Biomech Model Mechanobiol (2012) 11:751–758DOI 10.1007/s10237-011-0348-5

ORIGINAL PAPER

Analysis of the compressive strain below the removable and fixedprosthesis in the posterior mandible using a digital imagecorrelation method

Ivan Tanasic · Aleksandra Milic-Lemic ·Ljiljana Tihacek-Sojic · Ivica Stancic ·Nenad Mitrovic

Received: 2 April 2011 / Accepted: 31 August 2011 / Published online: 15 September 2011© Springer-Verlag 2011

Abstract It was the purpose of this study to determine andanalyse strains in the bone of posterior mandible below theremovable and fixed partial dentures using Digital ImageCorrelation Method. Dried mandible with shortened den-tal arch (Kennedy Class 1) was used in the experiment.The mandible model was prepared for accepting two typesof restorations for bilaterally missing molars conventionaltherapy, and removable and fixed partial dentures weremanufactured following standard prosthetic protocol. Themodels with prosthetic restorations placed in situ were sub-jected to loading of 50–300 N, and photographed using twodigital cameras as part of the digital image correlation methodequipment. Afterwards, the obtained data for strains withinrestored mandible during loading ciclus were analysed inthe software Aramis and graphically presented. Percentagesize of the mandible strains within the line section are from0.14 to 0.19% for the removable partial denture experimentalmodel and 0–0.14% for the fixed partial denture experimen-tal model. The study has demonstrated that Digital ImageCorrelation method can be used to measure strain on themandible surface and replacements during loading and thatfrom biomechanical perspective both therapeutic modalitiescan be equally useful for the restoration of the mandible withbilaterally distal edentulous spaces.

I. Tanasic (B) · A. Milic-Lemic · L. Tihacek-Sojic · I. StancicSchool of Dentistry, University of Belgrade, Belgrade, Serbiae-mail: doktorivan@hotmail.com

A. Milic-Lemice-mail: saskam@eunet.rs

N. MitrovicInnovation Center of Faculty of Mechanical Engineering,University of Belgrade, Belgrade, Serbia

Keywords Mandible · Removable partial denture ·Fixed partial denture · Digital image correlation method

1 Introduction

The success of therapeutic concepts in patients with short-ened dental arch is primarily determined with biomechan-ical compatibility of the restoration design with the oraltissues, named as biomechanical acceptance of the replace-ments (Kulchin et al. 2008). As a conservative procedure,removable partial denture (RPD) and Fixed Partial Denture(FPD), have been extensively used for the rehabilitation ofpartially edentulous patients over the years. However, themechanism of tissue behavior under mentioned restorationsis not fully analyzed or completely understood. Restoredpartial edentulous mandible may exhibit different transferof vertical loading through the mandible bone compared tounrestored mandible with intact dental arch. This may leadto various strain and resorption rate of posterior mandiblebone (Jin et al. 2004). Investigations of the restored posteriormandible are very important for better understanding the rela-tions between the prosthesis design and the effect of stressand strain acting upon the dental structure, abutment teethand residual ridge (Saito et al. 2003; Igarashi et al. 1999).Also, assumption for some functional relation between theform of mandible and masticatory stress may be established(Daegling and Hylander 2000). Therefore, additional knowl-edge about biomechanical behavior of mandible with differ-ent kind of replacements is imperativ. This type of study canprovide better prognosis in clinical practice.

Digital image correlation (DIC) as an optical measuringsystem is increasingly used for in vitro set-ups and may beutilized in dental biomechanical contexts (Shelton and Katz1991). DIC is an optical full-field technique for non-contact,

123

752 I. Tanasic et al.

3D deformation measurements (Kahn-Jetter and Chu 1990).A high contrast speckle pattern is applied onto the surface ofthe sample and observed by the charge-coupled device (CCD)cameras during loading. The entire field of view is dividedinto a number of unique correlation areas, or ‘facets’, whichtypically contain a square subset of pixels. Facets track thecharacteristic features of the speckle pattern during loadingand provide a progressive measurement of deformation.

