Manual
45 Pages
Preview
Page 1
Manual CDMAM 4.0
MANUAL CONTRAST DETAIL PHANTOM CDMAM 4.0 and CDMAM 4.0 Analyser Version 2.3
By M.J. Floor and R.J.M. van der Burght
Artinis Medical Systems Einsteinweg 17 6662 PW Elst The Netherlands Tel: + 31 481 350 980; Fax: + 31 842 105 702
© 2015 - Artinis Medical Systems BV. - Version 2.1
Manual CDMAM 4.0
Contents
1
1
Introduction to the CDMAM
4
2
Description of the CDMAM
6
3
4
5
6
7
2.1
The aluminium base
6
2.2
The PMMA cover
7
2.3
The PMMA plates
7
2.4
Absorption properties
8
Directions for use of the CDMAM
11
3.1
Making images
11
3.2
Handling instructions
12
3.3
The declaration of conformity
12
Evaluation of the CDMAM image
13
4.1
Correction scheme
13
4.2
Correction examples
14
4.3
Presentation of the results
15
CDMAM Analyser
17
5.1
Introduction
17
5.2
Image readout and scoring
17
5.3
Analysis method
17
Program structure
20
6.1
Projects
20
6.2
Groups
21
6.3
Images
21
6.4
Analysis
24
Analysis results
24
7.1
Image result
© 2015 - Artinis Medical Systems BV. - Version 2
24
2
Manual CDMAM 4.0
7.2
Group result
25
7.3
Project result
26
8
Toolbar
29
9
Menu
30
9.1
File
30
9.2
Edit
30
9.3
Project
30
9.4
View
30
9.5
Window
31
9.6
Help
31
10
System demands and installation
32
10.1
Requirements
32
10.2
Use of the USB Key
32
11
FAQ
33
11.1
Analysis
33
11.2
License key
33
12
Warranty Policy Artinis Medical Systems B.V.
34
13
Literature
36
14
Appendices
38
Appendix 1. SCORE FORM OF THE CDMAM PHANTOM
39
Appendix 2.EVALUATION FORM OF THE CDMAM PHANTOM
40
Appendix 3: Software License Agreement (CDMAM Analyser Software)
41
© 2015 - Artinis Medical Systems BV. - Version 2
3
Manual CDMAM 4.0
In mammography it is essential that objects with very small contrast and diameter can be distinguished from the background. Therefore the quality of the technical aspects of mammography equipment should be monitored with regular time-intervals. This is usually performed by measurement of the physical parameters of the X-ray equipment and observation conditions. The main item in quality control however should be the assurance of information transfer from the tissue under examination to the radiologist. Therefore the information content of an image should be monitored. To evaluate the image quality of mammography systems, a contrast detail phantom (CDMAM 3.2) was developed (Bijkerk, Lindeijer and Thijssen, 1995). With this phantom, which has been adapted from the Burger-Rose phantom (Burger, 1950), the threshold contrast of an imaging system is determined as a function of object diameter, by the detection of pairs of low-contrast objects. In the type 3.2 phantom, the objects (gold disks) ranged in diameter from 0.1 - 3.2 mm and in thickness from 0.05 to 1.6 m, resulting in a radiation contrast range of about 1 - 25 % at standard conditions (molybdenum anode, 30 m molybdenum filtration, 28 kV). The constant improvement of image quality in mammography (high contrast film-screen combinations, digital mammography) made it necessary to visualize lower contrast, as well as a higher spatial resolution (smaller objects) at a higher contrast. The CDMAM phantom was modified to facilitate the evaluation of these systems. In the CDMAM 3.4 version, the gold disks range in diameter from 0.06 to 2.0 mm and in thickness from 0.03 to 2.0 m, resulting in a radiation contrast range of about 0.5 - 30 % at standard mammography exposure conditions (Bijkerk, Thijssen and Arnoldussen, 2000). As automatic scoring makes time consuming manual readout obviate, the CDMAM 4.0 could be developed with an increased number of gold discs. With the CDMAM 4.0 the contrast detail (CD) curve of the mammography system under evaluation is determined with a higher accuracy (Floor, Colier and Van der Burght, 2013; Strudley and Young, 2014). More gold discs are available in the relevant areas for psychometric curve fitting. Not only where there added more diameters but also the thicknesses where optimized for each
The CDMAM phantom is a result of the project: “Quality Assurance in Mammography, at the Department of Diagnostic Radiology, University Medical Centre Nijmegen, St. Radboud”. It was designed by M.A.O. Thijssen, K.R. Bijkerk and J.M. Lindeijer, in cooperation with the Medical Instrumental Department. Molybdenum anode, 30 µm Molybdenum filtration, at 28 kV.
