|Year : 2013 | Volume
| Issue : 4 | Page : 261-268
Comparison of measurements from conventional, scanned, and photographed lateral cephalograms: An in vitro study
Sk Tanveer Ahamed, Revathi Peddu, Shyam Kumar Bandaru, Kalyani Mallavarapu, Sai Prakash Adusumilli, S. R. K. Reddy
Department of Orthodontics, Sibar Institute of Dental Sciences, Guntur, Andhra Pradesh, India
|Date of Web Publication||26-Nov-2013|
Sk Tanveer Ahamed
5-88-3, Ameen Mahal, 3rd Lane Lakshmipuram, Guntur - 522 007 Andhra Pradesh
Source of Support: None, Conflict of Interest: None
Background: The cephalograms are invaluable in diagnosis, treatment planning, and evaluation of treatment results. Digital cephalograms have several advantages. Conventional cephalograms can be converted to digital format by scanning on a flatbed scanner. More recently it was reported that conventional cephalograms photographed with a digital camera can be used for doing cephalometric analysis. However, the accuracy of the measurements is not very clear.
Aim: To compare the accuracy of angular and linear measurements obtained from scanned and photographed images of conventional cephalograms with that of manually traced original cephalograms.
Materials and Methods: Twenty cephalograms of patients taken in natural head position (NHP) were converted to digital format by scanning on a flatbed scanner and also were photographed with a high resolution digital camera. Both images were digitized with Vista Dent OC Imaging TM cephalometric software. Customized cephalometric analysis was performed on both images, and the measurements were recorded.
Results: Statistical analysis was done using paired Student's t-test and intraclass correlation coefficient tests. The results showed that; 1. There was significant distortion when analog cephalograms were converted to digital format by photographing. 2. Scanned images were more comparable to the original cephalograms than photographed images. 3. Among the linear and angular measurements, linear measurements showed significant variation than angular measurements.
Conclusion: It is acceptable to use digital photographs of cephalograms if angular measurements are primarily required. However, these images might not be acceptable if linear measurements are needed.
Keywords: Angular and linear measurements, conventional cephalograms, digital cephalograms, photographed images, scanned images
|How to cite this article:|
Ahamed ST, Peddu R, Bandaru SK, Mallavarapu K, Adusumilli SP, Reddy S. Comparison of measurements from conventional, scanned, and photographed lateral cephalograms: An in vitro study. J NTR Univ Health Sci 2013;2:261-8
|How to cite this URL:|
Ahamed ST, Peddu R, Bandaru SK, Mallavarapu K, Adusumilli SP, Reddy S. Comparison of measurements from conventional, scanned, and photographed lateral cephalograms: An in vitro study. J NTR Univ Health Sci [serial online] 2013 [cited 2020 Apr 8];2:261-8. Available from: http://www.jdrntruhs.org/text.asp?2013/2/4/261/122162
| Introduction|| |
Digital cephalometrics claim numerous advantages including improved landmark identification and faster cephalometric data analysis. ,,, Conventional cephalometric radiographs can be scanned and the images can be traced by computer software for cephalometric landmark identification and to perform analysis. 
As recently reported digital cameras available in most orthodontic practices can be used to make a digital photograph of the conventional cephalogram and trace images with orthodontic software for analysis.  As there is very little literature available on the accuracy of linear and angular measurements obtained with this new technique, the present study was undertaken with the aim to compare the accuracy of angular and linear measurements obtained from scanned and photographed images with manually traced original cephalograms.
| Materials and methods|| |
Twenty cephalograms of patients were taken in NHP, by asking the patient to keep the head erect and look into his eyes in a mirror kept 109 cm away [Figure 1]. Two calibration lead points were marked on nasal rod of the cephalostat at a distance of 50 mm from each other and the cephalograms were taken with Villa Sistemi Medicali Model No-82088150 with standardized peak of 80 kV and exposure time of 80 s [Figure 2]. All radiographs were manually traced with 3H pencil on 0.3 lead acetate tracing sheets [Figure 3]. Each radiograph was marked at four corner locations (P1, P2, P3, and P4) as four fiducial points [Figure 4] at a predetermined distance of 98 mm between P1-P2, P3-P4; and 100 mm between P1-P4, P2-P3 to determine any distortion in recording of images.