Biomechanical in vitro investigations of the human jawbehavior under static and dynamic loading (Rodriguez et al.2004) has shown the suitability of DIC as a method for assess-ing deformations of objects with complex geometries. DIC isparticularly suitable for biological applications because it canbe used for accurately determining strain in inhomogeneous,anisotropic, non-linear materials, such as mandible.

In the following study it was the aim to use DIC for eval-uating strains under two types of replacements (removablepartial denture and fixed partial denture) placed on the mandi-ble with bilaterally missing molars. The other aim was also topresent the method of DIC as an apropriate method for asse-sing strain distribution throughout the structures with com-plex geometries (Rodriguez et al. 2004) such as prosthesisrestored mandible.

2 The method and material

Dried (macerate) mandible was used in the experiment. Themandible was borrowed from the Laboratory for Anthropol-ogy, Institute of Anatomy, School of Medicine, Universityof Belgrade. Donor was men, 45 years of age from Serbia.For the experiment specimen selection, the exclusion crite-ria was that mandible was not with evident traumatic andpathological damages. In the inclusion criterion, the mandi-ble should have shortened dental arch (Kennedy class I) withabsence of the posterior molars, and the second premolar andmolars in the right and left side, respectively. Mandible waspreviously left immersed in the physiological solution (0.9%NaCl) in order to reach the volume and elasticity as in vivostudies (Arendts and Sigolotto 1990; Vollmer et al. 2000;DuChesne et al. 2003). After drying at the temperature of 27Celsius degrees, the mandible was prepared for accepting theprosthetic restorations.

Following preparation of the abutment teeth in accordancewith the main biomechanical principles of teeth preparation(Davenport et al. 2001; Shillingburg et al. 1973) and standardimpression procedures of the supporting tissues, two typesof restorations were made. The first restoration was com-plex removable partial denture (RPD) in combination withfull coverage cast porcelain fused to metal (PFM) crownson remaining teeth. Bredent ball attachments were used toconnect PFM crowns and RPD into functional union. Designof the RPD employed shortened buccal wings of denture

saddles and were modelled in such a manner for easier visualcontrol of strain and displacement during the experiment.Afterwards, eleven unit PFM cantilever FPD was fabricatedon the prepared mandible teeth.

Consequently, two experimental models of the same man-dible were obtained: the experimental model of the partialedentulous mandible with RPD positioned in situ and theexperimental model of the partial edentulous mandible withcantilever positioned in situ. The surfaces of the experimentalmodels with adequate restorations were sprayed with a thincoat of white layer, followed with a thin layer of high contrastblack paint placed on the top of the white layer in order toutilize the correct performance of the DIC method. The finegraning of the employed white and black spray (acrylic, man-ufacurer Motip) occupies certain mutual distances that werechanged under the loading and were registered by cameras.

Both models were placed in the standard tensile testingmachine-press system. The study employed simulated occlu-sal loading applied in vertical direction, with the intensityrange of 50–300 N (with the interval of 50 N). Although themaximal willing force in humans measured in the poster-ior region is 500–700 N, it was considered that the forceintensity decreases with the teeth loos and depends of mus-cular strength (Muller et al. 2001). The force was appliedto the buccal and lingual cusps of loading teeth in bothmodels, across the horizontal extention of the gnatodyna-mometer (Qian et al. 2009; Zaslansky et al. 2006). Gnatody-namometer was used for the purpose of precise and controlledforce (load) measuring. The loading tooth for both modelswas artificial (acrylic or metal–ceramic) left lower secondpremolar. The study included only the left part of the mandi-ble viewed from sagittal (lateral) aspect excluding the rightside of mandible. Additionally, the present analysis did nottake into account the viscoelasticity properties of the PDLtissue neither do mucosa layer.

