Following approval from local ethical committee,
thirteen (n=13) cadaver heads were scanned with a first generation 64 detector dual tube CT scanner (Definition,
Siemens Medical Systems; Erlangen,
Germany).
Scan range was determined to be from the upper border of frontal sinuses to the inferior border of maxillary alveolar process.
Each head was scanned 10 times with following parameters (Table 1) (Figure 1)
Scan No |
kV |
Tube Current (mAs) |
pitch |
1 |
120 |
80 |
1.2 |
2 |
120 |
80 |
3 |
3 |
120 |
50 |
3 |
4 |
120 |
20 |
3 |
5 |
120 |
CARE 50* (Average 9.15) |
3 |
6 |
100 |
50 |
3 |
7 |
100 |
20 |
3 |
8 |
100 |
CARE 50* (Average 8.90) |
3 |
9 |
80 |
50 |
3 |
10 |
80 |
CARE 50* (Average 9.40) |
3 |
Table 1: Scanning Parameters.
*:With automatic tube current modulation referance Current:50 mAs
All images were evaluated by two expert radiologists independently by double blind technique.
Evaluation and statistical analysis consisted of 4 steps:
STEP 1: The overall image quality of all images were scored according to a subjective scale (Table 2)
Score |
Meaning |
1 |
Unacceptable image quality |
2 |
Bad quality/too much artifacts |
3 |
Medium quality /some artifacts |
4 |
Good quality/ almost no artifacts |
5 |
Very good quality /no artifacts |
Table 2: Subjective scoring system used for quantification of image quality of all scans.
STEP 2: 20 osseous structures (Figure 2) were clustered into 4 groups and evaluated by two expert radiologists (Table 3)
Group No:1
Gross Structures
|
Zygomatic Arch |
Medial Pterygoid Plate |
Lateral Pterygoid Plate |
Mandibular Condyle |
Crista Galli |
Styloid Process |
Group No:2
Paranasal Sinuses
|
Posterior Wall Of Maxillary Sinus |
Base of orbita |
Lamina paprisea |
Fovea ethmoidalis |
Medial lamella of cribriform plate |
Group No:3
Nasal Cavity
|
Nasolacrimal Canal |
Nasal septum |
Nasal Bone |
Middle nasal concha |
Group No:4
Temporal Bone
|
Mastoid Air Cells |
Incudomalleolar Joint |
Cohlea Basal Turn |
Facial Canal |
Internal Acustic Meatus |
Table 3.
Osseous structures used to evaluate image quality
All structures were scored according to both a qualitative and a quantitative scale.
For qualitative image quality scoring each osseous structure was scored acording to Table 4
Score |
Meaning |
0 |
Osseous structure can not be demonstrated |
1 |
Osseous structure is demonstrated with insufficient image quality |
2 |
Osseous structure is demonstrated with sufficient image quality |
Table 4: Qualitative scoring system used to evaluate 20 osseous structure
For quantitave image quality scoring two Hounsfield Unit measurements were performed on similar areas from air,
bone and soft tissue. Attenuation values were used as signal values where the standard deviation was used as noise values (Figure 3)
Step 3: Radiation Dose
CTDIvol and DLP for all of the protocols were collected.
The standard protocol (pitch 1.2),
used for paranasal imaging was designed for head imaging scanner reported radiation doses based on a 16 cm diameter acrylic phantom.
As high pitch mode,
primarily designed for cardiac imaging,
scanner reported high pitch (3) protocols radiation doses based on a 32 cm diameter acrylic phantom which causes falsely higher dose reports on paranasal CT scans.
All CTDIvol and DLP values were converted to 16 cm diameter values by using conversion factors based on AAPM report no:204. Conversion was made with multiplying
120 kv protocols by the factor of 2,3
100 kv protocols by the factor of 2,4
80 kv protocols by the factor of 2,5
The conversion factors (coefficients) were taken from American Association of Physicists in Medicine (AAPM) report number 204
All 16 cm phantom DLP values were converted to effective dose by multiplying with a conversion factor (0.0021 mSv/mGy.cm ) taken from AAPM report 204
Step 4: Results
A)Results (Image quality,
qualitative)
A.1) Gross structures group:
- 80 Kv Care 50 protocol the image quality was scored below 1.5 (bad diagnostic quality)
A.2) PNS structures group:
- All structures except fovea ethmoidalis had scored lower than 1 on 80 Kv Care 50 (bad quality)
- None of CARE 50 protocols have diagnostic score higher than 1.5
A.3)Temporal bone group:
- Minimum mAs value of 50 should be used for optimal visualization of the facial canal.
(score >1.5)
- All other structures except facial canal were well demonstrated with all the protocols except 80 Kv CARE 50 protocol
A.4)Nasal cavity group:
- Middle concha could not be visualized with optimum quality in any of the protocols utilized in this study.
We consider that this is due to the use of cadaver heads .
- All other structures except for the middle concha received a diagnostic score of higher than 1.5
B)Results (Image quality,
overall) (Figure 4)
- There was no statistically significant difference (p=0,135) between the standard protocol,
120 kV/80 mAs,120 kV/50 mAs protocols
- 100 kV/50 mAs,
80 kV/50 mAs,120 kV/20 mAs,
120 kV CARE 50 protocols had good image quality (average score 3,5-4,1) .
And there was no statistically significant difference (p>0.5) between theese protocols.
- In 50 mAs group there was no statistically significant difference between 120 kV and 100 Kv (p=0,424)
- 80 kV CARE 50 and 100 Kv CARE 50 protocols were not sufficient for diagnostic purposes ,
100 kV 20 mAs scored higher than 3 but stastistically none of them were different (p=0,442)
C)Results (Image quality,
quantitative,
SNR) (Figure 5,6)
Regarding SNR:
- With the exclusion of 80 kV CARE 50 there was no significant difference in bone and soft tissue groups
- No significant differance in 50 mAs group
- 80 Kv group had significantly LESS SNR
D)Results (Radiation dose) (Figure 7)
Against long odds,
high pitch 120 kV 80 mAs group’s ED was higher than standart pitch 120 kV 80 mAs group
But high pitch 120 kV 50 mAs group’s ED was nearly half of standart pitch 120 kV 80 mAs group
Statistically insignificant EDs in 20 mAs group p=0.20
E)Results (Summary)
- In this study 1st generation 64 detector dual tube CT was used
- Best diagnostic quality was achieved with standard pitch 120 kV/80 mAs protocol
- High pitch 120 kV/80 mAs,
120 kV/50 mAs protocols achieved similar diagnostic quality with standard protocol
- 120 Kv/50 mAs protocols ED was %40 lower than the standard protocol
- Protocols with tube current of 20-50 mAs may provide enough diagnostic quality but should be used carefully as small osseous structures like temporal bone and soft tissues couldn't be visualized well
- In our study most of thin osseous structures could be well visualized with a tube current of 50 mAs.
Higher tube currents does not improve quality significantly,
but raises effective radiation doses.
Although higher voltage yields better diagnostic quality there is no significant difference between 120,
100 or 80 kV when a tube current of 50 mAs is used