RESEARCH Comparison of changes in dental and bone radiographic densities in the presence of different soft-tissue simulators using pixel intensity and digital subtraction analyses R S de Molon*,1, R G Batitucci1, R Spin-Neto2, G M Paquier1, C E Sakakura3, G M Tosoni1 and G Scaf1 1Department of Diagnosis and Surgery, School of Dentistry at Araraquara, Univ Estadual Paulista, Araraquara, S~ao Paulo, Brazil; 2Oral Radiology, Department of Dentistry, Aarhus University, Aarhus, Denmark; 3Department of Implantology, Barretos Dental School, Barretos Educational Foundation, S~ao Paulo, Brazil Objectives: To evaluate the influence of soft-tissue simulation materials on dental and bone tissue radiographic densities using pixel intensity (PI) and digital subtraction radiography (DSR) analyses. Methods: 15 dry human mandibles were divided into halves. Each half was radiographed using a charge-coupled device sensor without a soft-tissue simulation material (Wm) and with 5 types of materials: acrylic (Ac), wax (Wx), water (Wt), wood (Wd) and frozen bovine tissue (Bt). Three thicknesses were tested for each material: 10mm, 15mm and 20mm. The material was positioned in front of the mandible and the sensor parallel to the molar region. The radiation beam was perpendicular to the sensor at 30 cm focal spot-to-object distance. The digital images of the bone and dental tissue were captured for PI analyses. The subtracted images were marked with 14 landmark magnifications, and 2 areas of analyses were defined, forming the regions of interest. Shapiro–Wilk and Kruskal–Wallis tests followed by Dunn’s post-test were used (p, 0.05). Results: DSR showed that both the material type and the thickness tested influenced the gain of density in bone tissue (p, 0.05). PI analyses of the bone region did not show these differences, except for the lower density observed in the image without soft-tissue simulation material. In the dental region, both DSR and PI showed that soft-tissue simulators did not influence the density in these regions. Conclusions: This study showed that the materials evaluated and their thicknesses significantly influenced the density-level gain in alveolar bone. In dental tissues, there was no density-level gain with any soft-tissue material tested. Dentomaxillofacial Radiology (2013) 42, 20130235. doi: 10.1259/dmfr.20130235 Cite this article as: de Molon RS, Batitucci RG, Spin-Neto R, Paquier GM, Sakakura CE, Tosoni GM, et al. Comparison of changes in dental and bone radiographic densities in the presence of different soft-tissue simulators using pixel intensity and digital subtraction analyses. Dentomaxillofac Radiol 2013; 42: 20130235. Keywords: digital radiography; dental radiography; subtraction technique; bone density; radiography Introduction In vitro radiographic studies provide valuable and ac- curate diagnostic information in dental research.1 In those studies, soft-tissue simulators are often used to simulate what is present in vivo.2 In a clinical situa- tion, when an X-ray beam interacts with soft tissues and reaches the object, its intensity is attenuated, resulting in low-energy photons producing low-density areas in the image.3 The use of soft-tissue simulation materials having densities similar to the tissues found in patients in a real environment helps avoid exposure *Correspondence to: Dr Rafael Scaf de Molon, Department of Diagnosis and Surgery, School of Dentistry at Araraquara, Universidade Estadual Paulista, Rua Humaitá, 1680, 14801-903 Araraquara, S~ao Paulo, Brazil. E-mail: molon. foar@yahoo.com.br Received 6 July 2013; revised 19 August 2013; accepted 28 August 2013 Dentomaxillofacial Radiology (2013) 42, 20130235 ª 2013 The Authors. Published by the British Institute of Radiology http://dmfr.birjournals.org http://dx.doi.org/10.1259/dmfr.20130235 mailto:molon.foar@yahoo.com.br mailto:molon.foar@yahoo.com.br http://dmfr.birjournals.org of patients to excessive radiation during in vivo studies.4 Based on literature, the use of soft-tissue simulators, such as acrylic plates,2,4–8 wax,2,4,9,10 water,11–14 wood4,15,16 and bovine tissue17 has been tested in different ways with respect to the nature and/or the thickness of the material put around dry mandibles.2,5,16 It is important that these simulation materials are accurately reproducible, mea- surable and are ready for use in any circumstances.18 The influence of soft-tissue simulators on mandible bone radiographic density has been shown previously using digitial intraoral radiography.17 In a recent article, Schropp et al2 evaluated the thickness