|Year : 2022 | Volume
| Issue : 1 | Page : 68-74
Reliability and reproducibility of measurements in cephalometric radiography acquired by a charge-coupled device imaging system
Ala Mohamed Abdelrahim, Amal Hussein Abuaffan
Department of Orthodontics, Pedodontics and Preventive Dentistry, Faculty of Dentistry, University of Khartoum, Khartoum, Sudan
|Date of Submission||12-Mar-2022|
|Date of Decision||18-Apr-2022|
|Date of Acceptance||12-May-2022|
|Date of Web Publication||23-Jun-2022|
Amal Hussein Abuaffan
Department of Orthodontics, Pedodontics and Preventive Dentistry, Faculty of Dentistry, University of Khartoum, P. O. Box 1719, Khartoum
Source of Support: None, Conflict of Interest: None
Objective: To evaluate and compare the reliability, reproducibility, and speed of two cephalometric tracing methods computer-aided cephalometric tracing and manual tracing. Materials and Methods: This was an analytical, cross-sectional study. One hundred and three pretreatment cephalometric radiographs were randomly selected from the orthodontics department of a public university. Twelve cephalometric landmarks were identified, and fifteen measurements were calculated both manually and digitally using Vistadent OC software. The reliability of measurements was assessed for each method by applying the intraclass correlation coefficient (ICC). Paired t-test was used to compare the measurements' reproducibility and time difference between the two methods. Results: All angular and linear measurements for both the methods showed a range of moderate correlation (0.8 ≥ ICC ≥0.5) to strong correlation (ICC ≥0.8) except for L1-MAD, which displayed a poor correlation for both manual and digital tracing, (ICC = 0.36 and 0.33, respectively), as well as digital tracing of interincisal angle (ICC = 0.36). No statistically significant differences between the two methods were observed for all angular and linear measurements except upper anterior facial height (UAFH) and lower anterior facial height (P = 0.000). There was a statistically significant time difference between the two techniques (P = 0.000). The mean tracing time of the operator for single tracing was 18.02 min for manual tracing and 8.85 min when using the Vistadent program. Conclusion: Cephalometric measurements in conventional manual and digital cephalometric analysis are highly reliable. Although the reproducibility of some measurements between two methods showed statistically significant differences, most differences were considered minimal and clinically acceptable. Computer-assisted cephalometric analysis proved to be more time-efficient.
Keywords: Cephalometric analysis, conventional tracing, digital tracing
|How to cite this article:|
Abdelrahim AM, Abuaffan AH. Reliability and reproducibility of measurements in cephalometric radiography acquired by a charge-coupled device imaging system. J Head Neck Physicians Surg 2022;10:68-74
|How to cite this URL:|
Abdelrahim AM, Abuaffan AH. Reliability and reproducibility of measurements in cephalometric radiography acquired by a charge-coupled device imaging system. J Head Neck Physicians Surg [serial online] 2022 [cited 2022 Jun 28];10:68-74. Available from: https://www.jhnps.org/text.asp?2022/10/1/68/347983
| Introduction|| |
Cephalometric radiography is a critical tool for orthodontic diagnosis, treatment planning, prediction of growth, surgical procedures, and research purposes.
The manual tracing method was used for many years. Although there seems to be a consensus among orthodontists who attest to its lower cost and ease of application, this method presented some drawbacks: time consumption, being subjected to technical errors in landmark identification, the need for additional physical storage, film decay, and loss of information.,
The rapid development of computer-aided radiography appealed to many orthodontic professionals, who gradually progressed to digital tracing. This started indirectly using digitizing pads, followed by video or digital camera photographs, and finally evolved to scanned radiographs. These methods offer the advantages of a displayed image, which simplifies image manipulation, teleradiography, duplication, and superimposition.,, However, these techniques require film processing and have associated chemical hazards, whereby more errors arise with each step of digitization and requires more time to process.
The introduction of new direct digital forms provides the advantages of instant image access to databases and reduction of radiation exposure and avoids the need for film processing, along with the benefits of being capable of viewing older digital images. Gaining direct digital image can be achieved either using a semidirect technique using an intermediate (SP) storage phosphorus plate or by means of a charge-coupled device (CCD) and a direct sensor technique.
