Download Gutierrez_M_A_Compar..

A Comparison of the Accuracy of Cephalometric Analysis Using Conventional Hand Tracing vs. Computer Aided Analysis
M. A. Gutierrez, DDS, A. L. Shearer, DMD, R. Kleefield, DDS, J. Oppenheimer, DDS
St. Barnabas Hospital, Department of Dentistry, Division of Orthodontics, Bronx, NY
Abstract
The purpose of this study was to evaluate the accuracy and reproducibility of
cephalometric analysis using the most current versions of Vistadent® and
Dolphin® software, as compared to hand tracing. Due to the recent advances
in the new versions of these programs, we expect these methods to be as
accurate and predictable as the traditional manual tracings. Twenty lateral
cephalometric radiographs were examined by two operators to evaluate the
most current Vistadent® and Dolphin® software, as compared to hand
tracing. For the 660 angular and 480 linear measurements made there were
no statistical differences between conventional tracings and either Vistadent
3.1® or Dolphin 9®. The most current Vistadent® and Dolphin® software as
compared to conventional hand tracing do provide accurate and reproducible
analysis
Figure 5
A
B
Figure 3. A. Nasion point. B. Point A
Introduction
Broadbent introduced cephalometric radiology in 1931. In 1948 Downs introduced
the Downs analysis which is often cited as the first true orthodontic analysis. Since
then, orthodontic cephalometric analysis have been an integral part of orthodontic
diagnosis and treatment planning. Manual tracing has been regarded as the
standard for measuring cephalometric relationships. The manual tracing method is
time intensive and results in pencil tracings which are not very presentable and
difficult for the patient to understand during treatment planning sessions. (Fig 1 A).
A recent study by Dana and Goldstein in the May 2004 issue of the JCO compared Dolphin 8® and
VistaDent 9.0AT® to conventional manual tracings (4). The authors found no statistically significant
differences except in the Wits appraisal and the mandibular plane angle(4). Dolphin automatically
placed Gnathion point more superior than where the authors would have preferred affecting the
Mandibular plane(4). Because the Wits appraisal relies on construction of the occlusal plane users can
expect these digitized values to be unreliable if constructed automatically. With the use of modern
computer software Orthodontists are now able to quickly and efficiently compose multiple cephalometric
analyses. Because of the increased need of evidence based orthodontics, the accuracy and
reproducibility of various computer software programs in orthodontic cephalometric analysis should be
evaluated. The purpose of this study is to evaluate the accuracy and reproducibility of cephalometric
analysis measurements using the most current Vistadent®, and Dolphin® software, as compared to
hand tracing. Due to the recent updates in the new versions of these programs, we expect these
methods to be as accurate and predictable as the traditional manual tracings.
Materials and Methods
Figure 1 A. Pencil Tracing
Digitized cephalometric tracings allow for multiple analyses to be evaluated quickly
and efficiently. Presentation modes in the software are patient-education friendly
during treatment planning presentations. (Fig 2 A)
Thirty lateral cephalometric radiographs were picked at random from an orthodontic private practice.
The twenty best films were chosen for the study. A digital cephalometric orthodontic analysis was
composed using the VistaDent 3.1® and Dolphin 9®. Eleven well known angular measurements and
eight linear tracings from Downs, Ricketts, Steiner and Wits analyses where chosen. Hand tracings of
the cephalometric radiographs on acetate paper using a .5mm #2 pencil on a high quality light box in a
reduced light room were constructed. To evaluate inter-examiner error, 5 radiographs not included in
the study were traced by each operator at different times. No statistical differences were found
between the two operators (p>.05). For the computer software analysis, the radiographs were
scanned and calibrated into the computer. The analysis included the following measurements:
Linear:
WITS, U1-NA(mm),L1NB(mm),L1-Apo, E-line,N-ANS,ANS-GN,N-GN
Angular:
SNA,SNB,ANB,Y-AXIS,SN-MP,U1-SN,U1-NA,L1-MP,L1-NB,U1/L1,NLA
Data was then be analyzed statistically using a paired student t test.
Results
For the 660 angular measurements made there were no statistical differences between conventional
tracings and either Vistadent or Dolphin programs (P>.05).(Fig 4). For the 480 linear measurements
made there were no statistical differences between conventional tracings and either Vistadent or
Dolphin programs (P>.05), (Fig 4,5)
Figure 4
Figure 2. Vistadent digital tracing.
Various authors have concluded in their studies that the reliability of landmark
identification in digital images was comparable to that in original radiographs except
for certain points (8). A study by Forsyth concluded that with the majority of angular
and linear measurements there is a systematic error between the digital images
and the conventional radiographs (6). Landmarks on poorly defined edges such as
nasion and A point appear to have the greatest error (Figure 3. A,B), (5).
Discussion
Dana and Goldstein et al found that Dolphin 8® was significantly different from Vistadent 9® and
manual tracings in measuring the mandibular plane (GoGn), due to the automatic construction of
Gnathion(4). We did not have this problem with the mandibular plane using Vistadent 3.1AT®
and Dolphin 9®, because in the updated versions the operators plot Gnathion. Several authors
did see significant differences in the Wits appraisal due to differences in occlusal plane
construction. Old versions constructed this plane automatically(4). Our study showed no
statistical differences for the Wits appraisal. In the current versions of Dolphin® and Vistadent®
the operator is allowed to plot the occlusal points for construction of the occlusal plane.
Conclusion
The most current Vistadent, and Dolphin software as compared to Conventional Hand tracing do
provide accurate and reproducible cephalometric analysis. Therefore, Orthodontists can rely on
either computerized imaging software programs to be as accurate and applicable as manual
methods, which leads to a better and more reliable diagnosis.
References
1. www.physics.csbsju.edu/statistics/unpaired student t-test.
2. Y.J. Chen, S.K. Chen, H.F. Chang, K. Chen, Comparison of Landmark Identification in
traditional Versus Computer-Aided Digital Cephalometry, AJODO, Vol 70., October 2000., 387392.
3. Broadbent, B.H.: A new x-ray technique and its applications to orthodontia, Angle Orthod. 1:45,
1931.
4. J.M. Dana, M. Goldstein, J.G. Burch, P.C. Hardigan, Comparative study of manual and
computerized cephalometric analysis. JCO, May 2004 293-96.
5. Y.J. Chen, S.K. Chen, C.K. Chen, The effects of differences in landmark identification on the
cephalometric measurements in traditional vs digitized cephalometry.
6. D.B. Forsyth, W.C. Shaw, S. Richmond., Digital imaging of cephalometric radiography, part 1:
advantages and limitations of digital imaging., The Angle Orthodontist, Vol 66 No. 1. 1996, 37-42.
7. D.B. Forsyth, W.C. Shaw, S. Richmond., Digital imaging of cephalometric radiography, part 2:
Image Quality., The Angle Orthodontist, Vol 66, No 1. 1996, 43-52
8. Naini F.B., Otasevic M., Vasir S.N.A, A comparison of Manual tracing, digitizing, and computer
cephalometric analysis, Virtual Journal of Orthodontics 2001-03-15.
9. William R. Proffit, Contemporary Orthodontics, Third Edition, 2000.