Precise strain measurements were done using DIC sys-tem equipment of manufacturer GOM (Braunschweig, Ger-many). System consists of two digital cameras (50 mmlenses with the 25 mm distance ring, Schneider Kreuznach,Bad Kreuznach, Germany), trigger box, PC and softwareARAMIS (version 6.2.0., Braunschweig, Germany). In pre-cedence to the experiment measuring, the calibration of thesystem was performed. System calibration is an importanttask in 3D measuring assuring accurate dimensional mea-surements. The calibration objects can be the panels orcrosses of different sizes. The camera positioning on thestand is done manually according to parameters defined bymanufacturer. Parameters are chosen based on dimensions ofmeasuring object, i.e. measuring volume.

The project was specified for each new measuring step andthe photos were taken at different stages of loading. Through-out the measuring stages, software processing of measureddata was executed, enabling 3D presentation of the results,

123

Analysis of the compressive strain 753

analyzing obtained results, performing statistic data, creatingdiagrams, reports etc. As a result of the software processingphotos consisting the central figure of the structure underinvestigation and color scale shown on the right side wereobtained. The color scale serves for analysing the results,whereas each color represented on the scale correspondsto a certain percentage of strain obtained during loading.Figures that were gained in the following study included twoimportant parameters: section line and selection area. Sectionline consist of reference points selected by the software itselfissued to calculate the change in length. Thus, it was possibleto calculate the distance between any two points within thesurface of the tested structure (model). The stage-referencepoint was also set by software depending on the region underinvestigation (stage point position depends on the region ofbone that is explored). Stage point values are not constant butfluctuate (mutate) through the stages. Stage point in the pre-sented figures corresponds to the strain value obtained in onepoint of the third and the last stage. The highest referencepoint, used for comparison between models is stage point(stage points 3 and 5). The deepest (lowest) reference pointis the eqvivalent of the lowest point of the mandible basalpart viewed from sagital (lateral aspect) and this point is alsothe lowest point of the section line. Also it was consideredthat distance between this two points is the shortest distancebetween any two points. This distance changes its length withthe change of intensity and the slams point force.

The selection area was the area of interest for comparingthe figures of presented experimental models. Section line isthe part of selection area. The following study included theleft posterior mandible around the distally abutment teeth andthe region of mandible bone below prosthesis as selectionarea. Also the anatomically significant parts of the mandi-ble body such as foramen mentally and trigonum retromolarwere considered interesting and were included within selec-tion area.

Strains within selection area may be measured in a rangeof 0.01% up to several 100% and strain accuracy is upto 0.01% (Windisch et al. 2007). Both, small and largeobjects (from 1 to 2,000 mm) can be measured with the samesensor.

ARAMIS, software used in the study, is based on the prin-ciple of objective raster (fine-ground) procedure. It serves formeasuring 3D changes of shape and strain distribution on thesurface of statically and dynamically loaded objects. As non-contact and highly accurate method, ARAMIS determinesthe shape of the photographed object, its dimensions and 3Dstrain field. ARAMIS analyses and graphically presents mea-sured results providing optimal understanding of the testedobject behavior (Schmidt et al. 2003; Vendroux and Knauss1998). It is especially suitable for 3D strain measurementunder static and dynamic loading, in order to analyze thestrain field of real components (Goellner et al. 2010).

3 Results

The strain values within the section line (reference line,and the part of section area) obtained for the RPD modelwere higher than strain values obtained for the FPD model.According to deformation formula e = (L0 − L1)/L0 ×100,where L0 and L1 were the lengths before and after loading,respectively, the strain values were expressed in percents, andpresented on the scale by different colors (Figs. 1a, 2a). Also,further individual strain values were analyzed with ARAMISsoftware and are shown in Table 1.

The vertical line (Section 0) (Figs. 1a, 2a), was set bysoftware under the slams point of the load acting (in thedirection of load) on acrylic or metal-ceramics left secondlower premolar. Any increase in the intensity of load maybe presented with figures of corresponding stage. Since thestudy adopted simulated loading in range from 50 to 300 Nat intervals of 50 N, obtained multistage view and figuresof dimensional changes of structure under investigation ispresented in only six stages. However, for easy interpreta-tion of the results for both RPD and FPD models figuresincluded in the article visualize strain distribution when theload reached the intensities of 200 and 300 N, respectively.Strain intensitiy values are presented as gradient of col-ors in the scale on the right side of the figures. Exact val-ues of strain for each intensity of applied force are shownin Table 1. Separate strain values obtained for each pointof the line section in the last stage of the force acting ondental structures under investigation is shown in Diagrams(Figs. 3, 4).