Although images received by the CCD technique have avoided the previously mentioned drawbacks, there is still a shortage in the literature surrounding the efficiency of different cephalometric programs with this image-capturing mode. Therefore, the aim of this study was to compare the reliability and reproducibility of the cephalometric measurement using the Vistadent software program of a radiograph captured by CCD over the manual tracing method. In addition, it will address the time factor between the two methods.
| Materials and Methods|| |
This study was conducted on pretreatment digital cephalometric radiographs (hard copies and soft copies) of 103 randomly selected patients from the orthodontics department of a public university. The inclusion criteria were as follows:
- Good-quality pretreatment radiographs
- All permanent teeth must be present, with the exception of the third molars
- No impacted or partially erupted teeth
- No facial asymmetry or craniofacial defects.
All lateral cephalographs were taken in standardized conditions using the same machine (PROMAX 3D, D-054SB-C, 2014, PLNMECA OY, 00880 Helsinki, Finland). For manual tracing, each radiograph was overlaid with a transparent 0.003-inch acetate paper and attached to the hard copy printout with a duct tape. Manual tracing was then carried out using a 0.5-mm 3H pencil on a viewer box in a dark room. When visual inspection revealed a lack of superimposition between bilateral structural outlines, the average between the two structures was recorded.
As for digital tracing, soft copies of the cephalograms were transferred to a compact disk (CD) and imported to a personal computer loaded with Vistadent O. C. (TM – version 4.2.61 (177) 2006 – GAS International, Inc.) software. The images were calibrated by digitizing two fiducial points (10 mm apart) on the calibration ruler contained within the program's digital cassette. The observer utilized various image enhancement features, e.g., brightness, contrast, and magnification to identify cephalometric landmarks as accurately as possible with cursor-hair locator controlled by a mouse. Afterward, measurements were automatically registered by the Vistadent program, and the results were printed out for statistical evaluation.
A total of twelve cephalometric landmarks and four reference planes were identified on each radiograph [Figure 1] and [Figure 2]; fifteen measurements were calculated by the same operator [Figure 3] and [Figure 4]. The time consumed during manual and digital tracing processes was assessed with a digital chronometer and recorded on the data collection sheet for statistical analysis.
|Figure 1: Reference landmarks: 1: Nasion (Point N): The most prominent point of the frontonasal suture. 2: Sella (Point S): The midpoint of the cavity of Sella turcica. 3: Anterior nasal spine (ANS): The tip of the anterior nasal spine. 4: Posterior nasal spine (PNS): The tip of the posterior nasal spine. 5: Point (A): The innermost point on the contour of the premaxilla between anterior nasal spine and the incisor tooth. 6: Point (B): The innermost point on the contour of the mandible between the incisor tooth and the bony chin. 7: Pogonion (Pog): The most anterior point on the contour of the chin. 8: Menton (Me): The most inferior point on the mandibular symphysis. 9: Gonion (Go): The midpoint of the contour connecting the ramus and body of the mandible. 10: Pronasale (Pn): The most protruded point of the nasal tip. 11: Subnasale (SN): The point at which the columella merges with the upper lip in the midsagittal plane. 12: Soft tissue pogonion (Pog): The most prominent point of the soft tissue chin in the midsagittal plane|
Click here to view
|Figure 2: Reference plane:. 1: SN Plane: The line joins sella and nasion. 2: Maxillary plane: The line joins ANS and PNS points. 3: Functional occlusal plane: The line passes through the cusps of the upper and lower first molars and/or premolar teeth.. 4: Mandibular plane: The line joins Me to Go points|
Click here to view
|Figure 3: Angular measurements: 1: SNA: The angle between point S, point N, and point A. 2: SNB: The angle between point S, point N, and point B. 3: ANB: The angle between point A, point N, and point B.. 4: SN-MAX: the angle between SN and maxillary plane. 5: U1-Max: The angle between upper incisors and maxillary plane. 6: L1-MAD: The angle between the lower incisors and mandibular plane. 7: Interincisor angle: The angle between upper and lower incisors. 8: MMPA: The angle between The maxillary plane and mandibular plane. 9: Nasolabial angle: Is formed by two lines, a columella tangent and upper lip tangent|
Click here to view
|Figure 4: Linear measurements:. 1: Upper facial height (UFH): The distance between N to maxillary plane. UFH: Upper facial height. 2: Lower facial height (UFH): The distance between Me to maxillary plane. UFH: Upper facial height. 3: Lower Anterior Facial height proportion (LAFH %) = LAFH. UAFH + LAFH. UAFH: Upper anterior facial height, LAFH: Lower anterior facial height|
Click here to view
Ethical approval and consent to participate
The ethical approval was attained from the ethics research committee, faculty of medicine.
Informed consent from the participated patients was routinely obtained and drafted in the orthodontic department, faculty of dentistry.
