vandenberghe 2007 cone beam perio diagnosis

Vol. 104 No. 3 September 2007

ORAL AND MAXILLOFACIAL RADIOLOGY Editor: Allan G. Farman

Diagnostic validity (or acuity) of 2D CCD versus 3D CBCT-images for assessing periodontal breakdown

Bart Vandenberghe, DDS,a Reinhilde Jacobs, DDS, PhD, MS,b andJie Yang, DDS, MMS, MS, DMDc Leuven, Belgium, and Philadelphia, PAKATHOLIEKE UNIVERSITEIT LEUVEN AND TEMPLE UNIVERSITY

Objective. The objective of this study was to compare 2-dimensional intraoral digital images with 3-dimensional conebeam CT (CBCT) in assessment of periodontal bone levels and defects.Methods. Thirty periodontal bone defects of 2 adult human skulls (with soft tissue or substitute) were evaluated byusing intraoral digital radiography and CBCT. Digital radiographs were made with a size #2 CCD sensor and a 60-kVDC x-ray unit, with 0.28, 0.42, or 0.56 mAs as respective exposure settings. For CBCT, jawbone images were obtainedat 120 kV and 23.87 mAs. Periodontal bone levels and defects on both imaging modalities were assessed andcompared to the gold standard. Delineation of lamina dura, crater defect, furcation involvements, contrast, and bonequality were also analyzed.Results. Linear measurement deviations of periodontal bone levels from the gold standard ranged from 0.19 to 1.66mm for intraoral radiography versus 0.13 to 1.67 mm for CBCT. Accuracy was not significantly different between bothimaging modalities (P � .161). Intraoral radiography scored significantly better for contrast, bone quality, anddelineation of lamina dura, but CBCT was superior for assessing crater defects and furcation involvements (P � .018).Conclusions. CBCT images allowed comparable measurements of periodontal bone levels and defects as intraoralradiography. CBCT images demonstrated more potential in the morphological description of periodontal bone defects,while the digital radiography provided more bone details. These findings may offer perspectives for further studies onperiodontal diagnostics, prognostics, and presurgical planning with CBCT. (Oral Surg Oral Med Oral Pathol Oral
Radiol Endod 2007;104:395-401)
Intraoral radiography is the most common imagingmodality used for diagnosing periodontal bone defects.However, intraoral radiography is 2-dimensional (2D)

aPhD and MS Student, Oral Imaging Centre, School of Dentistry,Oral Pathology and Maxillofacial Surgery, Faculty of Medicine,Katholieke Universiteit Leuven; Division of Oral and MaxillofacialRadiology, Temple University School of Dentistry.bProfessor and Head, Oral Imaging Centre, School of Dentistry, OralPathology and Maxillofacial Surgery, Faculty of Medicine, Katho-lieke Universiteit Leuven.cAssociate Professor and Director, Division of Oral and MaxillofacialRadiology, Temple University School of Dentistry; Associate Pro-fessor, Department of Diagnostic Imaging, Temple University Schoolof Medicine.Received for publication Sep 21, 2006; returned for revision Jan 6,2007; accepted for publication Mar 13, 2007.1079-2104/$ - see front matter© 2007 Mosby, Inc. All rights reserved.
doi:10.1016/j.tripleo.2007.03.012
and the amount of bone loss can be underestimated dueto projection errors1-5 or observer errors in identifyingreliable anatomical reference points.4-6 Assessing pre-surgical bone levels and changes in post–periodontaltreatment often requires 3-dimensional (3D) informa-tion. A previous study showed that combination of 2Dwith 3D imaging provides a better preoperative assess-ment of the implant site.7 The introduction of digitalintraoral imaging and cone beam computed tomogra-phy (CBCT) may bring new potentials for periodontaldiagnosis and treatment planning.8

