Download ACCURACY OF COMPUTER-ASSISTED CEPHALOMETRIC

Orthodontics
ACCURACY OF COMPUTER-ASSISTED CEPHALOMETRIC
MEASUREMENTS: A COMPARATIVE STUDY
Mihnea IACOB1, Sorana ROŞU2, Irina ZETU3
1, 3. Lecturer, PhD, Department of Orthodontics and Pedodontics, „Grigore T. Popa” University of Medicine and Pharmacy - Iasi
2. PhD Student, Department of Preventive Dentistry, „Grigore T. Popa” University of Medicine and Pharmacy - Iasi
Corresponding author: [email protected]
Abstract
Aim. The aim of the present study was to comparatively
evaluate the accuracy of the results obtained by manual
and computer-assisted tracing of a group of cephalometric
radiographs.
Methods. A group of 60 initial cephalometric
radiographs from 60 orthodontic patients, obtained on the
same radiological device, was selected. The cephalometric
radiographs were traced and measured using the classical
(manual) method and the Orthalis cephalometric software,
by the same examiner, using the parameters of Steiner and
Tweed analysis.
Results. With a few exceptions, the computer-assisted
tracing technique gave results with an accuracy similar to
that of the manual, clinically acceptable technique.
Conclusions. The cephalometric tracing sofware
represents a useful tool in both orthodontic diagnosis and
treatment processes.
Keywords : cephalometric radiographs, computer-assisted
measurements
1. INTRODUCTION
The
cephalometric
radiography,
simultaneously introduced by Broadbent and
Brodie in the USA, and by Hofrath in Europe,
allows the orthodontist to evaluate the direction
and quantity of the growth and development
processes, as well the modifications owed to the
different types of orthodontic treatment [1,2].
The most frequent interpretation technique
for cephalometric radiographs is the manual
procedure, involving the use of acetate paper
and manual marking of the cephalometric
landmarks and anatomical structures. The
measurements are made with a ruler and a
protractor. Although this technique has been
long time utilized by orthodontic practitioners,
it was observed that it presents an appreciable
degree of error (projection errors, landmark
identification and measurement errors) [3-6].
International Journal of Medical Dentistry
Steiner cephalometric analysis offers a series
of specific measurements, useful in orthodontic
diagnosis and treatment planning. Steiner also
measured the necessary compromise of incisors
position, as a function of intermaxillary
discrepancy [7].
Charles Tweed determined the final position
of the lower incisors on the mandible, determining
the Tweed diagnostic triangle, which is the main
element in total space analysis, performed to
evaluate the need of orthodontic extractions [8].
In the final decades of the XXth century, a new
tracing technique emerged, in which the
orthodontist benefits from computer assistance.
The cephalometric landmarks are identified by
the human operator, yet all measurements are
computerized [9].
The aim of the present study is to comparatively
evaluate the accuracy of manual and computerassisted cephalometric measurements, using the
parameters of Steiner and Tweed cephalometric
anaysis.
2. METHODS
A group of 60 initial cephalometric
radiographs, obtained from 60 untreated
orthodontic patients (38 females, 22 males), with
ages between 8 and 23 years, was randomly
selected. The radiographs were taken with the
same radiological device. The exclusion criteria
were: poor head position, impacted or absent
incisors, poor film quality.
All radiographs were traced on a light-box, in
a dark room, using 0.003 inch acetate paper and
a 0.5 mm pen. The contours of the soft tissues
and the anatomical structures were traced, and
the cephalometric landmarks were marked. For
35
Mihnea IACOB, Sorana ROŞU, Irina ZETU
each radiography, the parameters of Steiner and
Tweed analysis were measured using a ruler and
a protractor.
Prior to computer-assisted tracing, all
radiographs were digitized by scanning (UMAX
Powerlook 1000 scanner), at 300 DPI, 60 digital
images, JPEG format, 2548x3510 pixels thus
resulting. In the first stage, the digital images
were calibrated by marking two points on the
scale present in the radiography. The cranial and
soft-tissue landmarks were marked on-screen by
the same operator. The measurements were
taken with the Orthalis cephalometric software.