As mentioned before the study primarily focused on thestrain distribution inside the bone below the prosthesis andaround the abutment teeth and is presented with respect to thedifferent colors. Therefore, the regions of sufficient interesetwere the alveolar part of the mandible, the abutments cani-nus and first premolar, the foramen mentally and trigonumretromolare.

3.1 Experimental model of the partial edentulous mandiblewith RPD positioned in situ

The major (principle) strain field for the third and the laststage is obtained during cusps loading of 200 and 300 N.Loaded tooth was left lower second premolar. Line sectionis just below the force slams (attack) point.

Percentage size of the lower jaw strains within the linesection are from 0.14 to 0.19%.

Separate values of strain (%) for each reference point ofsoftware installed line section in the last stage within theintensity force of 300 N are shown in diagram in Fig. 3. Theactual values of strain obtained during loading of differentintensity are given in Table 1. This strain values are showinga linear increase. The average strain of the line representing

123

754 I. Tanasic et al.

Fig. 1 Major strain field for thethird stage of the RPD modelduring vertical loading. a Majorstrain field for the fifth stage ofthe RPD model during verticalloading

mandible alveolar bone is 0.1%. The highest strain valuesare noticed just below the RPD. The average strain values ofthe anatomical structure-aperture foramen mentally and tri-gonum retromolare are between 0.5 and 0.7%, respectively.The buccal marginal periodontium of abutments premolarand canine deformes of about 0.5%.

3.2 Experimental model of the partial edentulous mandiblewith cantilever FPD positioned in situ

The major (principle) strain field (Fig. 2a) is obtained duringthe cusps loading on the distally extended cantilever ponticunit-premolar in the mandible. The values of strains in the

mandible are given in Table 1 with applied forces from 50to 300 N, increasing gradually by 50 N. This line joins thepoints of reference placed on the observed object, mandi-ble and cantilever, and changes its length with the changeof intensity and the slams point force (Fig. 4). The scale(Fig. 2a) enables registering of quantative changes in thebone volume, and it is presented in percents. Percentage sizeof the mandible strains within the section line are from 0 to0.14%. The average strain of the line surrounding mandiblebone tissue is 0.1%. The highest strain values are noticedin the region of the abutments marginal periodontium withthe mean strain value 0.4–0.5% and in the anatomical struc-ture-aperture (foramen mentale) with the maximum of 0.4%.

123

Analysis of the compressive strain 755

Fig. 2 Major strain field for thethird stage of the FPD modelduring vertical loading. a Majorstrain field for the fifth stage ofthe FPD model during verticalloading

Table 1 Strain values for RPD and FPD experimental models

Comparison of strain values (%)

Load [N] Stage RPD model FPD model

Section line Foramen mentally Trigonum retromolar Section line Foramen mentally Trigonum retromolar

50 0 0.14 0.20 0.25 0 0.16 0.05

100 1 0.15 0.24 0.36 0.03 0.18 0.08

150 2 0.16 0.27 0.44 0.05 0.2 0.1

200 3 0.17a 0.3 0.5 0.12a 0.2 0.1

250 4 0.18 0.45 0.6 0.13 0.33 0.15

300 5 0.19b 0.5 0.7 0.14b 0.4 0.2

a Strain values registered in the third stage of the RPD and FPD modelb Strain values registered in the fifth stage of the RPD and FPD model

123

756 I. Tanasic et al.

Fig. 3 Diagram of section linewithin the fifth stage of the RPDmodel during vertical loading

Fig. 4 Diagram of section linewithin the fifth stage of the FPDmodel during vertical loading

Bone strain value below the distal unit of cantilevered FPDis 1%.