Intraoral digital imaging not only reduces radiationexposure,9-15 but also optimizes assessment of oralstructures, improving the accuracy of periodontal diag-nosis.10-16 Conventional computed tomography (CT)provides 3D information, but the dose remains quitehigh. The recent development of CBCT reduces this
radiation exposure significantly.17-19
395

OOOOE396 Vandenberghe et al. September 2007

Over the past 15 years there have been manypublications concerning the applications of digitalintraoral radiography, but few of these have dealtwith its validity to monitor periodontal bone le-sions.10,16,20-25 The same scenario applies to the useof CBCT for periodontal indications.17-19,26-29 Manyquestions regarding both digital intraoral imagingand CBCT need to be addressed: Are periodontalbone levels, lamina dura, and bone craters well vi-sualized on both imaging modalities? How accurateare these imaging techniques in assessment of thebone levels and defects? Can the availability of 3Dimages assist the diagnosis of the bone loss anddefects? Therefore, the overall aim of this study wasto validate applications of digital intraoral imagingand CBCT in determination of the periodontal boneloss and defects. We hypothesized that both digitalintraoral radiography and CBCT would allow accu-rate assessment of periodontal bone levels.

MATERIALS AND METHODSThirty periodontal bone levels or defects of 2 adult

human skulls, a cadaver head and a dry skull, wereevaluated by using intraoral digital radiography (CCD,Schick Technologies, Long Island City, NY) andCBCT (I-Cat, 12 bit, Imaging Sciences International,Hatfield, PA). The upper and lower jaws of the cadaverhead were fixed with 10% formalin and functioned as aclinical subject. The adult human dry skull was coveredwith a soft tissue substitute, Mix D,30 and used as asimulation.

For the intraoral protocol, the paralleling techniquewas applied in a standardized exposure set-up. A filmholding system (XCP, RINN Corp, Elgin, IL) was used.To obtain identical images, bite blocks were coveredwith waxed imprints of the anterior, premolar, andmolar regions (see Fig. 1). Images were obtained witha size #2 charged coupled device (CCD) intraoral dig-ital sensor and a direct current (DC) x-ray unit (Helio-dent DS, Sirona Dental Systems GmbH, Bensheim,Germany). Exposure settings were 60 kVp with 0.28,0.42, and 0.56 mAs respectively (40, 60, and 80 ms �7 mA). A rectangular (4 cm � 3 cm) collimator (Uni-versal Collimator, RINN Corporation) was used. Thefocal-film distance was 30 cm.

For CBCT, the occlusal plane of the jawbones waspositioned horizontally to the scan plane and the mid-sagittal plane was centered. The field-of-view (FOV) orthe beam diameter at the surface of the image receptor(beam height) was adjustable. The protocols were set tovisualize the entire jaws, giving between 54 and 159slices of 0.4-mm thickness (approximately between 20-and 60-mm beam height). Images were obtained at 120
kVp and 23.87 mAs with a typical voxel size of 0.4 mm
(see Fig. 2). Periodontal bone levels and defects visu-alized with both imaging modalities were assessed by 3observers (postgraduate students at the Oral ImagingCentre). Images were viewed in a darkened room on 3notebook computers (Sony Vaio VGN A417m; SonyBelgium, Zaventem) with 17-inch LCD monitors andthe same screen resolution (1440 � 900 pixels). In-traoral 2D images were displayed in a random orderwith the Emago advanced V.3.5.2. software (Oral Di-agnostic Systems [ACTA], Amsterdam, The Nether-lands), in Tagged Image File Format (TIFF). CBCTimages were viewed with the I-CAT software (XoranCAT V.2.0.21, Xoran Technologies Inc., Ann Arbor,MI). Linear bone level measurements were carried outon a panoramic view obtained from an oblique line onthe axial plane with a standard slice thickness of 5.2mm. Measurement tools on both programs were used toobtain the data. Delineation of lamina dura, defectdescription, contrasts, and bone quality were also ana-lyzed by the 3 observers, using an ordinal scale.