The obtained data were introduced into an Excell
(Microsoft, Seattle,USA) spread sheet, and
evaluated statistically with the SPSS version 17
(SPSS Inc., Chicago, Ilinois, SUA) software.
All 60 radiographs were retraced, both by the
manual and computer-assisted technique, by the
same operator, at a 30 day interval. No statistically
significant differences were identified between
the two tracings.
For each cephalometric measurement
(variable), the mean, standard deviation,
standard error mean, 95% confidence interval of
the difference were calculated. The t Student test
was used to compare the manual tracing and
computer-assisted mean values.
3. RESULTS
Evaluations of manual and computerized
tracing measurements by Steiner and Tweed
analysis are presented comparatively in tables 1, 2.
Table 1 – Differences between manual and computerized tracings, Steiner analysis
Paired Samples Test
Paired Differences
Pair 1
Pair 2
Pair 3
Pair 4
Pair 5
Pair 6
Pair 7
Pair 8
Pair 9
Pair 10
SNA - SNA_M
SNB - SNB_M
ANB - ANB_M
SND - SND_M
1NA mm - 1/NA mm
1/NB mm - 1/NB mm
1/NA grade - 1/NA grade
1/NB grade - 1/NB grade
PG/NB - Pg/NB
SNOcl - SNOcl_M
Mean
-1.700
-.850
-1.056
-.883
-.117
.033
.517
-.729
.200
-2.533
Std. Deviation
3.346
2.065
2.087
1.984
3.499
2.393
4.942
6.888
2.276
3.149
Std. Error
Mean
.432
.267
.284
.256
.452
.309
.638
.897
.294
.406
95% Confidence
Interval of the
Difference
Lower
Upper
-2.564
-.836
-1.384
-.316
-1.625
-.486
-1.396
-.371
-1.020
.787
-.585
.652
-.760
1.793
-2.524
1.066
-.388
.788
-3.347
-1.720
t
-3.935
-3.188
-3.717
-3.449
-.258
.108
.810
-.813
.681
-6.232
df
59
59
53
59
59
59
59
58
59
59
Sig. (2-tailed)
.000
.002
.000
.001
.797
.914
.421
.420
.499
.000
Table 2 – Differences between manual and computerized tracings, Tweed analysis
Paired Samples Test
Paired Differences
Pair 1
Pair 2
Pair 3
Pair 4
Pair 5
Pair 6
Pair 7
Pair 8
Pair 9
Pair 10
Pair 11
Pair 12
36
FMIA - FMIA_M
FMA - FMA_M
IMPA - IMPA_M
SNA - SNA_M
SNB - SNB_M
ANB - ANB_M
Ao/Bo - AoBo_M
Z - Z_M
UL - UL_M
TC - TC_M
Hp - Hp_M
Ha - Ha_M
Mean
-2.267
.933
2.745
-.450
.367
-.683
1.183
-2.983
2.000
-.200
10.383
13.400
Std.
Deviation
8.309
4.769
5.653
3.181
2.617
2.432
2.819
7.299
2.123
2.413
6.134
5.126
Std. Error
Mean
1.073
.616
.792
.411
.338
.314
.364
.942
.274
.312
.792
.662
95% Confidence
Interval of the
Difference
Lower
Upper
-4.413
-.120
-.299
2.165
1.155
4.335
-1.272
.372
-.309
1.043
-1.312
-.055
.455
1.912
-4.869
-1.098
1.451
2.549
-.823
.423
8.799
11.968
12.076
14.724
t
-2.113
1.516
3.468
-1.096
1.085
-2.176
3.251
-3.166
7.296
-.642
13.111
20.248
df
59
59
50
59
59
59
59
59
59
59
59
59
Sig. (2-tailed)
.039
.135
.001
.278
.282
.034
.002
.002
.000
.523
.000
.000
Volume 4 • Issue 2 April / June 2014 •
ACCURACY OF COMPUTER-ASSISTED CEPHALOMETRIC MEASUREMENTS: A COMPARATIVE STUDY
4. DISCUSSION
Accuracy represents a critical aspect of every
scientific measurement. In orthodontic clinical
situations, a degree of accuracy below 2°,
respectively 2 mm, is considered acceptable [10].