4 Discussion

This study has demonstrated that Digital image correlationmethod (DIC) can be used to measure strain on the mandi-ble bone surface and restoration during loading. DIC havenot been widely used in dental research so far, although ithas several advantages over measurements by others digitalmethods. It is much less sensitive to ambient vibrations, candetect rigid body motion and simultaneously measure 3D dis-placements in a high dynamic range (microns to millimeters)of measuring capacity. The repeatability of the optical mea-suring is very good with the variation coefficient of 0.5%

(Windisch et al. 2007). Unlike single camera systems thathave difficulty measuring strain on complex surfaces geom-etries (Yang et al. 2007), the two camera system allows accu-rate measurements on flat surfaces and slightly less accuratemeasurement of the investigated curved surfaces. The studyincluded only the left part of the mandible model viewed fromsagittal (lateral) aspect where the right side of the mandiblewas excluded.

As all other cadaveric analyzes the experiment may sufferfrom the donor-related variability of the investigated boneproperties, the absence of soft tissue in the anatomic partinvestigated and impossibility of positioning mandible asin real situation. Nevertheless, this type of study analyz-ing using optical method (digital image correlation method),could help in understanding the nature of stress and straindistribution through the hierarchical structure of bone.

123

Analysis of the compressive strain 757

The modeling of soft tissue has not been done in theresearch, and viscoelastic properties of the periodontal liga-ment (PDL) and mucosa layer were not included when eval-uating the results. Knowing their dimensions and physicalcharacteristics it was considered that their biomechanicalbehavior might influence only the intensity of strain, butnot change the direction of strain distribution (Stokes andGreenapple 1985; Toms et al. 2002; Toms and Eberhardt2003). Nevertheless, it was the intention of this study to ana-lyze and present the DIC as a possible method for straindistribution analyze in the field of dental biomechanics. Thatis the reasons why strain values were presented qualitativelyas gradient of different colors and analyzed in percents, inorder to give full insight into the biomechanical behavior ofthe analyzed structures.

Study presented biomechanical view of two possible ther-apeutic solutions for the case of the mandible shorteneddental arch. Shortening the buccal wings of RPD saddleswas made for a better comparison of results between twoexperimental models. In accordance a wider field for opticalobservation of the alveolar ridge buccal side including themandible upper part was achieved. Since, treatment modal-ities employed in the study presented different biomechani-cal behaviour for possible comparison, the focus of analysewas placed on strains in the alveolar bone. According to theobtained strain values the software generated section linepositioned in the bone structure beneath the point of appliedload. Although, not very much in the same location for bothmodels, section lines were characterised by reference pointsin the bone that exibited greatest strain during loading. Thus,proper comparison of strain values between two experimen-tal models was enabled.

Differences are found between two experimental mod-els. Section line of the mandible restored with RPD is char-acterized with higher strain values. Force of 150 N causedstrains that were almost four times bigger than in the FPDmodel. Increase in values of strain in every single stage is alsoobserved, obviously confirming an existance of linear depen-dence between force (load) and strain. Observed difference instrain values for both models are probably a consequence ofdifferent kind of fixation existing within two types of restora-tions. In the RPD model all remaining teeth were splinted infull cast restoration in combination with attachment-retainedRPD. This type of RPD allows movement of the free-endsaddles toward the edentulous ridge when loaded. The move-ment resulted in transfer of a portion of the applied load tothe edentulous ridge, which led to observed stress and strainbeneath the denture saddle (see Fig. 2). Major strain fieldexisted around abutment tooth, but of less intensity comparedto major strain field on the alveolar ridge. The difference inthe obtained values may be attributed to the effect of splintingremaining teeth in the experiment. The improvement in stressdistribution to the supporting structures with fixed splinting

