Performance comparison between three intraoral image receptors of different technology at a variety of tube potential, tube current and exposure time settings using a stepwedge phantom

doi:10.4236/health.2011.31011

Paper Menu >>

Journal Menu >>

Vol.3, No.1, 56-65 (2011)

doi:10.4236/health.2011.31011

Health

Performance comparison between three intraoral image

receptors of different technology at a variety of tube

potential, tube current and exposure time settings using

a stepwedge phantom

Elli Katsoni1*, Ioannis A. Tsalafoutas2, Panagiotis Gritzalis3, Evripidis Stefanou3,

Evangelos Georgiou1, Emmanuel Yakoumakis1

1Department of Medical Physics, Medical School, University of Athens, Athens, Greece;

2Medical Physics Department, Agios Savvas Hospital, Athens, Greece;

3Department of Oral Diagnosis and Radiology, Dental School, University of Athens, Athens, Greece;

*Corresponding Author: [email protected]

Received 22 November 2010; revise 30 November 2010; accepted 3 December 2010.

ABSTRACT

Purpose: To comparatively evaluate the per-

formance of three intraoral image receptors of

different technology when exposed to different

X-ray beam spectra, dose and dose rate levels

using a stepwedge phantom. Materials and

methods: The intraoral radiographic receptors

evaluated were: the Kodak Insight F speed class

film, the Kodak RVG 6000, and the Duerr Vis-

tascan Combi PSP system. A dental quality

control phantom made of Plexiglas, containing

an aluminium stepwedge with 12 steps and 7

holes drilled in each step was radiographed

using a dental X-ray unit offering a wide range

of tube potential, tube current and exposure

time settings. The visibility of the holes in the

images produced with each one of the three

receptors was assessed by three independent

observers. For each image the total image qual-

ity score (TS) was derived from the summation

of the number of visible holes in each step. The

numbers of perceptible holes in each experi-

mental condition (TSs) were statistically ana-

lyzed through use of analysis of variance. In-

traobserver and interobserver agreement was

also measured. Results: Vistascan exhibited the

most extended useful exposure range, follow ed

by RVG 6000 and Insight. RVG 6000 exhibited

the largest TS values in all tube potential set-

tings except 70 kV where the Vistascan per-

formed better. Insight performed better than

Vistascan only at 60 and 63 kV. Vistascan per-

formed better at 66 and 70 kV, Insight at 60 and

66 kV, whereas RVG performed equally well at

all tube potential settings, except than at 52 and

70 kV. For the Insight the largest TS values w ere

obtained with the smallest ESAK values whe-

reas with the Vistascan the largest TS were ob-

tained with ESAK values that where the largest

observed. Conclusions: The performance of all

receptors tested was greatly dependent on the

exposure parameters and mainly on the kV set-

tings. Overall, the RVG 6000 offered the best

image quality at doses somewhere in between

those required by the Insight and t he Vistascan.

Keywords: Radiography; Dental; Digital; Dosage

1. INTRODUCTION

According to the UNSCEAR 2000 Report, dental in-

traoral radiography is among the most frequently per-

formed radiological procedures [1]. Although the patient

exposure associated with dental radiography is relatively

low, intraoral radiography should be optimised in order

to keep the radiation risk “as low as reasonably achiev-

able”, something that is widely known as the ALARA

principle [2].

Over the past 20 years both the X-ray units and the

X-ray receptors used in dental radiology have been

evolved. Modern dental X-ray units incorporate high

frequency generators, operate at higher tube potentials

and produce X-ray spectra that have higher mean energy

and therefore are more penetrating compared to those

produced by older dental X-ray units. These improve-

ments have contributed in the reduction of the radiation

dose to the entrance skin surface of the patient and the

E. Katsoni et al. / Health 3 (2011) 56-65

Table 1. Specifications of the receptors evaluated in this study according to the data provided by the manufacturers.

Model Manufacturer Pixel size (μm) TechnologySoftware Bit/pixelWidth Height File size (MB)

INSIGHT KODAK N/A SILVER

HALIDE N/A N/A 3.1 cm 4.1 cm N/A

RVG 6000 KODAK 18.5×18.5 CMOS

KODAK

WINDOWS

6.0.1

8 1200

pixels

1600

pixels 1.8

PSP VISTASCAN DUERR SCAN PITCH

12.5

IMAGE

PLATE

DBSWIN

V.3.3 16 2476

pixels

3195

pixels Up to 9.3

enhancement of image quality [3,4] Concerning the

X-ray receptors, new digital systems have been intro-

duced in to the clinical practice and nowadays digital

radiography is considered an accepted imaging tech-

nique in dentistry. Currently, solid-state detectors based

on CCD or CMOS technology, photostimulable storage

phosphor (PSP) systems, along with the old-fashioned

but still widely used silver halide based films, are com-

mercially available for intraoral radiography.

In the international literature many articles can be

found describing digital detector systems and presenting

comparisons among various systems with regard to their

diagnostic performance [5-9]. The characteristics of the

x-ray beam, especially the x-ray beam energy, can also

contribute to image quality. Earlier studies have found

that the x-ray beam energy can affect image contrast

[10-12] and the signal-to-noise ratio [13]. For those stu-

dies various x-ray energies were used maintaining the

same exposure. Most of the equipment for intraoral ra-

diography provides limited adjustments for tube currents

and x-ray energy values. On most machines, only the

exposure time is usually set based on experience and

according tooth type.