Investigations on the accuracy of both manual
and
computer-assisted
cephalometric
measurements have been developed by numerous
researchers [11-18].
The initial studies were focused on testing the
repeatability of each landmark position, the
landmark identification errors being the most
frequent.
More recent researches aim at comparing the
measurements results, rather than landmark
position [19-28].
A comparative evaluation revealed that, for
Steiner anlysis, five parameters: SNA, SNB, ANB,
SND and SNOcl angles, presented statistically
significant differences between the manual and
computer-assisted tracings. The differences seem
to be caused by the difficult identification of
point A, and especially Na (involved in all five
measurements), as also reported by other authors.
With the exception of the SNOcl angle, all other
measurementes evidenced differences below ±2º,
which are clinically acceptable.
Statistically significant differences for SNA,
SNB an ANB angles where also reported by
Gregston and McClure [7, 18, 29].
For Tweed analysis measuremennts, the
statistically significant differences involved :
FMIA, IMPA, ANB, Z angles and AoBo UL, Hp
and Ha distances.
The mean differences for FMIA, IMPA and Z
angles are higher than the clinical acceptance
interval.
The large differences reported for the linear
measurements might be caused by magnification.
These results agree with those reported by
Saynsu, Kublashvili, Uysal, Polat and Celik [9,
20, 21, 29].
The Orthalis computer-assisted tracing grants
a definite time economy, the operator assessing
that the time needed for computer tracing is only
20% of the time allocated for manual tracing. On
the other side, when working with Orthalis, the
operator must identify a considerable amount of
International Journal of Medical Dentistry
cranial and soft-tissue landmarks, 59 for Steiner
and 55 for Tweed analysis, respectively. Even
when all landmarks had been corectly drawn,
the automatically traced bony and soft-tissue
contours did not always perfectly superpose on
the real ones. It is not possible to mark two
landmarks placed at a distance smaller than 0.5
mm one from the other.
Previous comparative researches showed that
the two tracing techniques gave similar, clinically
acceptable results. For each cephalometric
landmark, a specific indentifing error pattern
exists, some landmarks (Na, Po, Ar, Ba, Go, Gn)
being more difficult to identify [4, 10, 13, 23, 26,
27, 30, 31].
5. CONCLUSIONS
1.The computer-aided tracing technique
measurements had an accuracy degree similar to
that of the manual technique. With some minor
exceptions, the results were clinically acceptable.
2. In the computer-assisted technique, Steiner
analysis parameters showed a higher accuracy
degree than Tweed analysis.
3.The cephalometric tracing sofware
represents a useful tool for both orthodontic
diagnosis and treatment processes. The main
advantages are time economy and elimination of
mathematical errors.
References
1. Broadbent B.H. (1931) A new X ray technique and
its application to orthodontics, Angle Orthod. 1:4566.
2. Hofrath H. (1931) Bedeutung der Rontgenfern und
Abstands Aufnahme fur die Diagnostik der
Keiferanomalien, Fortschr Orthod. 1: 231-258.
3. Chen J., Chen S., Chang H., Chen K. (2000)
Comparison of landmark identification in traditional
versus computer-aided digital cephalometry, Angle
Orthod. 70:387-392.
4. Chen J., Chen S., Chung J., Chang H. (2004) The
effects of differences in landmark identification on
the cephalometric measurements in traditional
versus digitized cephalometry, Angle Orthod. 74:155161.
5. Houston W.J.B (1983) The analysis of errors in
orthodontic measurements, American Journal of
Orthodontics. 83:382-390.
37
Mihnea IACOB, Sorana ROŞU, Irina ZETU
6. Houston W.J.B., Maher R.E., McElroy D., Sherriff M.
(1986) Sources of error in measurements from
cephalometric radiographs, European Journal of
Orthodontics. 8:149-151.
7. Steiner C. (1953) Cephalometrics for you and me,
Amer J Ortho. 39:729-755.
8. Tweed Ch. (1966) Clinical Orthodontics, The C. V.
Mosby Company, Saint Louis.
9. Athanasiou A.E. (1995) Orthodontic Cephalometry,
Mosby Edit.