Physical measurements of the skulls were consideredas the gold standards for further accuracy assessment ofboth imaging modalities. For the cadaver jaws, the goldstandard was obtained after image acquisition, by flapsurgery to allow physical measurements using a digitalsliding caliper (Mitutoyo, Andover, Hants, UK). Forthe dry skull however, gold standards were obtainedbefore adding soft tissue substitute and image acquisi-tion. Mesial, central, and distal bone levels and bonecrater depths on the oral and vestibular sides of eachselected tooth were measured. Because of dehydrationof the dry skull, the faded cemento-enamel junction(CEJ) could not be used as a reference point as in the

Fig. 1. Standardized exposure set-up: dry skull with gutta-percha markers and waxed imprints before covering the jawswith soft tissue substitute.

formalin-fixed cadaver jaws. Therefore, radio-opaque

which

OOOOEVolume 104, Number 3 Vandenberghe et al. 397

gutta-percha fragments with a small central indentationwere glued onto the respective teeth to serve as stan-dardized fiducials. A group of 19 teeth and 30 sites,including linear defects, 3D craters, and furcation in-volvements, was selected for comparison and statisticalanalysis.

STATISTICAL ANALYSISBone levels of the selected sites, measured on the

digital intraoral images, were compared with the goldstandards. Exposure settings, imaging methods, andobservers were used as independent variables and bonelevels and defects as the dependent ones. The goldstandard was obtained by averaging the scores of 2observers. Intraclass correlation showed no observereffect for these scores.

The acquired data were first scanned for outliers andtested for normality. As normality could not be foundeven after transformation, nonparametric statistics wereused for the analyses.31,32 The observer effect wastested with the Kruskal Wallis test and showed no

Fig. 2. Digital x-ray images of molar region from the lower cC, CBCT coronal slice. D, CBCT panoramic view (oblique). Oand CBCT) and comparison was done to the gold standard,

significant difference among the 3 observers (P � .05).

A 15% repeat of measurements was done at an intervalof 2 weeks and a high reliability was found amongevery observer (interval of 0.986 to 0.997 with 95%confidence and a single measure intraclass correlationcoefficient of 0.987). Those measurements were thenaveraged for further calculations (see Table I). Allstatistical analyses were carried out using the absolutevalues of these measurements. For comparison between2D CCD technique and CBCT, it was justified to usethe Wilcoxon Signed Rank test32 since the measure-ments were done on different imaging modalities. Totest the optimal intraoral exposure settings for compar-ison, the Friedman ANOVA test32 was applied.

Furthermore, subjective analysis of 5 dependent vari-ables (lamina dura, bone quality, contrast, craters, andfurcation involvements) on both imaging modalitieswas conducted. They were scored with an ordinal scalefrom 0 to 3 (where 0 � lack of visibility, 1 � poor, 2� medium, 3 � good). The ordinal data were processedusing nonparametric statistics and yielded no observereffect. Again the same tests were used; however, when

jaw. A, Two-dimensional CCD image. B, CBCT axial slice.ers described the defects using both imaging modalities (CCDwas obtained after removing the soft tissues (center).

adaverbserv

comparing 2D images versus 3D CBCT, the Wilcoxon


Signed Rank32 test was not used for the variablescraters and furcation involvements. CBCT image datasets for these variables were interpreted by the observ-ers using extra planes to describe the bone defects.Related to the discrepancy between 2D and 3D datasets, the Mann Whitney test was used to test thesevariables.32 The statistical analyses were done withSPSS V.13.0. statistical software (SPSS Inc, Chi-cago, IL).

RESULTSLinear bone level measurements

Table I shows absolute linear measurement devia-tions of periodontal bone levels from the gold standard.The deviations for intraoral radiography ranged from0.19 to 1.66 mm and 0.13 to 1.67 mm for CBCT.Further analyses revealed no significant difference be-tween the 2 imaging modalities (P � .161, Table II).The currently applied range of exposure settings forintraoral radiography yielded no significant differencein accuracy performance (P � .425, Table II) as suchthat it was justified to compare all of these to CBCTimaging.