The purpose of this study was to comparatively evalu-

ate, in a systematic inter-equipment manner, the per-

formance of three different intraoral image receptors

when exposed to different X-ray beam spectra, dose and

dose rate levels in detecting subtle radiographic density

differences using an aluminium stepwedge phantom.

2. MATERIALS AND METHODS

2.1. Image Acquisition

The intraoral radiographic receptors evaluated in this

study were: the Kodak Insight F speed class film (East-

man Kodak Company, Rochester, NY) the Kodak RVG

6000 (Eastman Kodak Company, Rochester, NY) and

the Duerr Vistascan Combi PSP system (Duerr Dental,

Bietingsheim-Bissingen, Germany). The first receptor is

a conventional silver halide film that requires chemical

processing, the second is a digital solid state receptor

based in CMOS technology that offers a direct display of

the digital image within seconds after the end of the ex-

posure and the third is a photostimulable storage phos-

phor (PSP) imaging plate which produces a digital image

after it has been scanned using a dedicated laser scanner

system. The main technical characteristics of the systems

tested are summarized in Table 1.

A modern dental X-ray unit (Prostyle Intra DC,

Plan-meca Oy, Helsinki, Finland) was used, offering

eight tube potential settings (ranging from 50 to 70 kVp),

seven tube current settings (ranging from 2-8 mA) and

26 exposure time settings (ranging from 0.01 to 3.2 sec).

The nominal total filtration was 2 mm Al and the focus

to collimator end distance was 30 cm.

A calibrated ion chamber dosimeter (Dosimeter 9010,

ionization chamber type 90x6-6; Radcal Corporation

Monrovia, USA) positioned at 30 cm from the tube fo-

cus was used to measure the dose in free air and deter-

mine the tube output at that distance. These measure-

ments were carried out for all the available tube potential,

tube current and exposure time selections, in order to

identify possible variations in output with different tube

loading values. The tube potential accuracy and repro-

ducibility were checked using a calibrated kilovolt peak

meter (Gammex RMI 245, 802108-1272, calibrated

31.01.2007, Gammex Inc., Middleton, USA).

Image quality was assessed using a quality control

phantom especially designed for dental radiography

(standard phantom for dental radiography, version 2.0),

shown in Figure 1. This phantom (henceforth referred to

as stepwedge phantom), is a Plexiglas parallelepiped

with a 30 mm by 40 mm base and 23 mm height, con-

taining a 12 step aluminium stepwedge which increases

in height from 1 to 12 mm, in 1 mm steps. In each step,

7 holes of 0.5 mm diameter have been drilled in depths

varying from 0.05 mm to 0.35 mm, in 0.05 mm incre-

ments. The stepwedge phantom has been described and

validated by Yoshiura et al. [14,15] The stepwedge

phantom was attached to the collimator end and the fo-

cus to receptor distance was kept constant at 35 cm.

The stepwedge phantom was radiographed using for

each receptor all possible tube potential, tube current and

exposure time selections combinations. The specific

X-ray unit used offered 182 mAs selections at each one

f the 8 different tube potential settings, that is, 1456

E. Katsoni et al. / Health 3 (2011) 56-65

26mm

20mm

1mm

12mm

10mm

1mm

2mm

0.5mm

1mm

5mm

Plexiglas

Aluminum steps

hole (depth:0.05, 0.1, 0.15,.....,0.3, 0.35mm

at each corresponding step)

10mm 30mm

40mm

Standard Phantom for Dental Radiography

(version 2.0)

0.5mm

[bottom view]

[lateral view]

5mm

Figure 1. Photographs (a,b) and schematic; (c) diagrams of the stepwedge phantom that was used for the

evaluation of image quality.

different exposure factor settings. The evaluation of im-

age quality was performed in two stages. In the first stage

the images for which all the 12 steps of the stepwedge

were visible were identified. The images satisfying this

criterion were included in the sample of images that were

further evaluated and will be henceforth referred to

as‘diagnosable images’ (see Figure 2). The images that

were not satisfying this criterion were rejected and will

be referred to as ‘non-diagnosable images’. The tube

current-exposure time combinations that produced diag-

nosable images determined the useful exposure range of

each receptor at each tube potential setting.

Concerning the exposures made with the Insight films,

all films were processed immediately after exposure,

using an automatic processor (Velopex Extra-X, Mediv-

ance Instuments, England) using the Readymatic dental

developer and Readymatic dental fixer solutions (East-

man Kodak, Rochester NY) at a temperature of 27℃.

This processor features an automatic replenishment sys-

Figure 2. Radiographic image of the

stepwedge phantom.

tem, however, in order to ensure that the processing

conditions remained fairly constant during the experi-

ments, the processing stability was repeatedly tested

every 50 films using sensitometry (Pehamed densitome-

ter Densinorm 21, PEHA med. Geräte GmbH-Sulzbach,

Germany). The acquired film radiographs were mounted

in opaque plastic holders and coded for later use. The

film radiographs were evaluated on a viewing box, with

all extraneous light masked. The observers were allowed

to use magnifying glasses (at multiple × 2 magnifica-

tion).

