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.
Noise spatial nonuniformity and the impact of statistical image reconstruction in CT myocardial perfusion imaging
Rent:
Rent this article for
USD
10.1118/1.4722983
/content/aapm/journal/medphys/39/7/10.1118/1.4722983
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/39/7/10.1118/1.4722983

Figures

Image of FIG. 1.
FIG. 1.

Reconstructions of the numerical phantom at different time frames. (a)–(d) Phantoms were reconstructed using FBP. (e)–(h)Maps of the standard deviation in the FBP reconstruction. (i)–(l) Phantoms were reconstructed using SIR. (m)–(p) The standard deviation maps calculated from SIR reconstructions. The noise is generally more uniform and of lower level in SIR images than in FBP reconstructions. The noise in SIR images has a negligible dependence on the source trajectory. The line segments superposed on the FBP reconstructions represent the short-scan source trajectory. The ROIs in (a) were used for the measurement of enhancement curves (Fig. 2). The display range was set to [0, 0.03] mm−1 for the reconstructions.

Image of FIG. 2.
FIG. 2.

Dynamic contrast enhancement curves in the noncardiac muscle ROI [A in Fig. 1(a)]. Also plotted is the relative noise standard deviation as a function of time in the same region. Notice that the level of noise-induced fluctuations in the enhacement curved match the trend observed in the relative noise standard deviation plots. SIR offers less temporal variations in noise and a lower level than FBP. The noise level simulated an incident fluence of 106 photons/detector element.

Image of FIG. 3.
FIG. 3.

Dynamic contrast enhancement curves in the myocardium ROI [B in Fig. 1(a)]. Also plotted is the relative noise standard deviation as a function of time in the same region. Notice that the level of noise-induced fluctuations in the enhacement curved match the trend observed in the relative noise standard deviation plots. SIR offers less temporal variations in noise and a lower level than FBP. The noise level simulated an incident fluence of 106 photons/detector element.

Image of FIG. 4.
FIG. 4.

Dynamic contrast enhancement curves in the left ventricle ROI [C in Fig. 1(a)]. Also plotted is the relative noise standard deviation as a function of time in the same region. Notice that the level of noise-induced fluctuations in the enhacement curved match the trend observed in the relative noise standard deviation plots. SIR offers less temporal variations in noise and a lower level than FBP. The noise level simulated an incident fluence of 106 photons/detector element.

Image of FIG. 5.
FIG. 5.

Dynamic contrast enhancement curves in the infarct ROI [D in Fig. 1(a)]. Also plotted is the relative noise standard deviation as a function of time in the same region. Notice that the level of noise-induced fluctuations in the enhacement curved match the trend observed in the relative noise standard deviation plots. SIR offers less temporal variations in noise and a lower level than FBP. The noise level simulated an incident fluence of 106 photons/detector element.

Image of FIG. 6.
FIG. 6.

Reconstructions of the in vivo porcine datasets at different tube currents. The images were reconstructed using FBP and SIR. The display range was [−1000, 900] HU.

Image of FIG. 7.
FIG. 7.

Definition of the regions of interest (ROI) used for the perfusion metric measurements. PM was located near the papillary muscle, while AW was situated in the anterior wall of the myocardium. Both ROIs were 3 × 5 voxels. The display range for this image was [−1000, 900] HU.

Image of FIG. 8.
FIG. 8.

Time attenuation curves (TAC) measured from reconstructions of the in vivo porcine datasets at different tube currents (a, b) and the standard deviation at different time frames (c, d). Images were reconstructed using FBP and SIR. The ROI were the measurements were performed was located in the papillary muscle, PM in Fig. 7. Note that in order to optimize the visualization of the data, the range of attenuation coefficient shown is not constant between the different plots. Also, some divergence from the 500 mA curves might be explained by the fact that a different scan was acquired for each tube current setting.

Image of FIG. 9.
FIG. 9.

Time attenuation curves (TAC) measured from reconstructions of the in vivo porcine datasets at different tube currents. The images were reconstructed using FBP and SIR. Two ROIs were used for the measurements; their location is shown above. Note that in order to optimize the visualization of the data, the range of attenuation coefficient shown is not constant between the different plots. Also, some divergence from the 500 mA curves might be explained by the fact that a different scan was acquired for each tube current setting.

Tables

Generic image for table
TABLE I.

Numerical simulation parameters.

Generic image for table
TABLE II.

In vivo dataset parameters.

Generic image for table
TABLE III.

Quantitative metrics of myocardial perfusion measured in images reconstruction using FBP and SIR in the anterior myocardial wall (AW) and in the papillary muscle (PM) as defined in Fig. 7. Each measurement (f) is presented with the percent deviation from the values calculated from the 500 mA scan (f ref), |ff ref|/f ref × 100% in parenthesis. Note that some divergence from the reference might be explained by the fact that a different scan was acquired for each tube current setting.

Loading

Article metrics loading...

/content/aapm/journal/medphys/39/7/10.1118/1.4722983
2012-06-08
2014-04-17
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Noise spatial nonuniformity and the impact of statistical image reconstruction in CT myocardial perfusion imaging
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/39/7/10.1118/1.4722983
10.1118/1.4722983
SEARCH_EXPAND_ITEM