© 2015 - Artinis Medical Systems BV. - Version 2
4
Manual CDMAM 4.0
individual diameter. The number of gold discs increased from 410 to 672 with improved tolerance specifications based on highly accurate production processes. In the CDMAM 4.0 version, each diameter has its own gold disk thicknesses. The thicknesses are optimized per diameter to the most interesting values per diameter according to the European guidelines for Quality Assurance in Mammography. The gold disks diameter range is from 0.08 to 2.00 in 21 steps. The thicknesses are between 0.012 m – 0.105 m for the 2.00 mm diameter gold discs to 0.590 m – 2.800 m for the 0.08 mm diameter gold discs, all in 16 steps. The phantom has been designed that about half of the disks will be detected by an experienced observer, when state-of-the-art mammography equipment is used with normal clinical settings. The phantom design is kept as close as possible to the CDMAM 3.4 for ease of use of the phantom and the translation ratio from automatic to human observer scores. The PMMA plates delivered by Artinis for simulation of breast thickness have a thickness accuracy of ±1%. Please, acknowledge Artinis Medical System B.V. in all published and/or presented work involving the phantom.
Figure 1. The CDMAM 4.0 (Contrast detail Mammography) phantom.
© 2015 - Artinis Medical Systems BV. - Version 2
5
Manual CDMAM 4.0
The CDMAM phantom, shown in Figure 1, consists of an aluminium base with gold-disks of various thickness and diameter, which is attached to a PMMA cover. It is recommended to use the phantom with PMMA plates to simulate the breast thickness. Usually the CDMAM phantom and 4 PMMA plates with a thickness of 10 mm are used to simulate a breast thickness of 5 cm. The dimensions of phantom and PMMA plates match the standard mammography size (180 x 240 mm). It is strongly recommended to use plates with a thickness precision of ± 1% to minimize attenuation differences, please contact your local distributor or Artinis Medical System for delivery information of such PMMA plates. For use with the automatic readout software CDCOM only use PMMA plates without any markers.
The aluminium base consists of Al 1050 (99.5% pure aluminium) and has a thickness of 0.5 mm. The base has been polished and was anodized black. Gold-disks (99.99 % pure gold) of variable thickness and diameter have been attached to the base by means of a vaporization technique. The disks are arranged in a matrix of 16 rows and 21 columns (Table 2). Within a column the disk-diameter is constant, with an exponential increasing thickness. The thickness steps were chosen largest in the high contrast part of the phantom, due to the nature of contrast detail curves of X-ray systems (Rose, 1973). During vaporization of the gold disks, the aluminium plate is covered with a mask with lasered holes of the desired diameter. Due to small irregularities of the mask and specific deviations related to the vaporization process, the disks can deviate slightly from the intended diameter, and might not be exactly circular. Also the thickness of the disks may deviate from the indicated values, mainly due to small variations in the gold vaporization profile. The disc diameters are logarithmic linear increasing from 0.08 mm to 2.0 mm. The diameter tolerance is ± 0.005 um (Mean ± 2SD).
© 2015 - Artinis Medical Systems BV. - Version 2
6
Manual CDMAM 4.0
The phantom contains 16 optimized thicknesses per diameter. For the 2.0 mm diameter the gold disc thickness varies between 0.012 m and 0.105 m. For the 0.08 mm diameter the gold discs thickness varies between 0.590 m and 2.800 m (Table 2). The thicknesses are optimized to be used in conjunction with the European Guidelines and for enlarging the number of discs which can be used for the fit of the psychometric curve. The thickness tolerance is ± 3% (mean ± 2 SD). The phantom is divided into four blocks for production purposes and inspection and to create space needed for reference discs. These reference discs are placed on the sides and between the blocks and used by CDCOM readout software to accurately localize the position of the gold discs. Each square contains two identical disks, one in the centre and one in a randomly chosen corner, to allow verification of the detection of each object. Easily memorisable patterns have been avoided for human image readout.