Since each observer made 30 intraoral radiographicmeasurements (3 settings), a total of 90 measurementswere obtained from all the observers. Among thesemeasurements, 48 (53%) were underestimated and 42(47%) were overestimated. For CBCT the 30 measure-ments (only 1 setting) had the same ratio of underesti-mation (16 of 30 measurements, 53%) and overestima-

Table I. Absolute differences between averaged obsereach exposure setting and both imaging modalities. Wunderestimations slightly predominated (53%) compare

Measurement on toothHeliodent

DS 0.56 mAs

Mandibular right first molardistal

0.70

Maxillary left centralincisor distal

1.03

Maxillary left lateral incisormesial

1.20

Maxillary left canine mesial 0.23Maxillary left canine distal 0.37Maxillary left first molar

mesial0.77

Maxillary left first molarcrater

0.26

Maxillary left first molarfurcation

0.64

Maxillary left second molarcrater

0.68

Mandibular left first molarcrater

1.63

tion (14 of 30 measurements, 47%) as intraoral

radiographs. Fig. 3 is a graphic representation of theexact unaveraged values in Table II, showing onlyminor difference between over- and underestimation.

Quality ratingData analysis of the 2 imaging modalities yielded a

significantly better outcome for the intraoral radio-graphic images regarding lamina dura, contrast, andbone quality. On the other hand, for the variables cra-ters and furcation involvements, the morphological de-scriptions of the periodontal defects were more clearlydepicted by using CBCT (P � .018). This implied thatCBCT was more accurate for 3D crater and furcationvisualization compared with intraoral digital imaging.Fig. 4 represents the average values scored for bothimaging modalities.

DISCUSSIONAs seen in the results, linear bone level measure-

easurements and gold standard of 10 selected sites foroking at the exact unaveraged values of these scores,verestimations (46%).Heliodent

DS 0.42 mAsHeliodent

DS 0.28 mAs I-CAT

0.40 0.20 0.20

0.76 0.99 0.13

1.33 1.23 0.42

0.49 0.19 0.180.04 0.31 0.181.17 0.67 0.26

0.46 0.19 0.96

1.14 0.94 0.17

0.65 0.82 1.67

1.66 1.66 0.37

Table II. Results of the Wilcoxon Signed Rank test forthe 2 different imaging modalities and of the Friedmantest for the different settings (the 3 different exposuretimes)

CCD�Heliodent DSvs I-CAT WilcoxonSigned Rank Test

Heliodent DS 40,60, and 80 msFriedman Test

Z –1.070 Chi square 1.712Exact Sig. 0.161 df 2.000

Asymp. Sig. 0.425

ver mhen lod to o

ments were similar with 2D intraoral digital and 3D

oxes),


CBCT images. Both imaging modalities had sameover- and underestimation rates for periodontal bonedefects. Bone craters and furcation involvements werebetter depicted on CBCT than on intraoral images. This

Fig. 3. Boxplot of exact differences between gold standard mI). The chart showed median (black line), interquartile rangeset-up or the CBCT data. The boxes range between –1 and 1Overall values were slightly more negative (especially the b

Fig. 4. Variable comparisons between 2D (CCD) and 3D(CBCT) imaging modalities. Observers found that the laminadura (LD) were well delineated on the CCD images and notvisible on CBCT (I-CAT). Contrast (CO) and bone quality(BQ) also scored better on the digital intraoral radiographs.Periodontal craters (CR) and furcation involvements (FU)were better visualized on CBCT.

could be because the CBCT provides multiplaner slices

and 3D information. However, because of the lowerresolution, CBCT scored less than the intraoral imagesin contrast, bone quality, and delineation of laminadura. This indicated that the current CBCT systemcould not replace intraoral radiography for periodontalassessment. In fact, combination of both imaging mo-dalities could benefit periodontal bone assessment andassist presurgical treatment planning.

