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.
Image filtering as an alternative to the application of a different reconstruction kernel in CT imaging: Feasibility study in lung cancer screening
Rent:
Rent this article for
USD
10.1118/1.3590363
/content/aapm/journal/medphys/38/7/10.1118/1.3590363
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/38/7/10.1118/1.3590363

Figures

Image of FIG. 1.
FIG. 1.

MTFs derived from PSFs measured in two scanners, Scanner1 (a) and Scanner2 (b), for three built-in reconstruction kernels (f SMOOTH, f STD, and f SHARP).

Image of FIG. 2.
FIG. 2.

Filter functions in the spatial-frequency domain obtained as the ratio of MTFs [Fig. 1(a)] in Scanner1 for all combinations of reconstruction kernels.

Image of FIG. 3.
FIG. 3.

A representative example for the optimization of Hann window settings of w 1 and Δw in Eq.(12). (a) Reconstructed f SMOOTHimage of the high-contrast CT test phantom. The phantom includes five cylindrical objects (diameters of 2–10 mm), (b) the f SHARP/f SMOOTHfiltered image obtained from the image (a), (c) reconstructed f SHARPimage, (d) the image obtained by subtracting the image (c) from (b). The SDs were obtained in five ROIs placed on the image and they were averaged, (e) the mean value of SDs depending on the Hann window settings. The smallest value is found (arrow).

Image of FIG. 4.
FIG. 4.

Filter functions in the spatial-frequency domain, obtained as the ratio of MTFs [Fig. 1(b)] in Scanner2 for all combinations of reconstruction kernels, multiplied by the Hann window.

Image of FIG. 5.
FIG. 5.

Schematic explanation of SDdiff, SDBG,filtered, and SDBG,recon, which were obtained as the standard deviations in the ROI1, ROI2, and ROI3, respectively. The elliptic shape of ROI1 is placed at the center of the FOV in which the major and minor axes of the ellipse are the sizes of FOV multiplied by 0.85 and 0.70, respectively.

Image of FIG. 6.
FIG. 6.

(a) Reconstructed f STD image obtained in CT screening for lung cancer. An abnormal shadow suspected as being lung cancer is detected (arrow), (b) the f SHARP/f STD filtered image obtained from the image (a), (c) reconstructed f SHARP image, (d) the image obtained by subtracting the image (c) from (b). (e) Reconstructed f STD image, (f) the f SMOOTH/f STDfiltered image obtained from the image (e), (g) reconstructed f SMOOTHimage, (h) the image obtained by subtracting the image (g) from (f). Window settings were C/W = −700/800 HU in (a)–(c) and 20/300 HU in (e)–(g).

Image of FIG. 7.
FIG. 7.

(a) Reconstructed f SMOOTHimage obtained in CT screening for lung cancer, (b) the f SHARP/f SMOOTHfiltered image obtained from the image (a), (c) reconstructed f SHARPimage, (d) the image obtained by subtracting the image (c) from (b). (e) CT value profile (dot) in the f SHARP/f SMOOTHfiltered image (b), the profile (line) in the reconstructed f SHARPimage (c), and the profile (dashed-line) in the reconstructed f SMOOTHimage (a) are shown along a horizontal line indicated by the dashed-line in the image (a). The window setting was C/W = −600/1000 HU for (a)–(c).

Image of FIG. 8.
FIG. 8.

MTFs obtained from the filtered images (dashed-line and open circle) compared with the MTF (solid line) obtained from the original reconstructed image, for each reconstruction kernel.

Image of FIG. 9.
FIG. 9.

(a) Reconstructed f SMOOTHimage of the chest phantom, (b) the f SHARP/f SMOOTHfiltered image obtained from the image (a), (c) reconstructed f SHARPimage, (d) the image obtained by subtracting the image (c) from (b). (e) CT value profile (dot) in the f SHARP/f SMOOTH filtered image (b), the profile (line) in the reconstructed f SHARPimage (c), and the profile (dashed-line) in the reconstructed f SMOOTHimage (a) are shown along a horizontal line indicated by the dashed-line in (a). The window setting was C/W = −700/500 HU for (a)–(c).

Tables

Generic image for table
TABLE I.

The values of SDdif f, SDBG,filtered, and SDBG,reco n (defined in Fig. 5) obtained for clinical chest images. The data are presented as mean ± standard deviation.

Generic image for table
TABLE II.

Spatial frequencies at which the MTFs (Fig. 8) have values of 50%, 10%, and 4%.

Loading

Article metrics loading...

/content/aapm/journal/medphys/38/7/10.1118/1.3590363
2011-06-16
2014-04-24
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Image filtering as an alternative to the application of a different reconstruction kernel in CT imaging: Feasibility study in lung cancer screening
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/38/7/10.1118/1.3590363
10.1118/1.3590363
SEARCH_EXPAND_ITEM