The cover consists of a 5 mm thick PMMA plate. In the cover a cavity with a depth of 2 mm has been milled, to accommodate the aluminium base with the gold-disks. The assembly ( 3 mm PMMA + 0.5 mm aluminium) has a PMMA equivalent thickness of 10 mm, under standard mammography-exposure conditions. Between the PMMA cover and the aluminium base a silkscreen printed transparent with X-ray contrasting paint has been placed, showing a grid, the disk thickness and disk diameter. Consequently the X-ray image will show a number of squares arranged in 21 columns and 16 rows, with the disk-diameter shown above most columns. The grid is not rotated for a more efficient use of the phantom area and to minimize effect of the heel effect on the CD Curve.
The CDMAM-phantom is delivered with 4 PMMA plates, for the simulation of different breast-thicknesses. The plates have a thickness of 10 mm each and are otherwise of the same dimensions as the phantom.
© 2015 - Artinis Medical Systems BV. - Version 2
7
Manual CDMAM 4.0
Under standard mammography-exposure conditions (Mo-anode, 30 m Mo-filtration, 28 kV), the phantom has a PMMA-equivalent thickness of 10 mm. These conditions will result in an HVL at the image plane of about 0.65 mm Al, when four PMMA plates are added. This indicates an effective energy of about 19 keV, according to Table 1. The linear attenuation coefficient of gold g = 175 mm-1 at this energy (Hubbell, 1982).
Energy [keV]
/ [cm2/g]
HVL [mm Al]
10
25.82
0.099
15
7.836
0.327
20
3.392
0.765
30
1.115
2.300
Table 1. The energy of monochromatic radiation with the mass attenuation coefficient / and the HVL of aluminium. The radiation contrast Cr of the gold disks can be calculated with the formula: Cr lblg 1 e gxg lb
where Ig and Ib represent the X-ray intensities with and without a gold disk and xg the thickness of a gold disk. In table 3 the radiation contrast of each gold disk is given for standard mammography exposure conditions.
© 2015 - Artinis Medical Systems BV. - Version 2
8
Disc Diameter (mm) 2
1.7
1.4
1.2
1
0.88
0.77
0.66
0.57
0.5
0.42
0.35
0.3
0.25
0.21
0.18
0.15
0.13
0.1
0.09
0.08
16
0.103
0.105
0.106
0.106
0.129
0.147
0.168
0.191
0.208
0.240
0.264
0.341
0.446
0.549
0.675
0.849
1.195
1.396
2.434
2.633
2.800
15
0.094
0.096
0.098
0.097
0.109
0.128
0.148
0.170
0.192
0.210
0.239
0.308
0.401
0.493
0.601
0.763
1.105
1.310
2.228
2.411
2.601
14
0.087
0.089
0.090
0.090
0.099
0.109
0.130
0.149
0.171
0.193
0.208
0.269
0.351
0.448
0.554
0.681
0.964
1.205
2.019
2.179
2.380
13
0.078
0.078
0.079
0.079
0.091
0.100