Concerning the RVG 6000 receptor, the original soft-

ware of the system was used for image capture. No im-

age processing was performed to enhance image quality

other than the system’s default pre-process. Finally,

concerning the Vistascan, the image plates were un-

packed in a dimly lit room and read out immediately

after exposure, using the Vistascan Combi system.[16,17]

The scanner’s resolution pitch settings were adjusted to

12.5 μm (corresponding to a theoretical resolution of 40

line pairs per mm).

2.2. Image Evaluation

Three certified dentists, postgraduate students in the

department of Oral Diagnosis and Radiology of the

Dental School of the University of Athens, served as

observers. They have over 5 years of experience inter-

preting analog and digital images. Before the actual im-

age quality evaluation session, the observers were first

trained in order to get familiar with the radiographic

images of the phantom and the rating process. During

E. Katsoni et al. / Health 3 (2011) 56-65

the image quality evaluation sessions, the observers had

to state the number of holes that they could perceive at

each step with 100% confidence level. The observers

rated the images independently (to account for in-

ter-observer variability), one image at a time and in a

random order, and dictated the ratings orally. Each image

was evaluated twice by each observer to account for

intra-observer variability. Due to the large number of

images, several sessions were required for each ob-

server to evaluate all images. For this reason, at the

beginning of each session and before the actual evalua-

tion, each observer was asked to observe 10 test control

images and rate them in order to get familiar with the

images again, before proceeding to the actual evalua-

tion session. For each image the average value of the

number of visible holes perceived by the three observ-

ers in each step was calculated and the summation of

these averages in all 12 steps was accounted as the total

image quality score (TS) of that image.

All digital images were viewed in fit to screen mode

on a 19-inch TFT monitor (Sony SDMHS95PR), with

1280 × 1024 resolution under subdued lighting condi-

tions. The monitor’s brightness and contrast were ad-

justed using the SMPTE (Society of Motion Picture and

Television Engineers) test pattern [18,19].

2.3. Statistical Analysis

The numbers of perceptible holes in each experimen-

tal condition (TSs) were statistically analyzed through

use of analysis of variance. Repeated measures ANOVA

were calculated for validity in relation to kV selection,

mA selection, exposure time and entrance surface air

kerma. Post hoc analysis was carried out using Tukey’s

test. P values less than 0.05 were considered to be statis-

tically significant. Intraobserver agreement was meas-

ured by Cohen’s Kappa and interobserver agreement was

measured by Fleiss kappa coefficient.

3. RESULTS

The output measurements revealed that the linearity of

output was within accepted limits at all tube potentials,

even though for small tube loadings (0.5 mAs) a reduc-

tion in output of up to 20% was observed with respect to

the mean value of output over the whole mAs range. The

reproducibility of output using the same exposure condi-

tions was better than 1%. The tube potential accuracy

and reproducibility were better than 3% and 1%, respec-

tively.

For all the images acquired, the receptor type, the ex-

posure parameters (kV, mA, s) were noted and the re-

spective entrance surface air kerma (ESAK) values were

then assigned. The number of diagnosable images was in

total 1230; 275 with Insight, 339 with RVG 6000 and

616 with Vistascan. From these values and Figure 3,

where the useful exposure range of each receptor is

shown, it is obvious that the VistaScan exhibited the

most extended useful exposure range for all tube poten-

tial settings. The useful exposure range of RVG 6000

was smaller compared to Vistascan but clearly larger

compared to that of the Inshight. It must be stressed

however, that the useful exposure ranges shown in Fig-

ure 3 are strictly valid only for the specific phantom and

geometric conditions used.

The useful exposure range of each receptor was de-

termined taking into account only the diagnosable im-

ages, that is, those images where all 12 steps were visi-

ble. However, within the sample of diagnosable images

great variations in the TS values were observed, which

are graphically represented in Figure 4. In this figure,

the lower bars extend from the minimum TS value ob-

served up to the 1st quartile value, the white boxes from

the 1st quartile value to the median value, the black

boxes from the median to the 3rd quartile value and the

upper bars from the 3rd quartile value to the maximum

TS value observed at the specific tube potential setting.

The black and white boxes represent the 50% of the ob-

served TS values at each tube potential setting. This fig-

ure clearly indicates that image quality was strongly de-

pended on the exposure factors used, not only for the

Insight, but for the digital receptors also. Even if the

exposure latitude of digital receptors is extended via

software manipulations of the recorded signals, still im-

age quality cannot be maintained constant within the

useful exposure range.

While Figure 4 is indicative of the large variations in

image quality with exposure factor settings, it is not so

useful for straightforward comparisons among receptors,

where the main interest is which receptor performs best

and at which tube potential setting and ESAK levels. To

facilitate comparisons among the receptors in terms of

the image quality offered and the respective radiation

dose required, the TS values of the best image obtained

with each receptor at each one of the tube potential set-

tings used and the respective exposure factors and ESAK

values required to obtain these images are given in Table

2. It can be seen that with the exception of 70 kV where

the largest TS value has been obtained with the Vistas-

can, for all the other tube potential settings, the largest

TS value was obtained with the RVG 6000 and it was

close to or larger than 60. The doses required were

roughly double compared to the respective doses re-

quired to obtain the maximum TS with the Insight. The

largest doses to obtain the maximum TS were observed

for the Vistascan and ranged from 3.2 to 4.8 mGy. It can

be seen that for the Insight the largest TS was obtained at

60 kV and at the same tube potential the largest TS

overall was obtained with the RVG 6000. For the Vis-

E. Katsoni et al. / Health 3 (2011) 56-65

Figure 3. Graphical representation of the exposure ranges of the three receptors tested in terms of mAs and in terms of

ESAK used to obtain the stepwedge phantom images at all tube potential settings.