of wax and acrylic used as soft-tissue simulation materials to obtain a radio- graphic density similar to that of the human cheek. This study showed that acrylic and wax with thicknesses of 14.5mm and 13–17mm, respectively, could simulate the human cheek in in vitro radiographic studies. Digital imaging helps analyse bone density and archi- tecture quantitatively and qualitatively19 and has advan- tages over film-based radiography, which is mainly connected to smaller radiation doses.20 Digital images are composed of an array of pixels. Pixel intensity (PI) anal- ysis is a measure of the blackness and whiteness on a scale from 0 (totally black) to 4096 (totally white)19 and could be a useful and simple method to evaluate and measure the bone density in mandibles and maxillae. Also, digital imaging makes the analyses of dental and bone density using digital subtraction radiograph (DSR) possible. In this context, DSR detects subtle changes in alveolar bone in the radiographic image, being a highly sensitive methodology,21 improving the detection of periodontal lesions. DSR is also used for the evaluation of teeth with carious lesions22 and dental implants.23 This method consists of a subtraction oper- ation between two sequential radiographs in which structures that do not change are eliminated. Thus, DSR increases the ability to detect small changes in the ra- diographic image, improving the detection of bone alterations and reducing the interference of anatomical structures.24,25 However, digital subtraction method is still not routinely used in clinics because of standardiza- tion problems, and most of the available literature is in respect to in vitro and ex vivo studies. To the best of our knowledge, there are no studies that correlate different soft-tissue simulation materials and thicknesses for radiographic dental and bone den- sities using PI and DSR. Therefore, the purpose of this study was to evaluate the influence of five soft-tissue simulation materials of three thicknesses on dental and bone tissue radiographic densities using PI and DSR, with an aim to facilitate radiological experiments. Material and methods The sample of this study comprised 15 partially eden- tulous dry human mandibles from the Department of Morphology, Laboratory of Anatomy, Araraquara Dental School, Araraquara, Brazil. The mandibles were divided into halves; thus, the selected sample size was 30 dry hemimandibles. Inclusion criterion was the presence of at least two teeth in the posterior mandibular area. Soft-tissue simulator materials For the soft-tissue simulations, 5 types of materials were tested: acrylic (Ac), wax (Wx), water (Wt), wood (Wd), and bovine tissue (Bt), of 3 thicknesses: 10 mm, 15 mm and 20 mm, and in a standard size of 163 16 cm. For the simulation using water (Wt), three acrylic containers were made with the same dimensions of height and width but with varying depths, according to the thicknesses tested, and were filled with water. The walls of the acrylic containers were made as thin as possible (2 mm) to avoid interference with the results. Samples of bovine tissue (Bt), muscle and fat were packed in a thin plastic wrapper and frozen and made to the same dimensions in height and width. Image acquisition The posterior region of each hemimandible was radio- graphed using a dental unit (Oralix 765 DC; Gendex, Milan, Italy) and a charge-coupled device (CCD) sensor system (Visualix® eHD; Gendex, Milan, Italy) without material (Wm) and with the five tested materials, in the three tested thicknesses. The hemimandibles were fixed and stabilized using an appropriate device, and the material was positioned in front of the mandible. The sensor was placed parallel to the molar and/or premolar regions. The X-ray beam was positioned perpendicular to the mandible, sensor and soft-tissue simulators, at a 30 cm focal-spot-to-sensor distance, to standardize image acquisition. The exposure settings were 65 kVp, 7 mA and 0.08 s (Figure 1). The digital images were captured and saved at a resolution of 1300 dpi and depth of 8 bits and 16 bits and stored in the tagged image file format. Neither the sensor nor the mandible Figure 1 Exposition standardization: (a) X-ray cylinder, (b) soft- tissue acrylic material, (c) digital sensor and (d) mandible Influence of soft-tissue materials in bone and dental tissue 2 of 7 R S de Molon et al Dentomaxillofac Radiol, 42, 20130235 was moved during the experiment (only the soft-tissue simulation material was changed between recordings). Pixel intensity analyses The images obtained with depths