Radiation dose is always a concern for using con-ventional CT. However, radiation dose of CBCT wasreported up to 15 times less than conventional CT.19

Recent studies reported that CBCT systems only re-quire 4 to 15 times the dose of a standard panoramicimage18 or only the dose of a film-based full-mouthradiographic examination (FMX).19 An FMX in theUnited States varies from 18 to 22 intraoral radiographswith a dose range of 13 to 100 �Sv.33,34 Effective doseof CBCT, starting at 36.9 �Sv, was in the range of theFMX.18,19 Furthermore, Scarfe et al.19 reported aboutdose reduction when using smaller FOV examinations.The 9-inch FOV of the I-CAT images (69 �Sv) shouldbe capable of visualizing both jaws and providing allnecessary information for periodontal treatment plan-

ements and observer linear bone level measurements (Tables), and extreme values within different settings of the CCDwhich represented minor differences from the gold standard.showing slightly more underestimation of bone loss.

easur (boxemm,

ning of implants. The images require 8 times the dose


of a standard panoramic image (1.9-11 �Sv). If moreinformation is required in a broader area, the 12-inchFOV (135 �Sv) should be used, but in that case theradiation dose would rise till 15 times a standard pan-oramic image dose.19

Because the radiation dose of CBCT is lower thanconventional CT, there is growing concern of its over-consumption and radiation safety. In our opinion, theuse of CBCT should still be carefully justified (diag-nostic benefit and risk to be balanced). The imagingsystem must be performed by experienced and trainedpractitioners. As low as reasonably achievable(ALARA) radiation safety principle must be followed.In the current study, a low exposure setting of CBCT(only 23.87 mAs and 0.4-mm voxel size) was used.More studies with a large sample size in the future willdetermine ideal exposure settings, which optimize theimage quality and lower the radiation exposure further.

The present study found that CBCT had a higherquality rating on bone crater and furcation involvementassessment, whereas contrast, bone quality, and delin-eation of lamina dura were rated lower than for digitalintraoral radiography. We would like to suggest that thecurrently tested model of CBCT should only be usedfor relatively more complex periodontal treatment plan-ning, such as prognostic planning and surgery of com-plex periodontal defects and potential use of dentalimplants.

Previous studies show that periodontal bone levelmeasurements are reproducible on film-based conven-tional radiography, while examiners’ agreement is notenhanced by using intraoral digital imaging systems.24

Nevertheless, the latter reduce radiation exposure andoffer potentials for image analysis, optimization, andquantification, such as contrast enhancement, periodon-tal filtering, and digital subtraction.8,12,13,20,21 Thesedynamic functions can aid periodontal diagnosis aswell; however, when compared with CBCT, digitalintraoral radiography is still a 2D technique with limi-tation of presenting 3D periodontal defects, particularlythe buccal and lingual aspects of bone loss.1-3 In thepresent study we actually attempted to reduce the radi-ation dose as much as possible while keeping fulldiagnostic capabilities to offer a clinically applicablecomparison to CBCT. The lowest settings applied (0.28mAs at 60 kV) were still able to visualize the periodon-tium with the same accuracy and thus these can befurther recommended for the present tube specifica-tions.

All linear measurements in this study were doneusing a standardized dry skull and a cadaver head, afterin vitro pilot-testing of the precision of the method. Astandardized repositioning and stabilization was guar-
anteed by an individually adapted stent material serving
as a rigid occlusal key during exposure. This setupallowed avoidance of any projection error and correctfiducial visualization. The standardized and reproduc-ible image of the reference points was confirmed by thehigh-accuracy scoring and the good intra- and interob-server agreement in the present report.

