Download focal spot size impact to digital x-ray image brightness

WO-7
FOCAL SPOT SIZE IMPACT TO DIGITAL
X-RAY IMAGE BRIGHTNESS STATISTICS
L. Bumbure, Y. Dekhtyar, T. Kirsanova, J. Mosara,
L. Shuvalova
Biomédical Engineering and Nanotechnology Institute,
Riga Technical University, Latvia
E-mail: [email protected]
Abstract
Quality control (QC) tests play a crucial role for patients' dose management in
diagnostic radiography. X-ray tube focal spot size is the most important index for QC of
x-ray machines. The focal spot size is typically estimated using pinhole or the star
pattern. However, the results by such the techniques are rather subjective. This
disadvantage could be decreased in digital radiography when an image brightness
distribution histogram is in use. Average value of brightness correlates to the focal spot
size (significance 0.99). This technique does not require any test object. The results
demonstrate that the above approach could become an important part of the QC for
digital x-ray diagnostic radiography equipment.
Key words: Digital radiography quality assurance, focal spot size
Introduction
Quality control (QC) tests play a crucial role for patients' dose
management in diagnostic radiography. Evaluation of changes in an x-ray tube
focal spot size is one of the most important measurements used for QC in
medical x-ray facilities. Usually focal spot size is estimated using pinhole or the
star pattern. However, these techniques are enough expensive, and the results
acquired in such measurements are rather subjective. In contrast, an x-ray
image brightness distribution histogram is a more reliable tool, which allows
estimating changes of focal spot size digitally, using mathematical/statistical
approach. The tool makes it possible to determine quite small (a decimal part of
millimeter) changes of focal spot size. Average value of brightness correlates to
the focal spot size (significance 0.99). This technique does not require any
phantom, test object or other special expensive equipment and any printed
hardcopy of a digital x-ray image being analyzed. The whole process goes on
digitally straight on the screen, thereby improving the reliability of the quality
test and reducing all related costs. The results demonstrate that the above
approach could become an important part of the QC for digital x-ray diagnostic
radiography equipment.
69
WO-7
Methods
The x-ray machine "Digital Diagnost" was employed to supply a flow of
radiation. Digital images were acquired in manual mode. The voltage was equal
to 81 kV and the milliampere seconds were 2mAs. All images were acquired
using both a small focal spot size (0.6 mm) and a large focal spot size (1.2
mm). The x-ray tube was placed at lm distance from digital detector.
Collimator was adjusted to size 19x25 cm for all images. No phantoms or test
objects were in use. All images were acquired in DICOM format. Using Print
Screen (PrtSc) button the images were transferred to Paint. Here, in Paint, each
image was divided to separate areas (50x50 pixels each area) in directions,
according to a transparent template (Fig. 1) fixed to the monitor screen. Each
area was saved as separate TIFF-image. The TIFF-images were transferred
using software designed, hence pixels' distribution over brightness for each
TIFF-image was produced. The data was processed using excel program. As a
result, an average brightness was calculated for each extracted area. The results
were compared using T-test (significance 0.99).
Anode >ide
\320°
-r
\
6
40^^
0°
5
29,
25
1
28
24
4
27
270°
31
23
90°
26, 2 22
21- 20 19- 18
1- 14 15 1Ó 17-
H34 8 30
1
35
9
36'
220 37' /
31
10
32
11
\
12
/
180
13
33
\
\
140\
Cathode side
Fig. 1. The transparent template
70
WO-7
Results and Discussion
The results represented on the figures 2 - 5 demonstrate that brightness of
images, acquired with large focal spot is more intense than brightness of
images, acquired with small focal spot.
aver age
Image bríghtness_directíon 0"
56 n
55
-A.
>
53
SS8U
-^"^^
51
J^—^
-""^^
1
•
V^
Large focus
—O—Small focus
46
45
C
1
2
3
4
5
6
7
Area's number on the template
Fig. 2. Image brightness changes
Image brightness.directioSO**
60-,
59
58
SI
1!
S
»
S
•
.^-•-.
/
56
55
54
53
52
51
^--•^
1 50
\
\
- Large focus
-Small focus
p
\
\
to ^y
46
45
44 -
^
^
—
o
•
V
13
14
15
16
17
Area's number on the template
Fig. 3. Image brightness changes
This happens, because x-ray beam intensity produced with large focus
reduces due to higher scattering in comparison with small focus beam. The
relevance is the same in all direction, showed on the template. Besides, so
called "heel effect" is well visible. The brightness of image on "anode side" is
superior (Fig. 2), because the beam intensity there is less, in turn, brightness at
"cathode side" is less (Fig. 4), for the beam intensity here is higher. Using Ttest (significance 0.99) it was determined that brightness distribution is
different for images acquired with large focal spot and for images acquired with
small focal spot.
71
WO-7
Image brightness-direction 180"
.
51 ,
50
49
^
i
S 47
S 46
—•— Large focus
^^—^..
^^\
g 44
•i'42
m 4,
40
39
1
^
^ *
>^^^
J
7
8
9
—0—Small focus
'
/vO^
•
V> i
10
11
12
13
Area's number on the template
Fig. 4. Image brightness changes
Image brightness_direction 270"
/
>
>
Cr~r\
V
\\
\
\*
o
-Large focus
-Small focus
Area's number on tbe template
Fig. 5. Image brightness changes
There is decided to make more investigations (using 60kV and lOOkV
energies) to enlarge application's range of the approach.
Conclusions. The results demonstrate that the above approach could
become an important part of the QC for digital x-ray diagnostic radiography
equipment, particularly, to observe changes in focal spot size. Brightness
average is the parameters, which characterize the changes.
72