Influence of Volume and Density of Computerized Tomography Images in Tube Current Automatic Exposure Control

 

Dong-Oh Kim¹, Cheong-Hwan Lim¹*, Hong-Ryang Jung¹, Ji-Hoon Choi², Jun-Seok Ha³

¹Dept. of Radiological Science, Hanseo University, Korea

²Dept. of Health Care, Hanseo University, Korea

³Dept. of Diagnostic Radiology Doungguk University Medical Center, Korea

 

ABSTRACT:

Background/Objectives: As computerized tomography (CT) is more frequently used, there is a growing interest about the patient dose. To administer an appropriate dose during examinations, most equipment use an automatic exposure control (AEC) system. This study aimed to affirm that volume and density are relevant factor of AEC which is the correct method for dose selection.

Methods/Statistical analysis: With regard to CT equipment, the MX-16 slice technology was used. A large phantom (35×35×10) and a small phantom (30×30×10) were produced using acrylic material. The phantom solution’s density was adjusted by diluting water with contrast medium. The study compared dose and image quality by volume and density, by two phantoms with different volume and density in identical penetrance.

Findings: With the dose increase, the noise decreased and the signal-to-noise ratio (SNR) rose. When the density increased, both dose and the noise increased. With the volume increased, both dose and noise increased. As for the measurement of phantoms with different volume and density in identical penetrance, dose of large phantom was 139mAs/slice, noise was 36.41HU while dose of small phantom was 104.3mAs/slice, noise was 38.83HU. To match with noise value in phantom with different volume and density, the small phantom showed 16mAs/slice increase.

Improvements/Applications: In an examination that applied automatic exposure control, the system selected another dose in case of identical penetrance with different volume and density levels. The image quality could not be maintained at a certain level. When a dose is chosen by an AEC system, it is appropriate to consider volume and density as factors to calculate the transmission dose.

 

 

 

 

1. INTRODUCTION:

A computerized tomography (CT) has an excellent resolution power and provides much diagnostic information. Due to such advantages, the number of CT examinations is growing every year.

 

However, since the exposure dose of CT is comparatively higher than of other radiographic means, it is important to obtain an optimal image while examining the appropriate dose to the patient.

 

To calculate appropriate doses, information about the patient is required, such as height, weight, BMI, chest/waist measurement¹, but this information collection is insufficient as the scan process problems. In an environment where the effectiveness of automatic exposure control (AEC) is demonstrated^2,3, radiological techologists depend on the CT equipment’s AEC function.

 

Various types of equipment differ by manufacturer, but the AEC function basically calculates doses by examining the energy penetrated in tomography, compares a standard phantom with the patient’s transmission dose to select a total dose, and then alters the dose according to examined part⁴.

 

It is the purpose of AEC system to find out appropriate dose for each patient to create similar quality image in different situations^5,6. However, when the quality of images differs, it is hard to tell whether the AEC system chose the appropriate dose.

 

This study formulated and verified the hypothesis that even with same penetrance, a patient’s different volume and density can cause differences in image quality. This compensates drawbacks of current AEC systems and provides basic material for the improvement of equipment in the future.

 

2. MATERIALS AND METHODS:

This study used 16-channel equipment from P brand, the MX 16-slice. Transparent acrylic material was used to make a large phantom (35×35×10 cm) and small phantom (30×30×10 cm). For the phantom solution, water was used as well as diluted contrast medium to adjust density.

 

2.1. Study method:

This study chose a protocol matched with the phantom size and used the large chest factory protocol of MX-slice 16 from P brand [Table 1].

 

Table 1. Experiment environment

	Scout	AEC Mode	Manual Mode
Lengh(mm)	300	300	300
Thikness(mm)	3	3	3
Increment(mm)	3	3	3
kVp	120	140	140
AEC	Off	On	Off
Collimation(mm)		16×1.5	16×1.5
Pitch		0.8631	0.8631
Rotation Time		1	1
FOV(mm)		300	300
Matrix		512	512
WW/WL		350/40	350/40

 

The image quality was compared with the images obtained by experiments with three parameters of density, volume, and permeability.

 

Experiments about the AEC mode and the manual mode were conducted. In the manual mode, mAs/slice was altered and other conditions were identical to those of AEC mode experiments.

 

2.1.1. Dose change:

Using large and small phantoms with the same penetrance, the dose was altered to 70 mAs, 100 mAs, 130 mAs, 160 mAs, and 190 mAs in manual mode to obtain images.

 

2.1.2. Density change:

Using the large phantom, the CT number was altered to 1HU, 50HU, 100HU, and 150 HU to obtain images.

 

2.1.3. Volume change:

Using large and small phantoms with different volumes, water with an identical CT number was put to obtain images.