When defining accuracy in terms of clinical measure-ment, a certain discrepancy between actual bone leveland radiographically estimated bone level has to beadmitted and considered as clinically acceptable. Smallor big errors in locating the CEJ and the alveolar crestcan respectively lead to over- and underestimation ofdisease prevalence.4 This can even lead to inappropri-ate planning for further treatment or unnecessary sur-gery. Considering that a 0.5-mm discrepancy can beadmitted clinically,4-5 both 2D CCD and 3D CBCT areaccurate enough in, respectively, 60% and 67% of themeasures. A 1-mm discrepancy even leads to, respec-tively, 82% and 90% accuracy.

Last, validation of these imaging modalities has beendone by a comparison in detectability of anatomical orpathological features, but the final test will be how wellany of them will effect treatment decisions and treat-ment outcome.

CONCLUSIONCBCT allowed similar periodontal bone level mea-

surements as digital intraoral radiography. Bone cratersand furcation involvements were better depicted onCBCT, while contrast, bone quality, and details oflamina dura scored better on digital intraoral radiogra-phy. A selective use of both imaging modalities mightthus aid periodontal diagnosis and treatment planning.However, selection criteria are needed to define theconditions and specific indications for use of 2D and/or3D imaging modalities in periodontology.

The authors thank Robslyn Gorin, MA (StatisticalApplications Manager, Computer Services, TempleUniversity, Philadelphia, PA), for her help with thestatistical analyses.

REFERENCES1. Eickholz P, Hausmann E. Accuracy of radiographic assessment

of interproximal bone loss in intrabony defects using linearmeasurements. Eur J Oral Sci 2000;108:70-3.

2. Schliephake H, Wichmann M, Donnerstag F, Vogt S. Imaging ofperiimplant bone levels of implants with buccal bone defects.Clin Oral Implants Res 2003;14:193-200.

3. Zulqarnain BJ, Almas K. Effect of x-ray beam vertical angula-tion on radiographic assessment of alveolar crest level. Indian JDent Res 1998;9:132-8.

4. Brägger U. Radiographic parameters: biological significance andclinical use. Periodontol 2000 2005;39:73-90.

5. Mol A. Imaging methods in periodontology. Periodontol 20002004;34:34-48.

6. Benn DK. A review of the reliability of radiographic measure-


ments in estimating alveolar bone changes. J Clin Periodontol1990;17:14-21.

7. Jacobs R, Adriansens A, Verstreken K, Suetens P, van Steen-berghe D. Predictability of a three-dimensional planning systemfor oral implant surgery. Dentomaxillofac Radiol 1999;28:105-11.

8. Jeffcoat MK, Reddy MS. Advances in measurements of peri-odontal bone and attachment loss. Monogr Oral Sci2000;17:56-72.

9. Jacobs R, Gijbels F. Final report of the minosquare concertedaction. Leuven, Belgium: European commission, EDMIN, De-partment of Periodontology, KULeuven; 2000.

10. van der Stelt PF. Principles of digital imaging. Dent Clin NorthAm 2000;44:237-48.

11. Sanderink GCH. Imaging: new versus traditional technologicalaids. Int Dent J 1993;43:335-42.

12. Mol A. Image processing tools for dental applications. Dent ClinNorth Am 2000;44:299-318.

13. Vandre RH, Webber RL. Future trends in dental radiology. OralSurg Oral Med Oral Pathol Oral Radiol Endod 1995;80:471-8.

14. Brettle DS, Workman A, Ellwood RP, Launders JH, Horner K,Davies RM. The imaging performance of a storage phosphorsystem for dental radiography. Br J Radiol 1996;69:256-61.

15. Lim KF, Loh EEM, Hong YH. Intra-oral computed radiogra-phy—an in vitro evaluation. J Dent 1996;24:359-64.

16. Kaepller G, Vogel A, Axmann-Krcmar D. Intra-oral storagephosphor and conventional radiography in the assessment ofalveolar bone structures. Dentomaxillofac Radiol 2000;29:362-7.

17. Cury PR, Araujo NS, Bowie J, Sallum EA, Jeffcoat MK. Com-parison between subtraction radiography and conventional radio-graphic interpretation during long-term evaluation of periodontaltherapy in Class II furcation defects. J Periodontol 2004;75:1145-9.