Figure 4. Graphical description of the statistical variations observed in the number of visible holes (TS)

at different tube potential settings for each receptor. The lower bars extend from the minimum TS value

observed up to the 1st quartile value, the white boxes from the 1st quartile value to the median value, the

black boxes from the median to the 3rd quartile value and the upper bars from the 3rd quartile value to the

maximum TS value observed at the specific tube potential setting.

Openly accessible at

E. Katsoni et al. / Health 3 (2011) 56-65

Table 2. The total score (TS), the tube loading (mAs) and the respective ESAK (in mGy) of the image with the largest TS are given.

Receptor¥ kV 50 52 55 57 60 63 66 70

Insight

46.7 @ 3.0 mAs

(6 mA × 0.5 s):

45.3 @ 2.4 mAs

(3 mA × 0.8 s):

1.0 mGy 1.2 mGy

46.7 @ 2.0

mAs (5 mA ×

0.4 s): 1.0 mGy

45.3 @ 1.6

mAs (8 mA ×

0.2 s): 0.9 mGy

63.0 @ 1.4 mAs

(7 mA × 0.2 s):

0.8 mGy

56.7 @ 1.4 mAs

(7 mA × 0.2 s):

0.9 mGy

57.0 @ 1.3 mAs

(2 mA × 0.64 s):

0.9 mGy

50.3 @ 1.0

mAs (6 mA ×

0.16 s): 0.8

mGy

RVG 6000

60.3 @ 6.3 mAs

(5 mA × 1.25

s): 2.5 mGy

59.7 @ 5.0 mAs

(4 mA × 1.25

s): 2.2 mGy

66.3 @ 4.8

mAs (3 mA ×

1.6 s): 2.3 mGy

64.3 @ 4.8

mAs (3 mA ×

1.6 s): 2.5 mGy

66.7 @ 2.5 mAs

(2 mA × 1.25

s): 1.4 mGy

63.3 @ 2.5 mAs

(2 mA × 1.25

s): 1.6 mGy

66.3 @ 3.2 mAs

(2 mA × 1.6 s):

2.2 mGy

55.3 @ 2.5

mAs (2 mA ×

1.25 s): 2.0

mGy

Vistascan

48.3 @ 8.0 mAs

(5mA × 1.6s):

3.2 mGy

48.7 @10 mAs

(4mA × 2.5s):

4.4 mGy

49.0 @ 6.4

mAs (8mA ×

0.8s): 3.2 mGy

53.0 @ 8.0

mAs (8mA ×

1s):

4.4 mGy

54.0 @ 8.0 mAs

(4mA × 2s):

50.3 @ 6.0 mAs

(6mA × 1s):

4.8 mGy 4.1 mGy

61.7 @ 5.1 mAs

(8mA × 0.64s):

3.9 mGy

58.3 @ 3.8

mAs (6 mA ×

0.64 s): 3.3

mGy

Table 3. Influence of the kV selection on the performance of the evaluated receptors. Bold marked digits indicate statistical

significance (p < 0.05).

INSIGHT

kV 50 52 55 57 60 63 66 70

50 - 0.001 0.963 <0.001 <0.001 <0.001 <0.001 0.797

52 0.001 - <0.001 0.968 <0.001 <0.001 <0.001 <0.001

55 0.963 <0.001 - <0.001 <0.001 <0.001 <0.001 1.000

57 <0.001 0.968 <0.001 - <0.001 <0.001 <0.001 <0.001

60 <0.001 <0.001 <0.001 <0.001 - 0.001 0.011 <0.001

63 <0.001 <0.001 <0.001 <0.001 0.001 - 0.999 <0.001

66 <0.001 <0.001 <0.001 <0.001 0.011 0.999 - <0.001

70 0.797 <0.001 1.000 <0.001 <0.001 <0.001 <0.001 -

RVG 6000

kV 50 52 55 57 60 63 66 70

50 - 0.079 <0.001 <0.001 <0.001 0.012 0.025 0.968

52 0.079 - 0.070 0.773 0.148 0.996 1.000 0.001

55 <0.001 0.070 - 0.900 1.000 0.413 0.149 <0.001

57 <0.001 0.773 0.900 - 0.977 0.993 0.917 <0.001

60 <0.001 0.148 1.000 0.977 - 0.618 0.284 <0.001

63 0.012 0.996 0.413 0.993 0.618 - 1.000 <0.001

66 0.025 1.000 0.149 0.917 0.284 1.000 - <0.001

70 0.968 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 -

VISTASCAN

kV 50 52 55 57 60 63 66 70

50 - 0.031 1.000 0.006 0.007 0.889 <0.001 <0.001

52 0.031 - 0.089 <0.001 <0.001 0.488 <0.001 <0.001

55 1.000 0.089 - 0.001 0.001 0.985 <0.001 <0.001

57 0.006 <0.001 0.001 - 1.000 <0.001 <0.001 <0.001

60 0.007 <0.001 0.001 1.000 - <0.001 <0.001 <0.001

63 0.889 0.488 0.985 <0.001 <0.001 - <0.001 <0.001

66 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 - 0.041

70 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.041 -

E. Katsoni et al. / Health 3 (2011) 56-65

Figure 5. The distribution of images with TS larger or

equal with the 90% of the largest total score observed

for each receptor overall, with respect to the mA setting

and the ESAK at all tube potential settings is shown.

tascan the largest TS value was observed at 66 kV.