0.110
0.130
0.150
0.171
0.191
0.242
0.312
0.401
0.493
0.606
0.865
1.111
1.854
2.000
2.180
12
0.069
0.069
0.071
0.071
0.080
0.091
0.100
0.110
0.130
0.150
0.169
0.209
0.272
0.350
0.442
0.554
0.777
0.970
1.680
1.830
1.958
11
0.056
0.056
0.057
0.057
0.071
0.080
0.092
0.100
0.110
0.130
0.148
0.192
0.243
0.310
0.396
0.490
0.690
0.860
1.560
1.655
1.787
10
0.048
0.048
0.049
0.049
0.057
0.071
0.080
0.091
0.100
0.109
0.129
0.170
0.210
0.270
0.344
0.440
0.610
0.770
1.440
1.531
1.623
9
0.042
0.042
0.043
0.043
0.049
0.057
0.071
0.080
0.091
0.100
0.108
0.147
0.190
0.240
0.304
0.390
0.550
0.678
1.330
1.415
1.508
8
0.037
0.037
0.038
0.038
0.043
0.049
0.057
0.070
0.080
0.090
0.098
0.127
0.167
0.206
0.264
0.338
0.485
0.598
1.210
1.304
1.386
7
0.032
0.032
0.033
0.033
0.038
0.043
0.049
0.057
0.070
0.079
0.089
0.107
0.144
0.186
0.234
0.297
0.429
0.538
1.100
1.184
1.273
6
0.027
0.027
0.028
0.028
0.033
0.038
0.043
0.049
0.056
0.070
0.077
0.096
0.125
0.163
0.200
0.256
0.379
0.467
0.947
1.081
1.165
5
0.023
0.024
0.024
0.024
0.028
0.033
0.038
0.042
0.048
0.056
0.069
0.088
0.104
0.142
0.181
0.227
0.327
0.416
0.833
0.928
1.047
4
0.020
0.021
0.021
0.021
0.024
0.028
0.032
0.037
0.042
0.048
0.055
0.076
0.093
0.121
0.158
0.194
0.285
0.366
0.731
0.810
0.890
3
0.017
0.018
0.018
0.018
0.021
0.024
0.027
0.032
0.037
0.041
0.047
0.067
0.085
0.101
0.136
0.173
0.245
0.313
0.640
0.716
0.782
2
0.015
0.015
0.015
0.015
0.018
0.021
0.024
0.027
0.032
0.036
0.040
0.053
0.073
0.090
0.116
0.150
0.216
0.273
0.557
0.617
0.684
1
0.012
0.012
0.012
0.012
0.015
0.018
0.021
0.023
0.027
0.031
0.035
0.045
0.064
0.081
0.096
0.129
0.165
0.231
0.493
0.535
0.590
Table 2. Thickness [μm] of the gold disks within the phantom for each row and diameter(Ø) [mm].
© 2015 - Artinis Medical Systems BV. - Version 2.1
Disc Diameter (mm) 2
1.7
1.4
1.2
1
0.88
0.77
0.66
0.57
0.5
0.42
0.35
0.3
0.25
0.21
0.18
0.15
0.13
0.1
0.09
0.08
16
1.79%
1.82%
1.84%
1.84%
2.23%
2.54%
2.90%
3.29%
3.57%
4.11%
4.51%
5.79%
7.51%
9.16%
11.14%
13.81%
18.87%
21.67%
34.69%
36.92%
38.74%
15
1.63%
1.67%
1.70%
1.68%
1.89%
2.22%
2.56%
2.93%
3.30%
3.61%
4.10%
5.25%
6.78%
8.27%
9.98%
12.50%
17.58%
20.49%
32.29%
34.42%
36.57%
14
1.51%
1.55%
1.56%
1.56%
1.72%
1.89%
2.25%
2.57%
2.95%
3.32%
3.57%
4.60%
5.96%
7.54%
9.24%
11.23%
15.52%
19.01%
29.77%
31.70%
34.06%
13
1.36%
1.36%
1.37%
1.37%
1.58%
1.73%
1.91%
2.25%
2.59%
2.95%
3.29%
4.15%
5.31%
6.78%
8.27%
10.06%
14.05%
17.67%
27.71%
29.53%
31.72%
12
1.20%
1.20%
1.23%
1.23%
1.39%
1.58%
1.73%
1.91%
2.25%
2.59%
2.91%
3.59%
4.65%
5.94%
7.44%
9.24%
12.71%
15.61%