18. Young SJ, Chaibi MS, Graves DT, Majzoub Z, Boustany F,Cochran D, et al. Quantitative analysis of periodontal defects ina skull model by subtraction radiography using a digital imagingdevice. J Periodontol 1996;67:763-9.

19. Kitagawa H, Scheetz JP, Farman AG. Comparison of comple-mentary metal oxide semiconductor and charge-coupled deviceintra-oral x-ray detectors using subjective image quality. Den-tomaxillofac Radiol 2003;32:408-11.

20. Eickholz P, Riess T, Lenhard M, Hassfeld S, Staehle HJ. Digitalradiography of interproximal bone loss: validity of differentfilters. J Clin Periodontol 1999;26:294-300.

21. Pecoraro M, Azadivatan-le N, Janal M, Khocht A. Comparisonof observer reliability in assessing alveolar bone height on directdigital and conventional radiographs. Dentomaxillofac Radiol2005;34:279-84.

22. Paurazas SB, Geist JR, Pink FE, Hoen MM, Steinman HR.
Comparison of diagnostic accuracy of digital imaging by using
CCD and CMOS-APS sensors with E-speed film in the periapicalbony lesions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod2000;89:356-62.

23. Tsiklakis K, Donta C, Gavala S, Karayianni K, KamenopoulouV, Hourdakis CJ. Dose reduction in maxillofacial imaging usinglow dose Cone Beam CT. Eur J Radiol 2005;56:413-17.

24. Ludlow JB, Davies-Ludlow LE, Brooks SL, Howerton WB.Dosimetry of 3 CBCT devices for oral and maxillofacial radiol-ogy: CB Mercuray, NewTom 3G and I-CAT. DentomaxillofacRadiol 2006;35:219-26.

25. Scarfe WC, Farman AG, Sukovic P. Clinical applications of conebeam computed tomography in dental practice. J Can Dent Assoc2006;72:75-80.

26. Almog DM, LaMar J, LaMar FR, LaMar F. Cone beam com-puterized tomography-based dental imaging for implant planningand surgical guidance, Part 1: Single implant in the mandibularmolar region. J Oral Implantol 2006;32:77-82.

27. Guerrero ME, Jacobs R, Loubele M, Schutyser F, Suetens P, vanSteenberghe D. State-of-the-art on cone beam CT imaging forpreoperative planning of implant placement. Clin Oral Investig2006;10:1-7.

28. Kobayashi K, Shimoda S, Nakagawa Y, Yamamoto A. Accuracyof measurement in distance using limited cone-beam computer-ized tomography. Int J Oral Maxillofac Implants 2004;19:228-31.

29. Misch KA, Yi ES, Sarment DP. Accuracy of cone beam com-puted tomography for periodontal defect measurements. J Peri-dontol 2006;77:1261-6.

30. White DR. Phantom materials for photons and electrons. TheHospital Physicists’ Association, Radiotherapy Topic Group.Scientific Report Series 1977;20:1-30.

31. Afifi AA, Clark V. Computer aided multivariate analysis, 2ndedition. Belmont, CA: Lifetime learning; 1984.

32. Siegel S. Non parametric statistics for the behavioural sciences.McGraw-Hill series in Psychology. New York: McGraw-Hill;1956.

33. Gibbs SJ. Effective dose equivalent and effective dose: compar-ison for common projections in oral and maxillofacial radiology.Oral Surg Oral Med Oral Pathol Oral Radiol Endod2000;90:538-45.

34. White SC. 1992 assessment of radiation risk from dental radi-ography. Dentomaxillofac Radiol 1992;21:118-26.

Reprint requests:

Reinhilde Jacobs, DDS, PhD, MSOral Imaging CentreKatholieke Universiteit LeuvenKapucijnenvoer 73000 Leuven, Belgium
[email protected]

vandenberghe 2007 cone beam perio diagnosis

Documents