|Figure 1: Patients head fixed in cephalostat in natural head position (NHP) Patient consent is provided along with figures|
Click here to view
|Figure 4: Cephalogram with four fi ducial points marked (P1, P2, P3, and P4)|
Click here to view
When cephalometric X-ray film was exposed, the lead points placed at 50 mm apart measured 54 mm and was taken as a standard for calibrating scanned and photographed images of the cephalogram to give 1:1 magnification and to prevent magnification error.
All 20 radiographs were photographed with an single-lens reflex (SLR) camera (Canon EOS 1000D) fixed on a tripod held at a predetermined distance of 107 cm from light box as per Collins et al.,  [Figure 5]. Light box illumination, flash, and fine settings were standardized for all photographed images. The film speed was ISO-100, aperture set at F/14, shutter speed at 1/30 s to avoid camera shake. Images were changed to grey scale by Adobe Photoshop (version 7), and saved with minimum compression in Joint Photographic Experts Group (JPEG) format and imported onto computer which were to be digitized.
|Figure 5: Radiographs photographing with a single-lens reflex (SLR) camera|
Click here to view
All 20 radiographs were scanned with a flatbed color scanner, Umax Power Look 2100xI scanner, with transparency adapter (Umax Scanners Ltd., Thailand) [Figure 6] at 300 dpi resolution with bit depth 8. After initially being saved as Tagged Image File Format (TIFF) documents, they were converted to JPEG format. The images were also converted from color to grey scale in Adobe Photoshop with minimum compression as landmarks are more easily identified by evaluation of differing shades of grey rather than by looking at color images.
|Figure 6: Umax Power Look 2100 XL-USB. Flatbed colour scanner used in the study|
Click here to view
All the 40 images of the cephalograms were then viewed on computer screen with an imaging software program (Vista Dent OC Imaging TM , cephalometric software) [Figure 7]. Six linear and eight angular measurements of fiducial points were recorded for testing the distortion of the photographed and scanned images and compared them with the original cephalograms. A customized cephalometric analysis was made, consisting of six linear and six angular measurements as follows:
- Linear measurements in mm used for customized analysis (original, photographed, and scanned) were: U1-NA, L1-NB, ANS-PNS, U1-NPOG, L1-NPOG, and N-ME.
- Angular measurements in degrees used for customized analysis (original, photographed, and scanned) were: SNA, SNB, ANB, U1-SN, L1-MndPL, and Mx-MndPL.
The linear and angular measurements of the fiducial points in mm and degrees used for original, photographed, and scanned cephalograms were:
Linear (in mm): P1-P2, P2-P3, P3-P4, P4-P1, P1-P3, and P2-P4.
Angular (in degrees): A: P1-P2-P4, A: P4-P2-P3, A: P1-P3-P2, A: P1-P3-P4, A: P2-P4-P3, A: P2-P4-P1, A: P4-P1-P3, and A: P2-P1-P3.
The cephalometric landmarks and fiducial points for all the 20 original hand traced cephalograms were marked with a sharp 3H pencil, keeping the cephalograms on illuminated light box and the linear and angular measurements were measured with a ruler and protractor [Figure 8] and [Figure 9].
|Figure 8: Cephalometric landmark and fiducial points traced on acetate sheet|
Click here to view
For all the 40 digital images of the original cephalograms customized cephalometric landmarks and fiducial points were digitized on the monitor screen and the linear and angular measurements for both cephalometric landmarks and fiducial points were measured using ruler option in Vista Dent OC Imaging TM cephalometric software [Figure 10].
