Quantitative Assessment of Lung Cancer Perfusion Using MDCT: Does Measurement Reproducibility Improve with Greater Tumor Volume Coverage?
Quan Sing Ng1,
Vicky Goh2,
Ernst Klotz3,
Heinz Fichte3,
Michele I. Saunders1,
Peter J. Hoskin1 and
Anwar R. Padhani2
1 Marie Curie Research Wing, Mount Vernon Hospital, Northwood, Middlesex, United
Kingdom.
2 Paul Strickland Scanner Centre, Mount Vernon Hospital, Rickmansworth Rd.,
Northwood, Middlesex, United Kingdom HA6 2RN.
3 Siemens Medical Solutions, Forchheim, Germany.
View larger version (63K):
[in a new window]
|
Fig. 1A —72-year-old man with lung tumor. Region of interest (ROI) is
drawn freehand around lung tumor.
|
|
View larger version (63K):
[in a new window]
|
Fig. 1B —72-year-old man with lung tumor. Color parametric map of
vascular blood volume or permeability is automatically generated by perfusion
software. Each pixel within ROI corresponds to single perfusion value,
allowing for pixel-by-pixel analysis.
|
|
View larger version (24K):
[in a new window]
|
Fig. 1C —72-year-old man with lung tumor. Color parametric maps of
vascular blood volume or permeability for four contiguous 10-mm axial levels.
By amalgamating data from all individual pixels from four levels, tumor
perfusion values can be derived for 40-mm z-axis coverage.
|
|
View larger version (8K):
[in a new window]
|
Fig. 2A —Bland-Altman agreement plots of permeability (mL/100 mL/min).
Plots show difference in permeability between two scans against mean of
permeability values for 10-mm (A) and 40-mm (B) tumor coverage.
Narrowing of 95% limits of agreement with 40-mm coverage indicates improvement
in measurement reproducibility.
|
|
View larger version (8K):
[in a new window]
|
Fig. 2B —Bland-Altman agreement plots of permeability (mL/100 mL/min).
Plots show difference in permeability between two scans against mean of
permeability values for 10-mm (A) and 40-mm (B) tumor coverage.
Narrowing of 95% limits of agreement with 40-mm coverage indicates improvement
in measurement reproducibility.
|
|
View larger version (8K):
[in a new window]
|
Fig. 3A —Bland-Altman agreement plots of blood volume (mL/100 mL).
Plots show difference in blood volume between two scans against mean of blood
volume values for 10-mm (A) and 40-mm (B) tumor coverage.
Narrowing of 95% limits of agreement with 40-mm coverage indicates improvement
in measurement reproducibility.
|
|
View larger version (8K):
[in a new window]
|
Fig. 3B —Bland-Altman agreement plots of blood volume (mL/100 mL).
Plots show difference in blood volume between two scans against mean of blood
volume values for 10-mm (A) and 40-mm (B) tumor coverage.
Narrowing of 95% limits of agreement with 40-mm coverage indicates improvement
in measurement reproducibility.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Copyright © 2006 by the American Roentgen Ray Society.