Statistical analyses were performed using (SPSS) Statistical Package for the Social Sciences, Version 220 software (IBM Corp, Armonk, NY). To determine the intraexaminer reliability, 15% of the total samples (n = 16) were randomly selected, retraced, and measured both manually and digitally by the same examiner the second time, with a 2-week interval between both the sessions. To evaluate this reliability, the intraclass correlation coefficient (ICC) was applied [Table 1]. These levels were used to determine the strength of the correlation: ICC ≥0.8 = strong, 0.8≥ ICC ≥0.5 = moderate, and ICC ≤0.5 = weak. The reproducibility and time difference of the measurements were calculated using the paired t-test, and the level of significance was adjusted by the number of tested measures according to the Bonferroni criteria.
| Results|| |
A correlation analysis for angular measurements by manual and digital tracings revealed that all angular measures showed moderate (0.8≥ ICC ≥0.5) to strong correlation (ICC ≥0.8) except for (L1-MAD) lower incisor-to-mandibular plane angle manual measurement, which revealed a weak correlation (ICC = 0.360), L1-MAD digital measurement (ICC = 0.33), and interincisal digital measurement (ICC = 0.36). A high correlation was observed for all linear manual and digital measurements (ICC ≥0.8) except for lower anterior facial height proportion (LAFH%) manual measurement, which presented only moderate correlation (ICC = 0.76).
Comparison of methods: Hand tracing versus computerized measurements for angular measurements [Table 2].
No statistically nor clinically significant difference was noted among all angular measurements. The highest magnitude of mean difference among angular measurements was 1.35° for (SN-MAX) sella–nnasion line and maxillary plane angle.
Comparison of methods: hand-tracing versus computerized measurements for linear measurements [Table 3].
Wits appraisal perpendiculars from points A and B on the maxilla and mandible, respectively, onto the occlusal plane, LL/E distance between lower lip and rickets E-line.
There was a statistically significant difference for upper anterior facial height (UAFH) and LAFH (P = 0.000) and a mean difference exceeding 2 mm for LAFH (mean difference = 3.12 mm).
Time difference [Table 4].
A statistically significant difference was detected between the two tracing methods in time, with a mean difference of 9.17 min.
| Discussion|| |
The global rise in the use of digital cephalometric software programs has stressed the need to assess their efficiency. Practitioners often face difficulties in choosing reliable and precise technology without access to proper studies.
Although numerous studies have evaluated the performance of different cephalometric software programs, most compared the cephalometric analysis of either scanned or photographed images to their equivalent hard copy or compared soft copies to their analog hard copies where cephalograms were obtained using the sandwich hybrid technique.,,,,,
To our knowledge, this cross-sectional study appeared to be the first to test the accuracy of the Vistadent OC software program using the direct digital image capture with a suitable global sample size. Although one Turkish study by Polat-Ozsoy et al. tested the program's precision with the same setup using only 30 radiographs, the efficiency of the Vistadent software program had not been tested regionally.
Direct digital images were chosen in this study to exclude any errors that may arise due to film processing, scanning, and imaging procedures, which may lead to vertical and horizontal distortion. However, it was unfeasible to use the hybrid sandwich technique in which digital and conventional radiographs are captured simultaneously.
In this study, measurements were used instead of landmarks because they are the end product of tracing and offer data for treatment planning. Moreover, this was recommended by Santoro et al. They pointed out that any analysis that aimed to reveal the consistency of digital cephalometric should concentrate on the use of measurements as an alternative to landmarks, as well as the sources of error. Landmark identification is considered the main source of errors, which is mostly linked to operator experience.
Reliability of angular manual measurements
The excellent and moderate correlation of the manual angular measurements recorded in this study indicates that the investigator faced minimal to no difficulty in correctly repeating these measures, except for the L1-MAD manual measurement, which expressed a weak correlation (ICC = 0.36). This finding could be attributed to the weak reliability of some landmarks included in the measurement, especially incisor root apex and the midpoint (gonion) of the contour connecting the ramus and body of the mandible (Geelen et al., 1998). In addition, the weak reliability of locating lower incisor landmarks could be attributed to the operator's lack of knowledge. Another contributing factor is the augmentation error in locating gonion due to its weak point definition, where the ramus and body of the mandible meet.
Chen et al. reported a significant difference in gonion localization, both horizontally and vertically, and stated that “the extent of difference for each landmark depends on the radiographic complexities, which are also associated with the reliability of landmarks.” In addition, Chen et al. classified gonion, (Porion) the midpoint on the upper margin of the external auditory canal, (Orbitale) the most inferior point on the infraorbital margin, (menton) the most inferior point on the mandibular symphysis.