of 8 bits and 16 bits were analysed using a dedicated software (ImageJ; National Institutes of Health, Bethesda, MD) for cal- culating the PI in 2 independent regions—bone and dental tissue. The area was limited to a square of 1003 100 pixels in the radiographic image. For the evaluation of bone tissue, the region of interest (ROI) was delimi- ted between the tooth roots and called Region 1 (Figure 2a). For the evaluation of dental tissue, the ROI was delimited between the healthy dental crown com- prising the enamel and dentin and called Region 2 (Figure 2b). The coordinates of each ROI were recorded, generating a macro for each jaw and each region, which was repeated for all images of the same hemimandible, to ensure that the same ROI was measured in all images. The average grey levels of the corresponding ROIs were analysed for both 8 and 16 bit images in the two ROIs, to compare the influence of the tested soft-tissue simulators on bone and tooth structure radiographic densities and if the depth of the materials would influence the pixel intensity results. Digital subtraction analyses For DSR analyses, the images were exported to a digital subtraction software (X-Poseit®, v. 3.1.17; Image In- terpreter System, Lystrup, Denmark). 2 digital images, 1 Wm and the other with 1 simulator material (Ac, Wx, Wt, Wd or Bt, in 3 diverse thicknesses—10mm, 15mm and 20mm) were imported to the software and posi- tioned side by side (Figure 3). Thus, the initial image was always the image without material (Wm), as the final image was the image with a soft-tissue simulator material, generating a subtracted image (Figure 4). To prepare the baseline (Wm) and the final image for the subsequent subtraction, 14 reference points were positioned in diverse places at the border of cement–enamel junction, enamel, dentin, and alveolar bone crest. The precision of the ref- erence points placed in the two radiographs was evaluated by means of an accuracy tool in the software. The points were repositioned if the distance between them in the two images was larger than 2 pixels. To restrict the areas of analyses to those interesting for the study, the following ROIs were outlined using the computer mouse: Region of interest: R1 related to the alveolar bone around the dental root and R2 related to the dental tissues. 1 rectangular window with 993 225 pixels and 1 square window with 733 73 pixels, respectively, were drawn for R1 and R2. Region of control: Two rectangular areas were drawn including the two areas of interest: dental tissue and al- veolar bone, to take into account the noise in the image, i.e., recording errors not accounted for by the program. ROI and region of control selected were made on the baseline radiographs and automatically transferred to the subtraction images. The standard deviation of the histogram distribution for the shades of grey in the region of control of the subtraction image was used to determine the threshold between what was interpreted as a signal and what could be determined as a noise in the subtraction image.26 The software performed the subtraction automatically, and all data concerning the subtraction in ROIs were auto- matically stored in the program data bank. Data analyses Statistical analysis was performed using GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA). Figure 2 Pixel intensity analysis in the ImageJ program (National Institutes of Health, Bethesda, MD) showing (a) Region 1, in bone tissue, and (b) Region 2, in dental tissues Influence of soft-tissue materials in bone and dental tissue R S de Molon et al 3 of 7 Dentomaxillofac Radiol, 42, 20130235 The results were comparatively evaluated in the different groups for all analysed parameters. Shapiro–Wilk and Kruskal–Wallis tests followed by the Dunn’s multiple comparisons post test were used to determine any sig- nificant differences between the groups, at a 5% signifi- cance level. Results Pixel intensity The results showed that no statistical difference was ob- served for PI among the diverse materials and thicknesses that were tested, in both Region 1 (with 8 and 16 bit images, respectively; Figure 5a,b) and Region 2 (with 8 and 16 bit images, respectively; Figure 6a,b). However, there was a statistically significant difference between the PIs obtained with the simulation materials and the ab- sence of a simulator material in Region 1 (bone tissue, for both 8 and 16 bit images). In Region 2 (dental tissue), there was no significant difference in the presence or the absence of a soft-tissue simulator. This result dem- onstrates