10. Gregston M., Kula T., Hardman P., Glaros A., Kula
K. (2004) A comparison of conventional and digital
radiographic methods and cephalometric analysis
software : I.hard tissue, Seminars in Orthodontics.
10(3):204-211.
11. Houston W.J.B. (1979) The application of computer
aided digital analysis to orthodontic records,
European Journal of Orthodontics. 1:71-79.
12. Houston W.J.B. (1982) A comparison of the reliability
of measurement of cephalometric radiographs by
tracing and direct digitization, Sweed Dent J Suppl.
15:99-103.
13. Richardson A. (1981) A comparison of traditional
and computerized methods of cephalometric
analysis, European Journal of Orthodontics, 3(1):15-20.
14. Oliver R.G. (1991) Cephalometric analysis comparing
five different methods, Br. J. Orthod, 18:277-283.
15.Davis D.N., Mackay F. (1991), Reliability of
cephalometric analysis using manual and interactive
computer methods, British Journal of Orthodontics,
18:105-109.
16. Nimkarn Y., Miles P.G. (1995) Reliability of computer
generated cephalometrics, Int J Adult Orthodon
Orthognath Surg. 10(1):43-52.
17. Ongkosuwito E.M., Katsaros K., Hof M.A., Bodegom
J.C. (2002) The reproducibility of cefalometric
measurements : a comparison of analogue and
digital methods, European Journal of Orthodontics.
24:655-665.
18. McClure S., Sadowsky L., Ferreira A., Jacobson A.
(2005) Reliability of digital versus conventional
cephalometric radiology : a comparative evaluation
of landmark identification error, Semin Orthod.
11:98-110.
38
19. Uysal T., Baysal A., Yagci A. (2009) Evaluation of
speed , repetability, and reproducibility dgital
radiography with manual versus computer-assisted
cephalometric analyses, European Journal of
Orthodontics. 31:523-528.
20. Polat-Ozsoy O., Gokcelik A., Memikoglu U. (2009)
Differences in cephalometric measurements : a
comparison of digital versus hand tracing methods,
European Journal of Orthodontics. 31:254-259.
21. Celik E., Polat O., Memikoglu U. (2009) Comparison
of cephalometric measurements with digital versus
conventional cephalometric analysis, European
Journal of Orthodontics. 31:241-246.
22. Cohen A.M. (1984) Uncertainity in cephalometrics,
Br J Orthod. 11:44-48.
23. Angus D., Brown J. (1998) A comparison between
hand drawn and computer aided cephalometric
analysis, Am J Orthod Dentofacial Orthop. 112:368.
24.Jacobson A. (2003) The Witts appraisal of jaw
disharmony, Am J Orthod DentofacialOrthop,
November.124(5):470-479.
25. Haynes S, Chau N.Y. (1995) The reproducibility and
repeatability of the Witts Analysis, Am J Orthod
DentofacialOrthop. 107(6):640-647.
26. Baumrind S., Frantz R.C. (1971) The reliability of
head film measurements 1.landmark identification,
Am J Orthod Dentofacial Orthop. 60:111-127.
27. Baumrind S., Frantz R.C. (1971) The reliability of
head film measurements 2.conventional angular
and linear measures, Am J Orthod Dentofacial Orthop.
60:505-517.
28.Santoro M., Jarjoura K., Cangialosi T. (2006)
Accuracy of digital and analogue cephalometric
measurements assesed with the sandwich technique,
Am J Orthod Dentofacial Orthop. 129:345-51.
29.Saynsu K, Isik F, Trakyali G, Arun T (2007) An
evaluation of the errors in cephalometric measurements
on scanned cephalometric images and conventional
tracings, Am J Orthod. 29:105-108.
30.Stabrun A.E., Danielsen K. (1982) Precision in
cephalometric landmark identification, Eur J Orthod.
4:185-196.
31. Trpkova B., Major P., Prasad N., Nebbe B. (1997)
Cephalometric landmark identification
and
reproducibility : a meta analysis, Am J Orthod
Dentofacial Orthop. 112:165-70.
Volume 4 • Issue 2 April / June 2014 •