And lower incisors apex as inconsistent and unreliable landmarks, irrespective of the method used as different reference planes may be constructed to locate the innermost point of a curve. Another study carried out by Gregston et al. justified the high degree of variation of the incisors by the overlapping of multiple incisors roots, thereby making the precise identification of their apices more challenging.
In this study, the lower incisor landmark was used in the manual tracing of interincisal angle and the distance between lower incisor tip to A-pog (L1-Apog). Interestingly, the measurements showed moderate-to-strong reliability (ICC = 0.768 and 0.981, respectively). Similarly, gonion and menton landmarks, which had been utilized to construct the mandibular plane used for maxillomandibular plane angle (MMPA) manual measurement showed a good correlation (ICC = 0.785).
Santoro et al. pointed out that studying angular and linear measurements with many landmarks is a more complicated procedure than the study of a single landmark. The difference in landmark location used in combination might nullify each other or increase the magnitude of discrepancy.
Reliability of angular digital measurements
The weakest angular digital correlations that stood out in this study were L1-MAD (ICC = 0.33) and interincisal angle (ICC = 0.36). Furthermore, the digital reliability of these measures appeared to be inferior to the manual one L1-MAD (ICC = 0.36) and interincisal angle (ICC = 0.76).
The weaker correlation of L1-MAD digital measurement could be linked to the difficulty in locating the gonion point digitally, as proposed by Santoro et al. Moreover, this point is related to poorly defined outlines correlated with bilateral anatomic structures. Although the location of this landmark can be simplified using the bisecting angle of the posterior border of the ramus and mandibular plane tangents, unfortunately, this could not be achieved using the Vistadent program.
In terms of the weak reliability of the interincisal angle digital measurement, Chen et al. reported that the accurately pinpointing the location of the maxillary and mandibular incisors root apices by digital means is challenging. In addition, Vistadent automatically superimposes the outline of incisors over the image once the crown tip and apex are located. This outline obscures most of the facial aspect of the incisor.
Reliability of manual and digital linear measurements
While the UAFH and LAFH are included in the calculation of LAFH%, all three measurements revealed an excellent agreement in both the tracing methods. Manual tracing of LAFH% showed a moderate correlation (ICC = 0.76). To the extent of our knowledge, no previous studies have tested neither the reliability nor the reproducibility of the LAFH% measurement.
Reproducibility of angular and linear measurements
With respect to comparing the angular measurement reproducibility between manual and digital methods, the study showed no statistically nor clinically significant difference among all angular measurements. However, linear measurement reproducibility revealed a statistically significant difference for only two linear measurements: UAFH and LAFH. Furthermore, the mean difference of LAFH measurement exceeded the cutoff from the mean of manual measurement (mean difference = 3.12 mm).
The results from a study conducted by Celik et al. supported the current results with a statistically significant difference in LAFH and a mean difference exceeding the cutoff (mean difference = 3.48). Their study considered the minimal and clinically acceptable excess relating the difference to low menton landmark reproducibility. A similar observation on different measurements was reported by Gregston et al., revealing a statistically significant difference reported for (FH-GoMe) Frankfort horizontal plane and mandibular plane angle and (L1-NB) lower incisors to Nasion-point B angle, with mean difference exceeding the cutoff and considered as clinically reliable.
The current study showed an excellent reliability for UAFH and LAFH measurements in both the tracing methods, separately (ICC >0.8), concluding that the measurements were consistently reproduced and that the landmarks' location was uncomplicated. The noncorrespondence between manual and digital UAFH and LAFH values could be due to variations in reading the image when displayed on the screen, as the operator identified anatomic structures differently in the digital sets, even if he/she could reproduce them consistently in each set.
The UAFH and LAFH statistical difference could be attributed to the effects of the flashing cursor (used for digitization), either by failing to contrast with the background making the landmarks imprecise or by obscuring certain landmarks.
Angular versus linear reproducibility
The current study revealed a statistically significant difference between linear measurements of the 6 measurements (upper facial height and lower facial height), while no angular measurement showed a statistical difference. A similar observation was previously noted by Celik et al. and Tikku et al.