Within each receptor the statistical significance of the

tube potential settings in regard to the overall perform-

ance is presented in Tab le 3 . In this table it can be seen

that for the Insight film and the Vistascan system the TS

was greatly varying with the tube potential selection.

This dependence of TS with tube potential selection can

be also appreciated from Figure 4, when looking at the

median and the maximum TS values obtained at each

tube potential. The Insight showed an increased per-

formance at 60 kV, Vistascan at 66 kV whereas the RVG

6000 could perform equally well at almost all tube po-

tentials except 52 and 70 kV. It must be noted finally,

that no straightforward correlation was observed be-

tween TS and ESAK in any of the receptors compared,

not even for the Vistascan where the image quality was

up to a point increasing with increasing dose.

An interesting observation was made during the

analysis of the results, concerning the dependence of TS

on the tube current selection, which for a given tube po-

tential selection is what determines the incident dose rate

to the receptor. For the RVG 6000 it was observed that

for a given kV selection, the same mAs that produced a

diagnosable image when a low mA selection was used

produced a non-diagnosable image when a high mA se-

lection was used. Table 2 is rather suggestive of this

behaviour since for all tube potential settings above 52

kV the image with the largest TS was obtained with ei-

ther the 3 or the 2 mA selection. Furthermore, the statis-

tical analysis also verified the dependence of the RVG

6000 performance on the mA selection, as the TS values

obtained with the 2 mA selection differed statistically

significant (p < 0.05) from the other selections available.

Unlike the RVG 6000, for the Insight and the Vistascan

not any dependence of their performance on dose rate

was documented.

The dependence of image quality on tube potential,

tube current and ESAK can be better appreciated from

Figure 5. In this figure the distribution of images with

TS of at least 90% of the largest TS value observed for

each receptor within the whole useful exposure range is

given with respect to the tube current settings and ESAK

at the various tube potential settings used. The threshold

of 90% was chosen in order to sort out the images with

the best quality overall. For the Insight almost all except

one best quality images were obtained at 60 kV, with

medium to high mA settings and with ESAK values

ranging from 0.52 to 1.36 mGy. For the RVG 6000 the

best quality images were obtained over a wide range of

tube potentials (50-66 kV) and with ESAK values rang-

ing from 0.55 to 2.82 mGy. However, it is characteristic

that 8 out of the 10 images were obtained using either

the 2 or the 3 mA tube current selections. For the Vis-

tascan, the best quality images were all obtained at 66

kV (using the 3 largest mA selections) and at 70 kV with

all mA selections, with the ESAK values ranging from

0.96 to 4.2 mGy.

Finally, in Figure 6 the variation of the average num-

ber of visible holes in each one of the 12 steps is given,

considering only the images with the highest TS at each

one of the tube current selection available, for all the

tube potential settings used. For the images obtained

with Vistascan the maximum number of visible holes

was observed in the first step and then the number of

visible holes gradually reduced when moving to the

thicker Aluminium steps, similarly with the images ac-

quired with the Insight, with the only exception that in

the Insight images the maximum number of visible holes

was most often observed in the second step rather than in

the first. In the images acquired with the RVG 6000 the

visibility of the holes was severely reduced in the first

two steps, reached a maximum in steps 5 and 6 and then

gradually reduced again. Evidently, the visibility of

holes in the thicker steps is poorer in the images ac-

quired with the Insight and the Vistascan than in the im-

ages acquired with the RVG 6000 system.

Due to the subjective nature of the image quality

evaluation procedure, in order to accept the validity of

the above results it was necessary to also validate statis-

tically the reliability of the observers’ evaluations. A

high level of agreement between the first and the second

evaluation and consequently high intra-observer reliabil-

ity scores (kappa value) were obtained, as can be seen in

Table 4. Concerning the inter-observer variability, kappa

values of 0.92, 0.88 and 0.96 were observed respectively

for the Insight, Kodak RVG 6000 and Vistascan, indi-

ating a very good inter-observer agreement. c

E. Katsoni et al. / Health 3 (2011) 56-65

Figure 6. The variation of the average number of visible holes in each one of the 12 steps (considering only the

images with the highest TS for each tube current setting) is given for the different tube potential settings used.

Table 4. Intra observer agreement measured by Cohen’s Kap-

pa.

overall Observer 1 Observer 2 Observer 3

Insight 0.95 0.94 0.95 0.96

RVG 6000 0.90 0.89 0.91 0.90

Vistascan 0.93 0.93 0.92 0.94

4. DISCUSSION

All comparisons presented above are based on the step-

wedge phantom and therefore their validity relies on the

assumption that this phantom is appropriate for simulat-

ing adequately clinical conditions. Yoshiura et al. [14,15]

who evaluated this phantom, concluded that the X-ray

attenuation range produced with this phantom is similar

to that produced in clinical practice. Furthermore, they

concluded that image quality could be quantitatively

evaluated by means of the number of visible holes, since

their detection is equivalent with the detection of small

lesions in an actual clinical situation. Although this

phantom does not represent all diagnostic tasks encoun-

tered in actual clinical situations, it is by all means suffi-

cient for the relative comparison of performance among

different systems.