25.47%
27.40%
29.01%
11
0.98%
0.98%
0.99%
0.99%
1.23%
1.39%
1.60%
1.73%
1.91%
2.25%
2.56%
3.30%
4.16%
5.28%
6.70%
8.22%
11.37%
13.97%
23.89%
25.15%
26.85%
10
0.84%
0.84%
0.85%
0.85%
0.99%
1.23%
1.39%
1.58%
1.73%
1.89%
2.23%
2.93%
3.61%
4.62%
5.84%
7.41%
10.12%
12.61%
22.28%
23.50%
24.73%
9
0.73%
0.73%
0.75%
0.75%
0.85%
0.99%
1.23%
1.39%
1.58%
1.73%
1.87%
2.54%
3.27%
4.11%
5.18%
6.60%
9.18%
11.19%
20.76%
21.93%
23.19%
8
0.65%
0.65%
0.66%
0.66%
0.75%
0.85%
0.99%
1.22%
1.39%
1.56%
1.70%
2.20%
2.88%
3.54%
4.51%
5.74%
8.14%
9.94%
19.08%
20.40%
21.54%
7
0.56%
0.56%
0.58%
0.58%
0.66%
0.75%
0.85%
0.99%
1.22%
1.37%
1.55%
1.86%
2.49%
3.20%
4.01%
5.06%
7.23%
8.99%
17.65%
18.71%
19.97%
6
0.47%
0.47%
0.49%
0.49%
0.58%
0.66%
0.75%
0.85%
0.98%
1.22%
1.34%
1.67%
2.16%
2.81%
3.44%
4.38%
6.42%
7.85%
15.27%
17.24%
18.44%
5
0.40%
0.42%
0.42%
0.42%
0.49%
0.58%
0.66%
0.73%
0.84%
0.98%
1.20%
1.53%
1.80%
2.45%
3.12%
3.89%
5.56%
7.02%
13.56%
14.99%
16.74%
4
0.35%
0.37%
0.37%
0.37%
0.42%
0.49%
0.56%
0.65%
0.73%
0.84%
0.96%
1.32%
1.61%
2.10%
2.73%
3.34%
4.87%
6.20%
12.01%
13.22%
14.42%
3
0.30%
0.31%
0.31%
0.31%
0.37%
0.42%
0.47%
0.56%
0.65%
0.71%
0.82%
1.17%
1.48%
1.75%
2.35%
2.98%
4.20%
5.33%
10.60%
11.78%
12.79%
2
0.26%
0.26%
0.26%
0.26%
0.31%
0.37%
0.42%
0.47%
0.56%
0.63%
0.70%
0.92%
1.27%
1.56%
2.01%
2.59%
3.71%
4.67%
9.29%
10.23%
11.28%
1
0.21%
0.21%
0.21%
0.21%
0.26%
0.31%
0.37%
0.40%
0.47%
0.54%
0.61%
0.78%
1.11%
1.41%
1.67%
2.23%
2.85%
3.96%
8.27%
8.94%
9.81%
Table 3. The radiation contrast Cr, calculated with formula 1, for standard mammography exposure conditions
© 2015 - Artinis Medical Systems BV. - Version 2
10
Manual CDMAM 4.0
To make an X-ray image, the CDMAM phantom should be positioned on the bucky with the thinnest disk-thicknesses at the thorax side, in combination with one or more PMMA plates (Figure 2). For use with the automatic readout software CDCOM only use PMMA plates without any markers. X-Ray apparatus
Figure 2. Reference configuration of the CDMAM phantom on the bucky of a mammography unit.
© 2015 - Artinis Medical Systems BV. - Version 2
11
Manual CDMAM 4.0
Store and use the phantom and its belongings at room temperature (15o-25o) and at normal humidity protected for fluid and moisture, dust, etc. preferable in the delivered case. Handle all products with care. PMMA scratches easily which might give that the phantom or plates is useless for evaluation. Clean the materials with non-aggressive general cleaner. The phantom cannot be used in MRI systems or in the neighbourhood of other magnetic materials.