 

2.1.4. Volume and density change in identical penetrance:

The large phantom was filled with water for a scout scan, and the CT number twice examined at the center of the scout image to get penetrance (HU). Penetrance (HU) of the small phantom to the large phantom was matched by diluting the water with contrast medium to obtain the scout image, and measured the CT number twice at the center. Two phantom images were obtained by the AEC.

 

2.1.5. Dose to achieve the same image quality:

Result of number 4(Volume and density change in identical penetrance), noises measured in each phantom were confirmed and the changed dose repeatedly measured to match noise values of the small and the large phantoms.

 

2.2 Measuring location:

The region of interest (ROI) in the form of an oval was drawn for measurement and its size was within 500mm²±10 mm² [Figure 1].

 

The ROI location was evenly measured from the center of the image four sides and in order to minimize error in measurement, it was measured more than three times with the average value used.

 

2.3 Analysis:

The average and a standard deviation of CT number were measured by software built in the P brand equipment and comparative analyses were conducted according to examination condition and phantom size.

 

A phantom with a different density also had a different CT number, so the image quality was compared to the noise, excluding the SNR.

 

When the examination conditions were different with an identical phantom, the noise and CT number were examined and then compared with the SNR.

 

 

Figure 1. Image quality measurement position

 

The CT number and noise from the phantom’s four sides were measured to obtain an average and the noise used for the standard deviation.

 

As for the SNR, the value divided by the ROI’s standard deviation was used as the ROI’s CT number ROI (Formula 1).

        ROI’s CT number

SNR = ------------------------           (1)

ROI Noise

 

3. RESULTS:

3.1 Image quality comparison by dose:

The image quality change by the dose change was assessed for the manual mode. Noise and the SNR were measured to analyze image quality.

As dose of small and large phantom increased, the noise value gradually decreased. When the dose of the small and large phantom increased, the SNR rose gradually[Table 2].

 

Table 2. Comparison of image quality according to dose change

	Small		Large
Dose	Noise	SNR	Noise	SNR
70(mAs/Slice)	47.88	2.773045	54.59	0.05394
100(mAs/Slice)	37.40	3.664459	44.23	0.08116
130(mAs/Slice)	31.78	4.286152	39.85	0.07484
160(mAs/Slice)	28.07	4.8845	33.26	0.08856

 

3.2 Dose and image quality comparison by volume:

The dose change by the volume change was measured for the AEC mode.

 

The dose 83.6 mAs/slice in the small phantom and 139 mAs/slice in the large phantom. The noise 28.32 HU in the former and 36,41 HU in the latter, respectively. The SNR showed a small rise compared to the volume increase [Table 3].

 

Table 3. Comparison of dose and image quality by volume

Division	Small	Large
CT Number	1	1
Dose (mAs/Slice)	83.6	139
Noise	28.32	36.41
SNR	0.055	0.098

3.3 Dose and image quality comparison by density:

The dose change by the density change of phantoms was measured in AEC mode and the density change was based on the CT number. Regarding image quality, noise was measured for analysis. With increased density, the dose and noise proportionally grew [Table 4].

 

Table 4. Comparison of dose and image quality according to density

CT Number(HU)	Dose (mAs/Slice)	Noise(HU)
1 (Avg. -3.5)	139	36.41
50 (Avg. 51.8)	163	44.12
100 (Avg. 102.8)	179	50.34
150 (Avg. 155.1)	197.3	58.22

 

3.4 Dose and image quality comparison by identical penetrance:

In case of the phantom’s identical penetrance but a different volume and density, measurements were performed in AEC mode. The dose and noise determined in AEC mode were measured to conduct a comparative analysis of image quality.

 

The dose was 104.3 mAs/slice in the small phantom and 139.0 mAs/slice in the large phantom. The noise was 38,83 HU in the small phantom and 36.41 HU in the large phantom [Table 5].

 

Table 5. Comparison of dose and image quality according to the same permeability

Division	Small	Large
Permeability(HU)	1,746	1,760
CT Number(HU)	133.48	-3.58
Dose(mAs/Slive)	104.3	139.0
Noise(HU)	38.83	36.41

 

3.5 Dose for identical image quality:

In the comparisons and measurements of dose and image quality by identical penetrance (Table 5), the small phantom measurement value was set as Samll-1. As for the noise value of Small-2, the Small-1 was changed to manual mode and the increased dose and 36.29 HU were measured which is close to the noise value (Table 5) of the large phantom.

 

The measured dose of Small-1 was 104 mAs/slice and that of Samll-2, which has a similar noise value as the identical large phantom (Table 5) was measured as 120 mAs/slice.

 

When the volume and density were different with identical penetrance, the dose of the small phantom increased 16 mAs/slice to reach an identical image quality [Table 6].