|Figure 10: Measuring cephalometric landmarks and fiducial points using ruler|
Click here to view
Intraclass correlation coefficient test was used to check the reliability of the measurements. Descriptive statistics of mean and standard deviations (SDs) were calculated for all the groups. Student's paired t-test was used to evaluate statistical significance for comparing mean values between original and photographed groups, original and scanned groups, and photographed and scanned groups setting significance value at *P < 0.05.
| Results|| |
Mean and SD values for the linear and angular measurements for the fiducial points of the three groups were shown in [Table 1].
|Table 1: Mean and SD Values for Linear and Angular Measurements of Fiducial Points for Different Parameters in Three Groups|
Click here to view
When the mean differences in the linear measurements between the original cephalograms and the photographed versions were compared, the photographed groups showed enlargement both in horizontal and vertical components; and when the original and scanned versions were compared, the scanned versions showed no distortion in the horizontal component and enlargement in the vertical component was observed.
When the photographed and scanned versions were compared, the photographed versions showed enlargement in both horizontal and vertical components. When the angular measurements of the fiducial points between original and photographed images were compared, two angular measurements of the photographed images out of eight showed increased values while the remaining 6 showed slight decrease.
When the angular measurements of the fiducial points between the original and scanned images were compared, two out of eight angular measurements of the scanned images showed increased values while the remaining 6 showed marginal decrease.
Mean and SDs of individual linear and angular cephalometric measurements of all the three groups were shown in [Table 2].
|Table 2: Mean and SD Values for Linear and Angular Measurements of Customized Cephalometric Analysis for Different Parameters in Three Groups|
Click here to view
The comparison of the mean values of the original with that of photographed and scanned groups showed that five among six linear measurement values showed higher values in both photographed and scanned versions as compared to the original and one linear measurement value was less in both photographed and scanned versions as compared to the original.
The angular measurements of photographed and scanned versions are comparable to the original cephalometric values, though both photographed and scanned groups showed slightly higher total mean value for the six parameters as compared to the original.
[Table 3] and [Table 4] shows intraclass correlation coefficient (reliability) of angular and linear measurements, respectively, among the three groups (original, photographed, and scanned), which showed good reliability at 95% confidence interval (CI).
|Table 3: Intraclass Correlation Coefficient (Reliability) Among the Three Groups (Angular Measurements)|
Click here to view
|Table 4: Intraclass Correlation Coefficient (Reliability) Among the Three Groups (Linear Measurements)|
Click here to view
| Discussion|| |
In the present study, 20 cephalograms of the patients were taken in natural head position (NHP). The advantages of taking cephalometric films in NHP and various methods to obtain NHP have been described by various investigators. ,,,,, In our study, we adopted a method suggested by Pereira  by asking the patient to stand relaxed and look into his/her own eyes in a mirror placed 190 cm away.
Four fiducial points were marked on the cephalogram (P1, P2, P3, and P4) at a predetermined distance of 98 mm between P1-P2 and P3-P4 (horizontal) and 100 mm between P1-P4 and P2-P3 (vertical). Both horizontal and vertical distances were kept almost equal to permit equal contribution between horizontal and vertical components. Collins et al.,  in their study fixed the distance between the points A-B and B-C (horizontal) as 70 mm and distance between A-D and B-C (vertical) as 90 mm and emphasized on vertical component, while Bruntz et al.,  in their study have not marked the fiducial points at a predetermined distance, but marked them at four corners of the original cephalogram which measured a mean horizontal distance of 177.4 mm and mean vertical distance of 155.8 cm and emphasized on horizontal components and compared them with the corresponding values obtained with the scanned images to access the distortion or magnification.
Bruntz et al.,  in their study scanned cephalograms at 150 dpi. They felt that during landmark digitization, magnification often was used to more accurately identify certain structures and in several instances the magnification caused significant pixilation and blurriness of images making it difficult to accurately identify the structures and suggested higher scanning dpi. Hence, all the 20 cephalograms in our study were scanned on a flatbed scanner standardized at 300 dpi resolution with bit depth 8.