was demonstrated for both mandibular and maxillary RPDs(Berg and Caputo 1993; Itoh et al. 1998). Major strain valuesfor the RPD model showed greater strain of the distal portionof the saddle and distal parts of the abutment tooth. In the FPDmodel all remaining teeth were splinted in full-arch recon-struction with cantilever FPD. The rigid construction of thecantilever FPD when loaded exhibited lesser strain valuescompared to the RPD in the first experiment. Also, loadingof the cantilever FPD caused only limited stress transfer todistant portions of the mandibular arch, while splinting ofthe remaining teeth allowed uniform distribution of load tothe supporting alveolar bone. However, it is shown that thebone underlying the FPD undergoes a higher resorption inthe buccal region, which is in correlation with findings ofField et al. (2010). It seems that when strain pattern for bothmodels were compared the FPD loading did not induce stron-ger dimensional alterations of mandible. However, the FPDtherapy influenced the higher strain magnitude on the mar-ginal bone of adjacent abutments when loaded on a distallyextended unit. This may be explained that when connect-ing two or more teeth rigidly, loading was distributed andtransmitted to both or each single abutments (and retainers)(Yamashita et al. 2006). The effect of connecting the teethseems to be limited locally, because the direction of the prin-cipal strains is within the upper part (marginal bone) of themandible in both cases (RPD and FPD therapy). Accordingly,it can be argued that FPD therapy does not alter the overallstrain pattern of the mandible during loading as RPD does,although higher strains within the facial marginal bone wereregistred next to FPD abutments than to RPD abutments.Also, major strain values observed in the marginal bone andabutment teeth emphasize the importance of avoiding ponticonly loading.

This in vitro study can not identically reproduce the realclinical situation, nevertheless the abovementioned resultsshowed that both therapeutic modalities (RPD, FPD) can beequally useful for the restoration of the mandible with bilat-erally distal edentulous spaces, in terms of biomechanics.It must be considered that cantilever FPD may not providethe best biomechanical solution on periodontally compro-mised abutment teeth due to their possible overloading dur-ing masticatory functions. In such cases, properly designedand adjusted RPD with fixed splinting of remaining teethmight be considered as therapeutic option. Although, in sit-uations with diminished periodontal support extensive full-arch splinting may not be effective or appropriate when usingRPD (Itoh et al. 1998).

5 Conclusion

After all the aforementioned and within the limitations of thisstudy the following conclusions may be provided:

123

758 I. Tanasic et al.

– Optical method of digital visualizing could be the methodof choise in biomechanical-dentistry research;

– There is existence of linear dependence between forceintensity and strain or displacement in both experimentalmodels;

– Cantilever pontic is biomechanically acceptable therapysolution in the case of splinting all the remaining teeth ina single unit in patients with mandibular shortened dentalarch;

– Findings provide that cantilever FPD therapy induce lessstrain in the bone of residual ridge. However, strains areconcentrated in the alveolar bone around the abutmentteeth.

References

Arendts FJ, Sigolotto C (1990) Mechanical characteristics of the humanmandible, and investigation of the ‘in-vivo’ reaction of the com-pact bone: a contribution to the descreption of the biomechanicsof the mandible-part II. Biomedizinische Technik 35:123–130

Berg T, Caputo AA (1993) Maxillary distal extension removable partialdenture abutments with reduced periodontal support. J ProsthetDent 70:245–250

Daegling DJ, Hylander WL (2000) Experimental observation, theoret-ical models, and biomechanical inference in the study of mandib-ular form. Am J Phys Anthropol 112:541–551

Davenport JC, Basker RM, Heath JR, Ralph JP, Glantz PO, HammondP (2001) Tooth preparation. Br Dent J 190:288–294

DuChesne A, Unnewehr M, Schmidt PF, Sotonyi P, Brinkmann B, Pif-fko J, Fischer G, Bajanowski T (2003) Deformation characteristicsof the human mandible in low impact experiments. Int J Legal Med117:257–262

Field C, Li Q, Li W, Thompson M, Swain M (2010) Prediction of man-dibular bone remodelling induced by fixed partial dentures. J Bio-mechanics 43:1771–1779

Goellner M, Schmitt J, Karl M, Wichmann M, Holst S (2010) Photo-grammetric measurement of initial tooth displacement under ten-sile force. Med Eng Phys 32:883–888