 

Table 6. Dose to indicate the same image quality

Division	Small-1	Small-2	Large
Dose	104	120	139
Noise	38.83	36.29	36.01

 

4. DISCUSSION:

This study examined the dose by volume and density in AEC mode and confirmed that the dose increased by the rise in volume and by the increase of density in phantoms with an identical volume. The dose increased proportionally to volume and density. About the phantom with identical volume and density, as for the comparative analysis result of noise and SNR by dose, the noise decreased and the SNR increased by the dose increase.

 

Based on this, small and large phantoms with identical penetrance but different in volume and density were produced to conduct an experiment on the AEC mode. While the penetrance was identical, in image quality, the small phantom’s average was 2.4 HU higher in noise than the large phantom. In order to get a noise value for the small phantom identical to the large phantom’s, an additional 16mAs/slice is needed.

 

By combining three AEC methods such as dose modulation by the X, Y axes⁷, dose modulation by the Z axis, and dose modulation by patient size, the dose is selected based on different standards (noise index, reference noise, reference mAs) by each CT equipment manufacturer⁸.

 

CT must be conducted with a minimum dose to obtain a constant quality of images. The examined part can be as short as 10 cm to a maximum of 60 cm, so the slight dose difference per slice leads to large amount of dose differences as a whole^9,10.

 

This study confirmed that even for phantoms with identical penetrance, the dose chosen by each equipment’s standard can lead to image differences and in order to get the same quality image, a higher dose is required for a small volume and high density. After the equipment protocol is decided, the dose selection is based on images obtained in scout and penetrance calculation. When calculating penetrance, it is recommended to consider the volume and density as an influential factor.

The study was limited because only the image quality difference by volume and density of the phantoms with identical penetrance was verified, but exact numbers could not be provided.

 

5. CONCLUSION:

This study used CT equipment with 16 channels to produce phantoms with identical penetrance but different density and volume to acquire images and it evaluated their image quality and doses. For the evaluation, noise, the SNR, and doses were compared.

 

A rise in density and volume led to a dose increase but the image quality could not maintain a same level. Also, when the volume and density were different with identical transmission doses, the dose selected by the AEC system was different and the image quality did not maintain that same level.

 

Since the AEC mode’s dose selection standard is the transmission dose, it is considered that the volume and density are influential factors in AEC system improvement and can contribute to an appropriate transmission dose calculation.

 

6. REFERENCES:

1.        Hamaguchi N, CT Auto Exposure Control on Cerebral 3DCT Angiography: Dose Reduction and Optimized Image SD for Inspection Purposes, Japanese Journal of Radiological Technology,2010, 166(4),pp.313-321

2.        Muramatsu Y, Performance Evoluation for CT-AEC(CT Automatic Exposure Control)System, Journal of Japanese Society of Radiological Technology, 2007, 63(5), pp.534-545

3.        McCollough CH, Automatic exposure in control in CT: are we done yet? Radiology, 2005, 237(3), pp.755-761.

4.        McCollough CH, Bruesewitz MR, Kofler Jm, Jr, CT dose reduction and dose management tools: overview of available options, Radiographics, 2006, 26(2), pp.503-515.

5.        Kubo T, Lin PJP, Stiller W, Radiation dose reduction in chest CT: a review, American Journal of Roentgenology, 2008, 190(2), pp.335-378.

6.        Kalra MK, Naz N, Rizzo SMR, Blake MA, Computed Tomography Radiation Dose Optimization: Scannig Protocols and Clinical Applications of Automatic Exposure Control, Current Problem in diagnostic Radiology, 2005, 34(5), pp.171-181.

7.        KB Lee, WH Lee, JH Lee, GBOh, BR Lee, Dose Reduction and image quality assessment in MDCT using AEC (D-DOM & Z-DOM) and in-plane bismuth shielding, Radiation Protection Dosimetry, 2010,141(4), pp.162-167

8.        Mannudeep K. Kalra, Michael M. Macher, Ravi S. Kamath et al, Sixteen-Detector Row CT of Abdomen and Pelvis: Study for Optimization of Z-Axis Modulation Technique Performed in 153 Patients, Radiology, 2004, 233(1), pp.241-250.

9.        Van Der Molen A J, Veldkamp W J H, Geleijns J, 16-slice CT: achievable effective doses of common protocols in comparison with recent CT dose surveys, The British Journal of Radiology, 2007,80(952), pp.248-303.

10.     Metter FAJ, WiestPW, Loken, JA, Kelsey CA, CT scanning: Patterns of use and dose, Journal of radiological protection, 2011, 20(4), pp.353-362.

 

 

 

 

Accepted on 17.06.2017           © RJPT All right reserved