Send back to manufacturer
The Declaration of conformity to the specifications states that the CDMAM 4.0 phantom is in compliance to the norms and product description. By heavy wearing out, misuse or mistreatment of the phantom, the phantom might become unusable or unreliable. In case one of the following symptoms is noticed, please contact your local distributor or Artinis Medical Systems. Large scratches and other damage to the body Large scratches and other damage to one of the PMMA Attenuation plates Dents, large scratches or other damage into the aluminium plate The aluminium plate gets loose from the body The grid paint is dissolved by moisture which is visible by high contrast particles at the Xray images and little white speckles between the foil and the body
© 2015 - Artinis Medical Systems BV. - Version 2
12
Manual CDMAM 4.0
CDMAM images can be evaluated at the classical way with images and human observer or in a digital environment by using automatic CDMAM Analyser scoring software. The Xray image of the CDMAM phantom needs to be evaluated by at least 3 experienced observers. The "Score form CDMAM phantom" (see Appendix 1) can be used for this purpose. The image should be evaluated in the area where the gold-disks are just visible, by indication of the location of the eccentric disks. At least 3 fields must be observed in each column and each row, in order to comply with the correction scheme, which is described in paragraph 4.1. The indicated positions of the eccentric disks have to be compared to the true diskpositions in the phantom, for which the "Evaluation form CDMAM phantom" can be used (see Appendix 2). To evaluate the observations certain rules have to be applied, taking into account the 4 nearest neighbours (defined by a common vertice) of the field under examination. The evaluation of a particular field must refer to the original observations for the nearest neighbours. Examples of the correction scheme are discussed in paragraph 4.2.
In the correction scheme, there are three possibilities for each observation: T: the eccentric disk was indicated at the true position F: the eccentric disk was indicated at a false position N: the eccentric disk was not indicated at all The two main rules within the correction scheme are: A True needs 2 or more correctly indicated nearest neighbours to remain a True. A False or Not indicated disk will be considered as True when it has 3 or 4 correctly indicated nearest neighbours. Exceptions on the two main rules are: A True which has only 2 nearest neighbours (at the edges of the phantom) needs only 1 correctly indicated nearest neighbour to remain True. A False or Not indicated disk which has only 2 nearest neighbours will be regarded True if both nearest neighbours are correctly indicated.
© 2015 - Artinis Medical Systems BV. - Version 2
13
Manual CDMAM 4.0
Six examples (shown in Figure 3) of the correction scheme are discussed below. Example 1: The common situation. T* remains T because of its 2 correctly indicated nearest neighbours. F* remains F because it has only 2 correctly indicated nearest neighbours. Example 2: F* is considered T because it has more than 2 correctly indicated nearest neighbours. Both T*'s however have only 1 correctly identified nearest neighbour, and thus are considered to be F's. Example 3: T* remains T because it has 1 out of 2 correctly indicated nearest neighbours. The absent corner will be considered as an F because it has only 1 correctly indicated nearest neighbour. Example 4: F* will be considered as a T because of its 2 out of 2 correctly indicated nearest neighbours. T* will be considered as an F because it has only 1 correctly indicated nearest neighbour. The absent corner will be considered as an F, because it has no correctly indicated nearest neighbours. Example 5: F* remains an F, because it has only 1 out of 2 correctly indicated nearest neighbours. Both T*'s are considered as F's because they have none respectively 1 correctly indicated nearest neighbour. Example 6: T* remains T because it has 1 out of 2 correctly indicated nearest neighbours. F* will be considered as a T because of 3 correctly indicated nearest neighbours. The absent corner will be considered as a T because of 2 correctly indicated nearest neighbours.
© 2015 - Artinis Medical Systems BV. - Version 2
14
Manual CDMAM 4.0
T T T T N T* T T N F* T T N N N T
T T T T N T* F* T N N T* T N N N T
Figure 3.1
Figure 3.2
F T* T T N N T T N N N T N N N N
F F* T T N T* T T N N T T N N N T
Figure 3.3
Figure 3.4
F F* T* T N N N T N T* N T N N N N
T T* F* T T T T T N T T T N N T T
Figure 3.5
Figure 3.6
Figure 3. Six examples for the use of the correction scheme.
