure 2a and b].

Bottom Line: After correction, the measured specific binding ratio (image contrast) increased significantly for healthy subjects, typically by more than 20%, while the background counts outside of all subjects were effectively removed.A weighting factor of 1.1-1.2 can be applied in clinical practice.This correction effectively removes downscatter and significantly improves image contrast inside the brain.

View Article: PubMed Central - PubMed

Affiliation: Department of Clinical Physiology, Nuclear Medicine, PET and Cyclotron Unit, Copenhagen University Hospital, Copenhagen, Denmark.

ABSTRACT
Correction for downscatter in I-123 SPECT can be performed by the subtraction of a secondary energy window from the main window, as in the triple-energy window method. This is potentially noise sensitive. For studies with limited amount of counts (e.g. dynamic studies), a broad subtraction window with identical width is preferred. This secondary window needs to be weighted with a factor higher than one, due to a broad backscatter peak from high-energy photons appearing at 172 keV. Spatial dependency and the numerical value of this weighting factor and the image contrast improvement of this correction were investigated in this study. Energy windows with a width of 32 keV were centered at 159 keV and 200 keV. The weighting factor was measured both with an I-123 point source and in a dopamine transporter brain SPECT study in 10 human subjects (5 healthy subjects and 5 patients) by minimizing the background outside the head. Weighting factors ranged from 1.11 to 1.13 for the point source and from 1.16 to 1.18 for human subjects. Point source measurements revealed no position dependence. After correction, the measured specific binding ratio (image contrast) increased significantly for healthy subjects, typically by more than 20%, while the background counts outside of all subjects were effectively removed. A weighting factor of 1.1-1.2 can be applied in clinical practice. This correction effectively removes downscatter and significantly improves image contrast inside the brain.

No MeSH data available.


Related in: MedlinePlus