No image enhancement was carried out in this study.

The software provided by the manufactures displayed

postprocessed images immediately after acquisition.

Several studies on the effects of image quality and ob-

server performance have been published. Further image

processing could improve the quality of the displayed

images according to some studies [20,21] whereas ac-

cording to other does not make any significant difference

[22,23]. The effect of image processing varies per sys-

tem but its impact on image quality was outside the

scope of the current study.

Concerning the Vistascan, the extended useful expo-

sure range observed in this study is in agreement with

what is already known from the applications of the PSP

technology receptors in general and dental radiology

[24-28]. Furthermore, the high ESAK values required to

produce images with the largest TS are in agreement

with the study of Berkhout et al. [24] who demonstrated

that PSP systems produce the best quality images at

doses up to ten times the minimum dose that produces a

diagnosable image. However, this characteristic of Vis-

tascan can lead to the systematic use of higher doses than

those actually needed for diagnosis, something that from

the aspect of radiation protection is considered disadvan-

tageous and in the general radiology applications it has

been reported as ‘the exposure factor creep’ [29].

In this study, the dependence of the performance on

the tube potential setting was observed mainly for the

Inshight and the Vistascan. Likewise, Svenson et al. [30]

demonstrated the differences in diagnostic accuracy for

different tube potential settings for two films differing in

speed. Concerning the Insight, Kaeppler et al. [31] also

concluded that the Insight film achieved the best results

at 60 kV.

In the present study it was seen that in the overall rat-

ing, the Kodak RVG6000 performed better in the detec-

tion of holes compared to the other two receptors tested.

This result is in agreement with the results of other re-

searchers who also demonstrated the superiority of

RVG6000 receptor [33].

It is well known that the majority of the dental X-ray

units currently in use worldwide offer a single tube po-

tential and tube current selection. Since specific recep-

tors perform better at certain tube potential and tube

current settings, it is important for the potential buyer to

take this into account when a new receptor or dental

X-ray unit is to be combined with existing equipment or

when both a new receptor and a new dental X-ray unit

are to be bought. Although, an image of adequate diag-

nostic quality can be achieved by all receptors tested in

many tube potential and tube current settings, the best

image quality was obtained with the least radiation ex-

posure only when specific exposure parameters were

used. For example, even the Vistascan which exhibited

the most extended useful exposure range at all tube po-

tential settings from 50 to 70 kV, performed better at 66

E. Katsoni et al. / Health 3 (2011) 56-65

and at 70 kV. Therefore, Vistascan should be combined

with a dental X-ray unit operating at 70 kV rather with

one operating at 60 kV. Unlike to Vistascan, RVG 6000

and Inshight should be combined preferably with an

X-ray unit operating at 60 kV.

The ALARA statement endorses the principle that (in-

dividual) doses should be as low as reasonably achiev-

able. In this context European guidelines [30] recom-

mended the establishment of diagnostic reference levels

(DRLs). The European guidelines [30] recommended a

DRL of 4 mGy absorbed dose in air measured at the end

of the spacer cone for a standard maxillary molar projec-

tion. In this study it was shown that for the specific

phantom used, images of good diagnostic quality could

be obtained with all receptors tested using just a fraction

of the proposed DRL (see Figure 5).

Similar surveys conducted in dental radiographic fa-

cilities [34-36] over the last 10 years have demonstrated

a trend for reduction of the entrance surface dose, with

the use of faster films and digital receptors, as well as

with modern x-ray units and rectangular collimation.

Compared with the previous guidelines, the 2007 ICRP

[37] recommendations for estimating risk associated

with exposure to radiation, increased the emphasis

given on the structures located in the oral region, par-

ticularly the salivary glands. According to the revised

recommendations for calculating effective dose, dental

radiography involves 32% to 422% more risk than that

previously thought [38]. Therefore, efforts should be

made to reduce dose as much as possible but not at ex-

pense of image quality and diagnostic accuracy. In this

context it is important to determine for each receptor the

exposure factors settings which can produce a good

quality image with the least radiation dose possible. For

this reason it would be ideal if receptor manufacturers

could determine, using a phantom like the one used in

this study, the optimal exposure factor settings and the

respective TS and ESAK values for each receptor that

becomes commercially available and include this in-

formation in their technical data sheets. In this way the

potential users would be able to determine if a given

receptor is well suited for the dental X-ray unit that may

already have or they intend to buy and furthermore they

would be able to compare receptors using performance

indices that relate to the clinical practice and therefore

are easy to comprehend.

The main results of this study in terms of the relative

performance of the three receptors studied can be sum-

marized as follows: Vistascan exhibited the most ex-

tended useful exposure range, followed by RVG 6000

and Insight. RVG 6000 exhibited the largest TS values in

all tube potential settings except 70 kV where the Vis-

tascan performed better. Insight performed better than

Vistascan only at 60 and 63 kV. Vistascan exhibited its

largest TS values at 66 and 70 kV, Insight at 60 and 66

kV, whereas RVG exhibited large TS values at all tube

potential settings, except than at 52 and 70 kV. For the

Insight the largest TS values were obtained with the

smallest ESAK values whereas with the Vistascan the

largest TS were obtained with ESAK values that where

the largest observed. RVG 6000 was the only system that

exhibited a dependence on the dose rate since most of

the large TS values were obtained at low mA selections.