The results can be presented in a logarithmic graph, in which the disk-thickness is plotted against the disk-diameter. The curve through the threshold fields is called the contrast detail (CD) curve (Thijssen, 1993). The image quality can be expressed in a figure by calculation of the ratio of correctly identified disk-positions to the total number of squares:
𝑐𝑜𝑟𝑟𝑒𝑐𝑡 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑖𝑜 =
𝐶𝑜𝑟𝑟𝑒𝑐𝑡 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 ∙ 100% 𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑞𝑢𝑎𝑟𝑒𝑠
Another method to quantify image quality is called the Image Quality Figure (IQF)-method (Thijssen et al, 1989), which is defined by 16
𝐼𝑄𝐹 = ∑ 𝐶𝑖 ∙ 𝐷𝑖,𝑚𝑖𝑛 𝑖=1
where Di,min denotes the threshold diameter in contrast-column i. Summation over all contrast-columns yields the IQF. Image quality increases with an increasing number of correctly identified disk-positions. In this case the IQF will become smaller because the values of diameter and thickness of the threshold-disks are smaller.
© 2015 - Artinis Medical Systems BV. - Version 2
15
Manual CDMAM 4.0
An alternative method for calculation of the IQF has been presented (Thijssen et. al., 2000), giving a higher IQFinv at better image quality. Since for CDMAM phantoms the threshold thickness is determined per diameter, we suggest to calculate the IQF Inv according to
𝐼𝑄𝐹𝑖𝑛𝑣 =
100 ∑21 𝑖=1 𝑡𝑡ℎ𝑟,𝑖 ∙ 𝐷𝑖
where tthr,i is the threshold thickness of diameter i. Di (mm)
2
1.7
1.4
1.2
1
0.88
0.77
0.66
0.57
0.5
0.42
All invisible: tthr,i
0.113
0.116
0.118
0.118
0.140
0.167
0.195
0.218
0.235
0.265
0.295
All visible: tthr,i
0.010
0.010
0.010
0.011
0.013
0.016
0.018
0.020
0.023
0.027
0.030
Di (mm)
0.35
0.3
0.25
0.21
0.18
0.15
0.13
0.1
0.09
0.08
All invisible: tthr,i
0.384
0.505
0.608
0.744
0.951
1.349
1.513
2.662
2.890
3.061
All visible: tthr,i
0.038
0.057
0.071
0.083
0.114
0.144
0.200
0.433
0.468
0.518
Table 4. tthr,i for completely invisible columns or completely visible comlumns Since the phantom contains more diameters, the IQFinv of the CDMAM 4.0 cannot be compared to the IQFinv of the CDMAM 3.4. For comparing results between phantoms we recommend to use the contrast threshold values of the common diameters of both phantoms or comparing CD Curves.
© 2015 - Artinis Medical Systems BV. - Version 2
16
Manual CDMAM 4.0
The CDMAM Analyser software enables fast automatic readout, scoring and analysis of digital CDMAM 4.0 images. The program uses CDCOM for CDMAM 4.0 for image readout and scoring and analyses the scores according to the described analysis method in the supplement to the European Guidelines fourth edition (Perry et al., 2013). The analysis results are observer independent. As a prerequisite for reading this manual and using the program correctly the reader should also study the previous chapters on the CDMAM carefully.
For automatic readout and scoring of digital CDMAM images the software tool CDCOM for CDMAM 4.0 is used. The CDCOM software automatically identifies the gold discs on images of the CDMAM phantom and is available from the EUREF website [www.euref.org]. The program attempts to correctly locate the position of the gold discs by first identifying the grid position and reference dots and subsequently the gold discs positions. CDCOM performs the following actions: It reads a single digital CDMAM version 4.0 image and tries to detect the gold disks in the phantom. CDCOM determines in which corner of each cell the gold disc is most likely to be The program checks for each cell if the correct corner was chosen An additional detection step is done using the 3 corners where no marker is present and the centre dot More information regarding CDCOM can be found in the CDCOM Manual. The CDCOM executable can be found in the installation folder of this software (CDCOM4.exe). If EUREF releases a new version of CDCOM for CDMAM 4.0 you can download the executable from the EUREF website and replace the current by the new executable file. Please notice you have to give the new executable file the same name, so “CDCOM4.exe”.