The cephalograms were photographed with a Canon EOS 1000D SLR camera held at a distance of 107 cm from the light box as advocated by Collins et al.,  and camera was fixed on a tripod to avoid camera shake. The images were saved in JPEG format with minimum compression and imported on to a computer screen for digitization.
All the 20 original cephalograms were hand traced carefully on matte acetate overlaid by the same operator to minimize interoperator tracing and land mark identification errors, and images were viewed on a computer screen by the same operator and digitized the fiducial and chosen cephalometric landmarks using Vista Dent OC Imaging TM cephalometric software.
Collins et al.,  presumed that photographing an image before digitization makes no significance difference with regard to the angular measurements and the angular measurements using fiducial points was not considered in his study. However, to verify the reproducibility and accuracy of the angular measurements in our study, eight angular measurements A: P1-P2-P4, A: P4-P2-P3, A: P2-P3-P1, A: P1-P3-P4, A: P3-P4-P2, A: P1-P4-P2, A: P4-P1-P3, and A: P3-P1-P2 were considered based on four fiducial points P1, P2, P3, and P4.
A customized cephalometric analysis was designed in the present study and six linear and six angular measurements were chosen. Chen et al.,  and Bruntz et al.,  showed discrepancies in the vertical component when identifying the landmarks porion, orbitale, and gnathion on digital media and also indicated that FH plane was highly unreliable with digital media. Hence, both linear and angular measurements based on identification of cephalometric landmarks porion and orbitale was avoided in our study. The six linear cephalometric readings chosen were U1-NA, L1-NB, ANS-PNS, U1-NPOG, L1-NPOG, and N-ME; and the six angular measurements chosen were SNA, SNB, ANB, U1-SN, L1-MndPL, and MX-MNDPL.
Fiducial points P1, P2, P3, P4 and customized cephalometric landmarks were marked on the cephalometric tracing sheet for digitization on the computer screen for analysis. Collins et al.,  in their study collected the cephalograms randomly from the patients files which were taken on the same X-ray machine to reduce calibration errors and for this reason all cephalometric radiographs in our study were taken on the same machine.
The data from the fiducial points (linear and angular) and cephalometric landmarks were recorded and was statistically analyzed using Statistical Package for Social Sciences (SPSS) 10 software (SPSS Inc. Chicago, IL).
The results showed that the photographed version of the original cephalograms showed an enlargement of 0.30 mm horizontally and vertical enlargement of 0.41 mm, while photographed groups showing distortion in both horizontal and vertical components, in the linear measurements between the original and scanned versions showed an enlargement of 0.13 mm in vertical component and no distortion in the horizontal component. Marcμ and Wenzel  reported a 2% enlargement in the vertical plane and no distortion in horizontal plane. Our results showed that the scanned images showed no distortion in horizontal plane and 0.13 mm distortion in vertical plane.
When the distortion between photographed and scanned images with that of the original cephalograms were compared; both photographed and scanned versions showed higher values than original, but scanned images are much closer and comparable to the original, while significant difference was shown between photographed and original versions.
The comparison of angular measurements of fiducial points of photographed and scanned images with that of the original showed a mean difference of 0.06° increase and 0.09° decrease in the individual measurements of photographed and scanned versions, respectively which has no clinical significance and hence our results concur with Collins et al.  The marginal variations of the angular measurements with that of the original may be attributed to the point location, tracing, and measurement error than the variation due to distortion of the images.
Comparison of the mean values of cephalometric readings of the original with that of the photographed and scanned groups showed that five among six linear measurements U1-NA, ANS-PNS, U1-NPOG, LI-NPOG, and N-ME showed higher values in both photographed and scanned versions as compared to the original and one measurement value of L1-NB was less in both photographed and scanned versions.