REFERENCES

[1] United Nations Scientific Committee on the Effects of

Atomic Radiation (2000) Sources and effect of ionizing

radiation. UNSCEAR publications, Report, 1.

[2] European Union. Council Directive 97/43 Euratom (1997)

On health protection of individuals against the dangers of

ionizing radiation in relation to medical exposures, and re-

pealing directive 84/466 Euratom. Official Journal of the

European Communities, 22, L180.

[3] Kitagawa, H. and Farman, A.G. (2004) Effect of beam

energy and filtration on the signal-to-noise ratio of the

Dexis intraoral X-ray detector. Dentomaxillofacial Radi-

ology, 33, 21-24. doi:10.1259/dmfr/26493631

[4] Wenzel, A. (2006) A review of dentists' use of digital

radiography and caries diagnosis with digital systems.

Dentomaxillofacial Radiology, 5, 307-314.

doi:10.1259/dmfr/64693712

[5] Hellen-Halme, K. (2007) Quality aspects of digital radi-

ography in general dental practice. Swedish Dental

Journal. Supplement, 184, 9-60.

[6] Hellen-Halme, K., Nilsson, M. and Petersson, A. (2007)

Digital radiography in general dental practice: A field

study. Dentomaxillofacial Radiology, 36, 249-255.

doi:10.1259/dmfr/95125494

[7] Cowen, A.R., Kengyelics, S.M. and Davies, A.G. (2008)

Solid-state, flat-panel, digital radiography detectors and

their physical imaging characteristics. Clinical Radiology,

63, 487-498. doi:10.1016/j.crad.2007.10.014

[8] Hintze, H. (2006) Diagnostic accuracy of two software

modalities for detection of caries lesions in digital radio-

graphs from four dental systems. Dentomaxillofacial Ra-

diology, 35, 78-82. doi:10.1259/dmfr/50356588

[9] Lanca, L., Silva, A., Alves, E., Serranheira, F. and Cor-

reia, M. (2008) Evaluation of exposure parameters in

plain radiography: A comparative study with European

guidelines. Radiation Protection Dosimetry, 129, 316-320.

doi:10.1093/rpd/ncn144

[10] Harada, T., Nishikawa, K., Shibuya, H., Hayakawa, Y.

and Kuroyanagi, K. (1995) Sens-A-Ray characteristics

with variation in beam quality. Oral Surgery, Oral Medi-

cine, Oral Pathology, Oral Radiology & Endodontics, 80,

120-123. doi:10.1016/S1079-2104(95)80028-X

[11] Goshima, T., Goshima, Y., Scarfe, W.C. and Farman, A.G.

(1996) Sensitometric response of the Sens-A-Ray, a charge

coupled imaging device, to changes in beam energy. Den-

tomaxillofacial Rad io lo gy, 25, 17-18.

[12] McDonnell, D. and Price, C. (1993) An evaluation of the

Sens-A-Ray digital dental imaging system. Dentomaxil-

lofacial Radiology, 22, 121-126.

[13] Kitagawa, H. and Farman, A.G. (2004) Effect of beam

energy and filtration on the signal to noise ratio of the

E. Katsoni et al. / Health 3 (2011) 56-65

Dexis intraoral x-ray detector. Dentomaxillofacial Radi-

ology, 33, 21-24. doi:10.1259/dmfr/26493631

[14] Yoshiura, K., Kawazu, T., Chikui, T., Tatsumi, M., To-

kumori, K., Tanaka, T., et al. (1999) Assessment of image

quality in dental radiography, part 1: Phantom validity.

Oral Surgery, Oral Medicine, Oral Pathology, Oral Ra-

diology, and Endodontics, 87, 115-122.

doi:10.1016/S1079-2104(99)70304-5

[15] Yoshiura, K., Kawazu, T., Chikui, T., Tatsumi, M., To-

kumori, K., Tanaka, T., et al. (1999) Assessment of image

quality in dental radiography, part 2: Optimum exposure

conditions for detection of small mass changes in 6 in-

traoral radiography systems. Oral Surgery, Oral Medi-

cine, Oral Pathology, Oral Radiology, and Endodontics,

87, 123-129. doi:10.1016/S1079-2104(99)70305-7

[16] Ang, D.B., Angelopoulos, C. and Katz, J.O. (2006) How

does signal fade on photo-stimulable storage phosphor

imaging plates when scanned with a delay and what is

the effect on image quality? Oral Surgery, Oral Medicine,

Oral Pathology, Oral Radiology, and Endodontics, 102,

673-679. doi:10.1016/j.tripleo.2005.11.002

[17] Ramamurthy, R., Canning, C.F., Scheetz, J.P. and Farman,

A.G. (2004) Impact of ambient lighting intensity and du-

ration on the signal-to-noise ratio of images from photo-

stimulable phosphor plates processed using DenOptix

and ScanX systems. Dentomaxillofacial Radiology, 33,

307-311. doi:10.1259/dmfr/91373164

[18] Parsons, D.M., Kim, Y. and Haynor, D.R. (1995) Quality

control of cathode-ray tube monitors for medical imaging

using a simple photometer. Journal of Digital Imaging, 8,

10-20. doi:10.1007/BF03168051

[19] Jervis, S.E. and Brettle, D.S. (2003) A practical approach

to soft-copy display consistency for PC-based review

workstations. The British Journal of Radiology, 76,

648-652. doi:10.1259/bjr/25693100

[20] Alpoz, E., Sogur, E. and Baksi Akdeniz, B.G. (2007)