The analysis method is according to the Supplement to the European Guidelines. All diameters are included during the analysis. The score results of the CDCOM software are
© 2015 - Artinis Medical Systems BV. - Version 2
17
Manual CDMAM 4.0
averaged and filtered by a Gaussian filter. The probability of gold disc detection by CDCOM is between 25% and 100% by the 4-alternative forced choice method. The gold disc thicknesses having 62.5% detection probability are the threshold thicknesses and are determined by applying psychometric curve fitting by a least squares procedure over the probabilities. The psychometric curve is fitted by (Veldkamp, 2003):
𝑝(𝑑) =
0.75 + 0.25 1 + 𝑒 −𝑓(𝐶−𝐶𝑡 )
C = logarithm of signal contrast C = log(1 − e−μd) Ct = signal contrast at the threshold of 62.5% f = fitting parameter p(d) = the probability of detection p(d) of an object with thickness d The attenuation coefficient of gold depends on the beam quality used for the exposure. These values were calculated by DR Dance and KC Young at NCCPM with 3mm PMMA representing the compression paddle, using spectra from Boone (et al., 1997) and attenuation coefficients for materials in the test objects (aluminium, gold, PMMA) from Berger (et al, 2005). In the software these threshold thicknesses are called the Automatic Threshold Gold Thicknesses. The Automatic Threshold Gold Thicknesses are converted to Predicted Human Gold Thicknesses by correction for human read out by 𝑇𝑝𝑟𝑒𝑑𝑖𝑐𝑡𝑒𝑑 = 𝑎𝑇𝑎𝑢𝑡𝑜 𝑛 Tpredicted =predicted human threshold gold thickness Tauto = Automatic Threshold Gold thickness a, n = fitted parameters with a = 1.441, n = 0.895 The Fit to Predicted Human Gold Thickness are determined by fitting a third order polynomial through the Predicted Human Gold Thicknesses resulting in the CD Curve by
𝑇𝑓𝑖𝑡 = 𝑎 +
𝑏 𝑐 𝑑 + 2+ 3 𝑥 𝑥 𝑥
Tfit = Fit to Predicted Human Gold Thickness (um) x = detail diameter (mm)
© 2015 - Artinis Medical Systems BV. - Version 2
18
Manual CDMAM 4.0
a, b, c and d = coefficients adjusted to achieve a least squares fit, and are ≥ 0 The IQFinv number is computed according to
𝐼𝑄𝐹𝑖𝑛𝑣 =
100 ∑21 𝑖=1 𝑡𝑖,𝑡ℎ𝑟 ∙ 𝐷𝑖
where ti,thr denotes the threshold thickness (Fit to Predicted Human Gold Thickness) in diameter-column i. The thickness is given in µm whereas the diameter is taken in mm. The total detected (%) is determined by
𝑇𝑜𝑡𝑎𝑙 𝑑𝑒𝑡𝑒𝑐𝑡𝑒𝑑 (%) =
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑑𝑒𝑡𝑒𝑐𝑡𝑒𝑑 𝑔𝑜𝑙𝑑 𝑑𝑖𝑠𝑐𝑠 × 100% 𝑡𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑔𝑜𝑙𝑑 𝑑𝑖𝑠𝑐𝑠
© 2015 - Artinis Medical Systems BV. - Version 2
19
Manual CDMAM 4.0
The data and images in the program are organized in projects with groups (Figure 4). The project may contain several groups, each having by example other settings during imaging of the phantom, allowing easy comparison of the influence of the settings to the image quality. A group should contain multiple images made with the same settings. We recommend to use at least 8 images, but preferable 16 or more per group for reliability of the results. The readout and score result can be viewed per project, group or image. Analysed images are marked with a green sign. In case the analysis failed, the image is marked with a yellow question mark, mostly caused by wrong or incomplete DICOM header information (see chapter 6.3).
Figure 4. Program structure
A new project can be made via the menu, File, New project. Existing projects can be opened via menu, File, Open project or via the toolbar opening button. The project can be saved via menu, File, Save project (as) or via the toolbar saving button, which also changes the project name from the standard “new project” to the project file name. Latest projects can be opened via menu, File. The project has properties, which can be viewed via menu, Project, properties or via the right mouse button at the project. The first tab offers to add comments for project description and the second to use relative file paths or changing the setting if DICOM files are opened or not by addition to a project. The relative file paths save the image file paths relatively to the project file path, which is easy for copying project with images from a computer to another computer, USB or CD by example.
© 2015 - Artinis Medical Systems BV. - Version 2
20