Igarashi Y, Ogata A, Kuroiwa A, Wang CH (1999) Stress distributionand abutment tooth mobility of distal-extension removable partialdentures with different retainers: an in vivo study. J Oral Rehabil26:111–116

Itoh H, Caputo AA, Wylie R, Berg T (1998) Effects of periodontal sup-port and fixed splinting on load transfer by removable partial den-tures abutments with reduced periodontal support. J Prosthet Dent79:465–471

Jin X, Sato M, Nishiyama A, Ohyama T (2004) Influence of loadingposition of mandibular unilateral distal extension, removable par-tial dentures on movements of abutments tooth and denture base.J Med Dent Sci 51:155–163

Kahn-Jetter Z, Chu T (1990) Three-dimensional displacement mea-surements using digital image correlation and photogrammic anal-ysis. Exp Mech 30:10–16

Kulchin Y, Vitrick OB, Lantsov AD, Vorobyev VA, MoskvinY (2008) Measuring the deformity of dentofacial bone tissuesusing a speckle correlation method. Optoelectron Instrum DataProcess 44(3):99–105

Muller F, Heath MR, Ott R (2001) Maximum bite force after thereplacement of complete dentures. Gerodontol 18:58–62

Qian L, Todo M, Morita Y, Matsushita Y, Koyano K (2009) Deforma-tion analysis of the periodontium considering the viscoelasticityof the periodontal ligament. Dent Mater 25:1285–1292

Rodriguez D, Moreno V, Gallas M, Abeleira MT, Suarez D (2004) In-plane electronic speckle pattern of interference (ESPI) with opticalfibre system applied to the study of the human jaw. Med Eng Phys26:371–378

Saito M, Miura Y, Notani K, Kawasaki T (2003) Stress distributionof abutments and base displacement with precision attachment-and telescopic crown-retained removable partial dentures. J OralRehabil 30:482–487

Schmidt T, Tyson J, Galanulis K (2003) Full-field dynamic displace-ment and strain measurement using advanced 3D image correlationphotogrammetry: part 1. Exp Tech 27:47–50

Shelton JC, Katz DM (1991) Laser interferometric techniques in ortho-paedics. J Med Eng Technol 15:209–221

Shillingburg HT, Hobo S, Fisher DW (1973) Preparation design andmargin distortion in porcelain-fused-to-metal restorations. J Pros-thet Dent 89(6):527–532

Stokes I, Greenapple DM (1985) Measurement of surface deformationof soft tissue. J Biomech 18:1–17

Toms S, Eberhardt A (2003) Nonlinear stress-strain behavior of peri-odontal ligament under orthodontic loading. Am J Orthod Dento-fac Orthop 123:657–665

Toms SR, Dakin GJ, Lemons JE, Eberhardt AW (2002) Quasi-linearviscoelastic behavior of the human periodontal ligament. J Bio-mech 35:1411–1415

Vendroux G, Knauss W (1998) Submicron deformation field measure-ments: part 2. Improved digital image correlation. Exp Mech38:86–92

Vollmer D, Meyer U, Joos U, Vegh A, Piffko J (2000) Experimentaland finite element study of a human mandible. J Cranio Maxillo-fac Surg 28:91–96

Windisch S, Jung R, Sailer I, Studer S, Ender A, Hämmerle C (2007) Anew optical method to evaluate threedimensional volume changesof alveolar contours: a methodological in vitro study. Clin OralImpl Res 18:545–551

Yamashita J, Wang Q, Dechow PC (2006) Biomechanical effects offixed partial denture therapy on strain patterns of the mandible. JProsthet Dent 95:55–62

Yang L, Zhang P, Liu S, Samala PR, Su M, Yokota H (2007) Mea-surement of strain distributions in mouse femora with 3D-digitalspeckle pattern interferometry. Opt Lasers Eng 45(8):843–851

Zaslansky P, Shahar R, Friesem AA, Weiner S (2006) Relationsbetween shape, materials properties, and function in biologicalmaterials using laser speckle interferometry: in situ tooth defor-mation. Adv Funct Mater 16(15):1925–1936

123