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.
Automated bone removal in CT angiography: Comparison of methods based on single energy and dual energy scans
Rent:
Rent this article for
USD
10.1118/1.3651475
/content/aapm/journal/medphys/38/11/10.1118/1.3651475
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/38/11/10.1118/1.3651475

Figures

Image of FIG. 1.
FIG. 1.

Left: anthropomorphic thorax phantom with home made insert for the evaluation of bone removal techniques. The insert contains six cylinders which can be filled with iodinated contrast agents and bony structures. Right: schematic drawing of three bony structures located in central three cylinders with various holes representing vessels traveling through bone. Numbers are diameters of holes in mm. The concentrations of HA for these inserts are 408 mg HA/ml (top), 800 mg HA/ml (left), and 380 mg HA/ml (right).

Image of FIG. 2.
FIG. 2.

Determination of the dual energy ratios DE iodine and DE calcium for dual energy scans with and without tin filtration (a) Scan without tin filtration. See Fig. 2(b) for legend. (b) Scan with tin filtration at 140 kV. The solid line represents the threshold used to identify the bone in the phantom experiment.

Image of FIG. 3.
FIG. 3.

CNRD n after bone removal for different methods as a function of the dose fraction δ (solid lines) or the weighting factor f (dashed line). The symbols correspond to the values used in the quantitative comparison of the methods. See text for details.

Image of FIG. 4.
FIG. 4.

Qualitative analysis of the performance of dual energy based bone removal methods with tin filtration added. (a) Image scanned at U low = 80 kV with CTDIvol = 12 mGy. (b) Image scanned at U high = 140 kV with CTDIvol = 9 mGy. (c) Mixed image after bone removal by dual energy masking with threshold R < 2.46. (d) Image after bone removal by dual energy subtraction. Figs. (a)–(c): window width/center (W/C) 800/450 HU, Fig. (d): W/C 400/0 HU.

Image of FIG. 5.
FIG. 5.

Qualitative analysis of the performance of dual scan based bone removal methods. (a) Registered nonenhanced image scanned at U low = 80 kV with CTDIvol = 4 mGy. (b) Enhanced image scanned at U low = 80 kV with CTDIvol = 17 mGy. (c) Enhanced image after bone removal by dual scan masking. (d) Image after bone removal by dual scan subtraction. Figs. (a)–(c): window width/center (W/C) 800/450 HU, Fig. (d): W/C 400/0 HU.

Image of FIG. 6.
FIG. 6.

Scatter plot of CT numbers of the images at 80 and 140 kV with tin filtration depicted in Figs. 4(a) and 4(b), respectively. The upper and lower dashed lines represent R = DE iodine and R = DE calcium, respectively. The three solid lines represent the thresholds R = 2.05, R = 2.46, and R = 2.88. Voxels with CT numbers below the dashed-dotted line are excluded from the bone bask to avoid masking the background tissue.

Image of FIG. 7.
FIG. 7.

Influence of the threshold value R on the performance of the dual energy based bone masking method with tin filtration added. (a) Threshold R < 2.05. (b) Threshold R < 2.88. See also Fig. 4(c) for threshold R < 2.46. Figures (a) and (b): window width/center (W/C) 800/450 HU.

Tables

Generic image for table
TABLE I.

Overview of the four investigated bone removal methods. Each method uses a fraction δ of the total dose for either the nonenhanced scan or the high voltage enhanced scan. The dose for each bone removal method equals the dose D total, normalized to 1, of the reference scan.

Generic image for table
TABLE II.

Dual energy ratios with 95% confidence intervals for iodine and calcium with and without tin filtration at 140 kV.

Generic image for table
TABLE III.

Noise scaling factors with 95% confidence intervals. N.a. = not applicable.

Generic image for table
TABLE IV.

Optimal CNRDs and corresponding dose factors for the four bone removal methods. The CNRDs are normalized to the CNRD of the reference scan. For the dual scan based methods, the values are derived fully analytically. For the dual energy based methods, 95% confidence intervals are given.

Loading

Article metrics loading...

/content/aapm/journal/medphys/38/11/10.1118/1.3651475
2011-10-24
2014-04-19
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Automated bone removal in CT angiography: Comparison of methods based on single energy and dual energy scans
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/38/11/10.1118/1.3651475
10.1118/1.3651475
SEARCH_EXPAND_ITEM