Manual versus digital time difference
The study time table showed that the time taken by the operator to perform manual tracing was nearly twice the amount of time required for digital tracing. The mean tracing time for a single tracing was 8.85 min for Vistadent versus 18.02 min for manual tracing. Uysal et al. confirmed this finding, whereby the manual time consumed by the two examiners was nearly twice that of the digital method. Moreover, Chen et al. also found that the time needed for tracing cephalometric measurements could be reduced using the digital technique.
| Conclusion|| |
- Cephalometric measurements showed high-to-moderate reliability for both the tracing methods, except for L1-MAD and interincisal angle; both are dependent on low reproducibility anatomical points (incisors apexes, gonion, and menton)
- Although statistically significant differences were noted between the two techniques for UAFH and LAFH, the differences were considered minimal and clinically acceptable
- Computer-assisted cephalometric analysis provides time benefit to clinicians
- Computer-aided cephalometric analysis can be considered a safe and valuable method of cephalometric analysis.
The main data in this paper will be shared on reasonable demand of the main author as well as corresponding author.
This material has never been published and is not currently under evaluation in any other peer reviewed publication.
The permission was taken from the Institutional Ethics Committee before starting the project. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent was obtained from all individual participants included in the study.
We acknowledge the assistants of the orthodontic department college of dentistry University of Khartoum, Sudan for their help in collection of cephalometric radiograph.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Albarakati SF, Kula KS, Ghoneima AA. The reliability and reproducibility of cephalometric measurements: A comparison of conventional and digital methods. Dentomaxillofac Radiol 2012;41:7-11.
Chen SK, Chen YJ, Yao CJ, Chang HF. Enhanced speed and precision of measurement in a computer-assisted digital cephalometric analysis system. Angle Orthod 2004;74:501-7.
Forsyth DB, Shaw WC, Richmond S. Digital imaging of cephalometric radiography, Part 1: Advantages and limitations of digital imaging. Angle Orthod 1996;66:37-42.
Tikku T, Khanna R, Maurya RP, Srivastava K, Bhushan R. Comparative evaluation of cephalometric measurements of monitor-displayed images by Nemoceph software and its hard copy by manual tracing. J Oral Biol Craniofac Res 2014;4:35-41.
Polat-Ozsoy O, Gokcelik A, Toygar Memikoglu TU. Differences in cephalometric measurements: A comparison of digital versus hand-tracing methods. Eur J Orthod 2009;31:254-9.
Brennan J. An introduction to digital radiography in dentistry. J Orthod 2002;29:66-9.
Houston WJ. The analysis of errors in orthodontic measurements. Am J Orthod 1983;83:382-90.
Riffenburgh RH. Statistics in Medicine. 3rd
ed. San Diego: Academic Press; 2012.
Erkan M, Gurel HG, Nur M, Demirel B. Reliability of four different computerized cephalometric analysis programs. Eur J Orthod 2012;34:318-21.
Gregston MD, Kula T, Hardman P, Glaros A, Kula K. A comparison of conventional and digital radiographic methods and cephalometric analysis software: I. hard tissue. Semin Orthod 2004;10:204-11.
Oliver RG. Cephalometric analysis comparing five different methods. Br J Orthod 1991;18:277-83.
Santoro M, Jarjoura K, Cangialosi TJ. Accuracy of digital and analogue cephalometric measurements assessed with the sandwich technique. Am J Orthod Dentofacial Orthop 2012;
Uysal T, Baysal A, Yagci A. Evaluation of speed, repeatability, and reproducibility of digital radiography with manual versus computer-assisted cephalometric analyses. Eur J Orthod 2009;31:523-30.
Bruntz LQ Palomo JM, Baden S, Hans MG. A comparison of scanned lateral cephalograms with corresponding original radiographs. AM J Orthod Dentofacial Orthop 2006;130:340-7.
Geelen W, Wenzel A, Gotfredsen E, Kruger M, Hansson LG. Reproducibility of cephalometric landmarks on conventional film, hardcopy, and monitor-displayed images obtained by the storage phosphor technique. Eur J Orthod 1998;20:331-40.
Baumrind S, Frantz RC. The reliability of head film measurements. 2. Conventional angular and linear measures. Am J Orthod 1979;60:505-17.
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.
Celik E, Polat-Ozsoy O, Toygar Memikoglu TU. Comparison of cephalometric measurements with digital versus conventional cephalometric analysis. Eur J Orthod 2009;31:241-5.
Turner PJ, Weerakone S. An evaluation of the reproducibility of landmark identification using scanned cephalometric images. J Orthod 2001;28:221-9.
Power G, Breckon J, Sherriff M, McDonald F. Dolphin imaging software: An analysis of the accuracy of cephalometric digitization and orthognathic prediction. Int J Oral Maxillofac Surg 2005;34:619-26.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4]