For the six angular cephalometric readings, the total mean value for all the six angular measurements SNA, SNB, ANB, U1-SN, L1-MP, and MX-MP put together was 400.25° for photographed and 399.95° for scanned groups as compared to the original value which was 399.25°. Hence, it showed that the angular measurements of both photographed and scanned versions are comparable to the original, though both photographed and scanned groups showed slightly higher total mean values for the 6 parameters as compared to the original which was in agreement with the results of Bruntz et al.,  and Collins et al. 
No direct comparison of our results of distortion/magnification following conversion of analogue cephalograms to digital format by scanning or photographing was possible as no direct data was available. Collins et al.,  compared the measurements between photographed lateral cephalograms and scanned cephalograms only and Bruntz et al.,  made comparison of scanned lateral cephalograms with corresponding original radiographs.
| Conclusions|| |
- Distortion was found when the original cephalograms were converted to digital format whether by scanning on a flatbed scanner or when photographed with a digital camera.
- Linear and angular measurements of both scanned and photographed versions showed higher mean values as compared to original hand traced cephalograms.
- Measurements of scanned images were more comparable to the original cephalograms than the photographed images.
- Among the linear and angular measurements, linear measurements showed significant variation than angular measurements.
Caution has to be exercised when linear measurements from the photographed images are used for cephalometric analysis.
| References|| |
|1.||Roden-Johnson D, English J, Gallerano R. Comparison of hand-traced and computerized cephalograms: Landmark identification, measurement, and superimposition accuracy. Am J Orthod Dentofacial Orthop 2008;133:556-64. |
|2.||Chen YJ, Chen SK, Chang HF, Chen KC. Comparison of landmark identification in traditional versus computer-aided digital cephalometry. Angle Orthod 2000;70:387-92. |
|3.||Baumrind S, Frantz RC. The reliability of head film measurements. 2. Conventional angular and linear measures. Am J Orthod 1971;60:505-17. |
|4.||Chen YJ, Chen SK, Yao JC, Chang HF. The effects of differences in landmark identification on the cephalometric measurements in traditional versus digitized cephalometry. Angle Orthod 2004;74:155-61. |
|5.||Bruntz LQ, Palomo JM, Baden S, Hans MG. A comparison of scanned lateral cephalograms with corresponding original radiographs. Am J Orthod Dentofacial Orthop 2006;130:340-8. |
|6.||Collins J, Shah A, McCarthy C, Sandler J. Comparison of measurements from photographed lateral cephalograms and scanned cephalograms. Am J Orthod Dentofacial Orthop 2007;132:830-3. |
|7.||Siersbaek-Nielsen S, Solow B. Intra- and interexaminer variability in head posture recorded by dental auxiliaries. Am J Orthod 1982;82:50-7. |
|8.||Showfety KJ, Vig PS, Matteson S. A simple method for taking natural-head-position cephalograms. Am J Orthod 1983;83:495-500. |
|9.||Lundström F, Lundström A. Natural head position as a basis for cephalometric analysis. Am J Orthod Dentofacial Orthop 1992;101:244-7. |
|10.||Preston CB, Evans WG, Todres JI. The relationship between ortho head posture and head posture measured during walking. Am J Orthod Dentofacial Orthop 1997;111:283-7. |
|11.||Raju NS, Prasad KG, Jayade VP. A modified approach for obtaining cephalograms in the natural head position. J Orthod 2001;28:25-8. |
|12.||Rogers MB. Duplication of X-rays by scanning. J Clin Orthod 2002;36:208-9. |
|13.||Pereira AL, De-Marchi LM, Scheibel PC, Ramos AL. Reproducibility of natural head position in profile photographs of children aged 8 to 12 years with and without the aid of a cephalostat. Dental Press J Orthod 2010;15:65-73. |
|14.||Marcì V, Wenzel A. Reliability of landmark recording on film and digital lateral cephalograms. Eur J Orthod 1993;15:137-48. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10]
[Table 1], [Table 2], [Table 3], [Table 4]