Perceptibility curve test for digital radiographs before

and after application of various image processing algo-

rithms. Dentomaxil lofacial Radiology, 36, 490-494.

doi:10.1259/dmfr/20897311

[21] Li, G., Sanderink, G.C., Berkhout, W.E., Syriopoulos, K.

and van der Stelt, P.F. (2007) Detection of proximal car-

ies in vitro using standard and task-specific enhanced

images from a storage phosphor plate system. Caries

Research, 41, 231-234. doi:10.1159/000099324

[22] Koob, A., Sanden, E., Hassfeld, S., Staehle, H.J. and

Eickholz, P. (2004) Effect of digital filtering on the

measurement of the depth of proximal caries under dif-

ferent exposure conditions. American Journal of Den-

tistry, 17, 388-393.

[23] Li, G., Engström, P.E. and Welander, U. (2007) Meas-

urement accuracy of marginal bone level in digital radio-

graphs with and without color coding. Acta Odontologica

Scandinavica, 65, 254-258.

doi:10.1080/00016350701452089

[24] Berkhout, W.E., Beuger, D.A., Sanderink, G.C. and van der

Stelt, P.F. (2004) The exposure range of digital radio-

graphic systems: Dose reduction or risk of overexposure?

Dentomaxillofacial Radiology, 33, 1-5.

doi:10.1259/dmfr/40677472

[25] Borg, E., and Grondahl, H.G. (1996) On the exposure

range of different X-ray photon detectors in intra-oral ra-

diography. A comparison of image quality in film,

charge-coupled device and storage phosphor systems.

Dentomaxillofacial Radiology, 25, 82-88.

[26] Borg, E., Attaelmanan, A. and Grondahl, H.G. (2000)

Image plate systems differ in physical performance. Oral

Surgery, Oral Medicine, Oral Pathology, Oral Radiology,

and Endodontics, 89, 118-124.

doi:10.1016/S1079-2104(00)80026-8

[27] Farman, A.G. and Farman, T.T. (2005) A comparison of

18 different x-ray detectors currently used in dentistry.

Oral Surgery, Oral Medicine, Oral Pathology, Oral Ra-

diology, and Endodontics, 99, 485-489.

doi:10.1016/j.tripleo.2004.04.002

[28] Kitagawa, H., Farman, A.G., Scheetz, J.P., Brown, W.P.,

Lewis, J., Benefiel, M., et al. (2000) Comparison of three

intra-oral storage phosphor systems using subjective im-

age quality. Dentomaxillofacial Radiology, 29, 272-276.

doi:10.1038/sj.dmfr.4600532

[29] Willis, C.E. (2004) Strategies for dose reduction in ordinary

radiographic examinations using CR and DR. Pediat ric Ra-

diolog y, 34, 196-200. doi:10.1007/s00247-004-1269-6

[30] Svenson, B., Grondahl, H.G., Petersson, A. and Olving, A.

(1985) Accuracy of radiographic caries diagnosis at dif-

ferent kilovoltages and two film speeds. Swedish Dental

Journal, 9, 37-43.

[31] Kaeppler, G., Dietz, K. and Reinert, S. (2007) Influence

of tube potential setting and dose on the visibility of le-

sions in intraoral radiography. Dentomaxillofacial Radi-

ology, 36, 75-79. doi:10.1259/dmfr/75743437

[32] Ramamurthy, R., Scheetz, J.P., Clark, S.J. and Farman,

A.G. (2006) Effects of imaging system and exposure on

accurate detection of the second mesio-buccal canal in

maxillary molar teeth. Oral Surgery, Oral Medicine, Oral

Pathology, Oral Radiology, and Endodontics, 102, 796-802.

doi:10.1016/j.tripleo.2006.02.009

[33] Radiation Protection (2004) European guidelines on ra-

diation protection in dental radiology. The safe use of ra-

diographs in dental practice. Issue No. 136.

[34] Yakoumakis, E.N., Tierris, C.E., Stefanou, E.P., Pha-

nourakis, I.G. and Proukakis, C.C. (2001) Image quality

assessment and radiation doses in intraoral radiography.

Oral Surgery, Oral Medicine, Oral Pathology, Oral Ra-

diology, and Endodontics, 91, 362-368.

doi:10.1067/moe.2001.111940

[35] Hatziioannou, K., Psarouli, E., Papanastassiou, E., Bous-

bouras, P., Kodona, H., Kimoundri, O., et al. (2005)

Quality control and diagnostic reference levels in in-

traoral dental radiographic facilities. Dentomaxillofacial

Radiology, 34, 304-307. doi:10.1259/dmfr/38802780

[36] Gonzalez, L. and Moro, J. (2007) Patient radiation dose

management in dental facilities according to the X-ray

focal distance and the image receptor type. Dentomaxil-

lofacial Radiology, 36, 282-284.

[37] Valentin, J. (2007) The 2007 recommendations of the

international commission on radiological protection. El-

sevier, Oxford. doi:10.1259/dmfr/67494525

[38] Ludlow, J.B., Davies-Ludlow, L.E. and White, S.C.

(2008) Patient risk related to common dental radio-

graphic examinations: The impact of 2007 International

Commission on Radiological Protection recommenda-

tions regarding dose calculation. Journal of American

Dental Association, 139, 1237-1243.