1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Sinogram bow-tie filtering in FBP PET reconstruction
Rent:
Rent this article for
USD
10.1118/1.3096707
/content/aapm/journal/medphys/36/5/10.1118/1.3096707
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/36/5/10.1118/1.3096707

Figures

Image of FIG. 1.
FIG. 1.

2D Fourier transform of a sinogram (left) and basic bow-tie mask in the FT domain for (right).

Image of FIG. 2.
FIG. 2.

Top: Image quality phantom used for contrast and noise measurement. Bottom left and center panels show one slice of the VOIs defined for contrast assessment (, hot; , background). Bottom right panel shows a different part of the phantom where the VOI for noise measurement was drawn .

Image of FIG. 3.
FIG. 3.

One slice of the point source study. The profile was obtained from an ROI of and fitted to a Gaussian function (thick line).

Image of FIG. 4.
FIG. 4.

Results of stackgram filtering for the three interpolation schemes tested. Filled symbols show the results of just transforming to the stackgram domain and back, with no filtering applied.

Image of FIG. 5.
FIG. 5.

Results in contrast vs noise for the filters in Table I.

Image of FIG. 6.
FIG. 6.

Results in radial and tangential resolution, measured as FWHM of a point placed from the center of the FOV vs noise for the filters in Table I.

Image of FIG. 7.
FIG. 7.

Left: Results in radial vs tangential resolution, measured as FWHM of a point placed from the center of the FOV for radial, angular, stackgram, and binary bow-tie filters. Right: (a) Four point sources filtered with a Gaussian angular filter . (b) Four point sources filtered with a bow-tie filter for a value of 0.4 (too low for the points far from the center). The same position-dependent blurring can be observed in both cases.

Image of FIG. 8.
FIG. 8.

Coronal, sagittal, and axial views of an FDG rat brain study. Upper panel: Radial filter (Butterworth, 12th-order, cutoff frequency 35% Nyquist)+axial filter (Gaussian, FWHM 0.94 mm). Lower panel: Proposed filtering scheme: Bow-tie filter filter (Butterworth, 12th-order, cutoff frequency 35% Nyquist)+axial filter (Gaussian, FWHM 0.94 mm).

Image of FIG. 9.
FIG. 9.

Coronal, sagittal, and axial views of an FDG mouse brain study. Upper panel: Radial filter (Butterworth, 12th-order, cutoff frequency 35% Nyquist)+axial filter (Gaussian, FWHM 0.94 mm). Lower panel: Proposed filtering scheme: Bow-tie filter filter (Butterworth, 12th-order, cutoff frequency 35% Nyquist)+axial filter (Gaussian, FWHM 0.94 mm).

Image of FIG. 10.
FIG. 10.

Coronal, sagittal, and axial views of an FDG mouse heart study. Upper panel: Radial filter (Butterworth, 12th-order, cutoff frequency 45% Nyquist)+axial filter (Gaussian, FWHM 0.94 mm). Lower panel: Proposed filtering scheme: Bow-tie filter filter (Butterworth, 12th-order, cutoff frequency 45% Nyquist)+axial filter (Gaussian, FWHM 0.94 mm).

Image of FIG. 11.
FIG. 11.

Normalized profiles corresponding to the dotted lines in Fig. 10: Dashed line, filtering; solid line, filter (proposed method). Arrows indicate the peak-to-valley ratio.

Tables

Generic image for table
TABLE I.

Filters used in the comparison.

Loading

Article metrics loading...

/content/aapm/journal/medphys/36/5/10.1118/1.3096707
2009-04-14
2014-04-23
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Sinogram bow-tie filtering in FBP PET reconstruction
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/36/5/10.1118/1.3096707
10.1118/1.3096707
SEARCH_EXPAND_ITEM