that the material used interfered in the X-rays attenuation, increasing the PI in the bone region. Digital subtraction radiography DSR analysis for bone tissue showed that for a thick- ness of 10 mm, Ac, Wt and Bt presented a significantly higher gain of grey levels than Wd and Wx. For a thickness of 15 mm, the tested material did not in- fluence grey-level gain for Ac, Wt and Wd, whereas the Bt group presented a higher grey-level gain than the Wx group. For the thickness of 20 mm, Bt showed significantly higher grey-level gain, whereas Ac and Wd presented lesser grey level gain, compared with all other tested materials (Figure 7a). In general, Bt presented higher grey-level gain in the 3 tested thicknesses (Figure 8). Density gain area (by means of DSR) in R1 showed that the different soft-tissue simulator materials and thicknesses tested did not influence what was assessed for bone. Therefore, there was no statistically significant difference between the materials and thicknesses tested in this region (Figure 7b). In R2, there was not enough gain of grey-level data to perform any statistical anal- ysis because we found a majority of null results. Discussion In this study, we used PI and DSR to evaluate the in- fluence of soft-tissue simulation materials in the de- termination of dental and bone densities. To simulate soft tissues, we used five types of materials in three di- verse thicknesses. Both material type and thickness have been widely used2,5,6,9,10,15,16 with aims to simulate the real anatomical structures.27,28 In this study, we showed that both the material type and the thickness tested influenced the gain of density in bone tissue as demon- strated by DSR. PI analysis in the bone region showed low density in the image without simulating material. In the dental region, both DSR and PI revealed that soft- tissue simulators did not influence the teeth density. DSR is often used to detect subtle changes in bone and dental structures. However, the reproducibility of geometric factors on the images to be subtracted is Figure 3 Examples of the reference points marked using X-poseit, v. 3.1.17 (Image Interpreter System, Lystrup, Denmark) subtraction program. 14 reference points were positioned at the border of cement–enamel junction, enamel, dentin, and alveolar bone crest, standardizing the subtracted images. R1 is representative of the alveolar bone around the dental root and R2 related to the dental tissues Influence of soft-tissue materials in bone and dental tissue 4 of 7 R S de Molon et al Dentomaxillofac Radiol, 42, 20130235 essential for accurate digital subtraction radiogra- phy.29,30 Therefore, the digital sensor system (CCD) was chosen for being a direct digital radiographic system that does not need to be removed after exposure to an X-ray beam, assuring the standardization of geometric factors during the experiment. When an X-ray beam passes through soft tissues, some of these photons are absorbed by the tissues and some have its trajectory altered, leading to a dispersion or scattering of the beam, changing the incidence in radiographic film or sensor and, therefore, the measured radiographic density. The phenomenon of attenuation depends on atomic interactions, so materials with diverse compositions may lead to different degrees of attenuation, described as linear attenuation coefficient.3 As radio- graphic tissue density is derived from this attenuation coefficient degree, quantification of the density becomes more accurate when the influence of soft-tissue simulator materials is taken into consideration.17 According to Souza et al,17 different materials with the same thickness showed different optical densities in the mandibular bone area, using digitized film-based radiographic images. However, the differences with the methodology used in our study make it difficult to compare the findings. PI results for Region 1 showed that soft-tissue materials and thickness did not affect the ra- diographic density of the bone tissue. On the other hand, when there was no soft-tissue simulator material, the pixel intensity was lower, influencing the grey-level val- ues. Our results suggest that PI analysis of a specific ROI is a simple and efficient method that may be applicable to an analysis of density level in digital images. Figure 4 Representative subtracted image Figure 5 Pixel intensity gain of grey among different soft-tissue materials of each thickness in Region 1 with (a) 8 bits and (b) 16 bits (Kruskal–Wallis test–Dunn’s test). * indicates significant statistical difference (p, 0.05). ∙ represents the outliers—data more than two standard deviations Figure 6 Pixel intensity gain of grey among different soft-tissue materials of each thicknesses in Region 2 with (a) 8 bits and (b) 16 bits (Kruskal–Wallis test–Dunn’s test). ∙ represents the outliers—data more than two standard deviations Influence of soft-tissue materials in bone and dental tissue R S de Molon et al 5 of 7 Dentomaxillofac Radiol, 42, 20130235 In Region 2 (for PI) and R2 (for DSR), our results showed that the materials and thicknesses tested presented no statistically significant difference in density gain. This must be considered carefully when dealing with cavitation or demineralization detection. In studies related to dental caries,6,10,28,31 the simulation of soft tissue can lead to a reduction in radiographic image contrast, impairing the diagnosis. In this way, we cannot extrapolate the present results to that type of clinical situation. This finding may be related to the high density (lighter images) observed in the dental tissue radio- graphs, mainly for enamel. However, the attenuation of X-ray beams for soft-tissue simulation materials may interfere with the image of low-density structures, such as a bone. The bone marrow spaces in the radiographic image presents lower radiographic density. Thus, the attenuation of X-ray beams in the regions of bone, promoted by the interposition of soft-tissue simulation materials, is more noticeable than those of enamel and dentin. If the study includes teeth with cavitation or demineralization process, the loss of dental material might set the experiment more susceptible to density changes caused by the soft-tissue simulators. DSR could detect different grey-level gains among the soft-tissue simulator materials and thicknesses. Our results showed that bovine tissue as a soft-tissue simu- lator material presented the highest grey-level gain compared with the others. These results are in accor- dance with a previous study.17 Furthermore, Wx, Wd and Bt showed an increased grey-level gain according to the increase in thickness. In this experimental model, there was no difference between the materials and thicknesses related to the gain area analysed with DSR. A possible explanation for these results would be the exposure latitude of the used CCD sensor. A previous article32 has shown a wide range of latitudes among 18 X-ray detectors, including the 1 used in the present study and film radiographs. Moreover, photostimulable storage phosphor plate systems were shown to possess even wider latitudes than solid-state sensors. This variation of the dynamic range between film and digital systems may give rise to the assumption that a different result would have been obtained if a film was used in the present study because subtle differences in pixel intensity/image density related to the various types of simulation materials may have been revealed with a film. However, the use of digital imaging is increasing and replacing film-based radiog- raphy, and it is likely that most future assessments of new radiographic detectors using an in vitro model will be on digital receptors. Another aspect to be considered is that despite the assumption that conventional radiographic film would be able to present statistically significant differences between the use/non-use of soft-tissue simulation materials, the use of direct digital radiographic system has increased in recent years, and therefore it is nec- essary to know the differences that occur among the diverse radiographic digital systems. In this way, it is important to emphasize that the results of the present study should be considered specific to the intraoral digital system evaluated. In conclusion, according to the limits of this study, the materials tested were showed to significantly influence grey-level gain in the alveolar bone, especially when us- ing Bt. The thickness of the soft-tissue simulator tested also influenced the gain of radiographic density, in al- veolar bone, except for Ac and Wt. In dental tissues, Figure 7 (a) Digital subtraction radiography analyses among different soft-tissue materials of each thickness in Region 1 (Kruskal–Wallis test–Dunn’s test). * indicates significant statistical difference compared with the others (p, 0.05). (b) Digital subtraction analysis of gain area among the different soft-tissue simulator materials of three thicknesses. ∙ represents the outliers—data more than two standard deviations Figure 8 Representative subtracted images of 5 soft-tissue materials with thickness of 10 mm: acrylic (Ac), wax (Wx), water (Wt), wood (Wd) and bovine tissue (Bt) Influence of soft-tissue materials in bone and dental tissue 6 of 7 R S de Molon et al Dentomaxillofac Radiol, 42, 20130235 there was no grey-level gain with any soft-tissue simula- tor material tested, suggesting that it is not necessary for any type of material to simulate the soft tissues in teeth, and further studies, considering teeth with cavitation or demineralization processes, should be carried out. Funding This study was supported by a grant and fellowship from FAPESP (State of S~ao Paulo Research Founda- tion) numbers 2008/10680-0 and 2008/10145-8. References 1. de Molon RS, Morais-Camillo JA, Sakakura CE, Ferreira MG, Loffredo LC, Scaf G. Measurements of simulated periodontal bone defects in inverted digital image and film-based radiograph: an in vitro study. Imaging Sci Dent 2012; 42: 243–247. doi: 10.5624/isd.2012.42.4.243 2. Schropp L, Alyass NS, Wenzel A, Stavropoulos A. Validity of wax and acrylic as soft-tissue simulation materials used in in vitro radiographic studies. Dentomaxillofac Radiol 2012; 41: 686–690. doi: 10.1259/dmfr/33467269 3. Pietrobelli A, Formica C, Wang Z, Heymsfield SB. Dual-energy X-ray absorptiometry body composition model: review of physi- cal concepts. Am J Physiol 1996; 271: E941–E951. 4. Caldas Mde P, Ramos-Perez FM, de Almeida SM, Haiter-Neto F. Comparative evaluation among different materials to replace soft tissue in oral radiology studies. J Appl Oral Sci 2010; 18: 264–267. 5. Benchimol D, Näsström K, Shi X. Evaluation of automatic ex- posure control in a direct digital intraoral system.Dentomaxillofac Radiol 2009; 38: 407–412. doi: 10.1259/dmfr/22255081 6. Wenzel A, Haiter-Neto F, Gotfredsen E. Influence of spatial resolution and bit depth on detection of small caries lesions with digital receptors. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007; 103: 418–422. doi: 10.1016/j.tripleo.2006.05.016 7. Castro VM, Katz JO, Hardman PK, Glaros AG, Spencer P. In vitro comparison of conventional film and direct digital im- aging in the detection of approximal caries. Dentomaxillofac Radiol 2007; 36: 138–142. doi: 10.1259/dmfr/15558921 8. Shi XQ, Li G. Detection accuracy of approximal caries by black- and-white and color-coded digital radiographs. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009; 107: 433–436. doi: 10.1016/j.tripleo.2008.07.008 9. Mora MA, Mol A, Tyndall DA, Rivera EM. In vitro assessment of local computed tomography for the detection of longitudinal tooth fractures. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007; 103: 825–829. doi: 10.1016/j.tripleo.2006.09.009 10. Ulusu T, Bodur H, Odabaş ME. In vitro comparison of digital and conventional bitewing radiographs for the detection of approximal caries in primary teeth exposed and viewed by a new wireless handheld unit. Dentomaxillofac Radiol 2010; 39: 91–94. doi: 10.1259/dmfr/15182314 11. Borg E, Gröndahl HG. Endodontic measurements in digital radiographs acquired by a photostimulable, storage phosphor system. Endod Dent Traumatol 1996; 12: 20–24. 12. Gomes-Filho IS, Sarmento VA, de Castro MS, da Costa NP, da Cruz SS, Trindade SC, et al. Radiographic features of periodontal bone defects: evaluation of digitized images. Dentomaxillofac Radiol 2007; 36: 256–262. doi: 10.1259/dmfr/25386411 13. Halse A, White SC, Espelid I, Tveit AB. Visualization of stannous fluoride treatment of carious lesions by subtraction radiography. Oral Surg Oral Med Oral Pathol 1990; 69: 378–381. 14. Kullendorff B, Nilsson M, Rohlin M. Diagnostic accuracy of direct digital dental radiography for the detection of periapical bone lesions: overall comparison between conventional and direct digital radiography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996; 82: 344–350. 15. de Morais JA, Sakakura CE, Loffredo LC, Scaf G. Accuracy of zoomed digital image in the detection of periodontal bone defect: in vitro study. Dentomaxillofac Radiol 2006; 35: 139–142. doi: 10.1259/dmfr/31981949 16. Scaf G, Sakakura CE, Kalil PF, Dearo de Morais JA, Loffredo LC, Wenzel A. Comparison of simulated periodontal bone defect depth measured in digital radiographs in dedicated and non- dedicated software systems. Dentomaxillofac Radiol 2006; 35: 422–425. doi: 10.1259/dmfr/61300663 17. Souza PH, da Costa NP, Veeck EB. Influence of soft tissues on mandibular gray scale levels. Braz Oral Res 2004; 18: 40–44. doi: /S1517-74912004000100008 18. Farquharson MJ, Spyrou NM, al-Bahri J, Highgate DJ. Low energy photon attenuation measurements of hydrophilic materials for tissue equivalent phantoms. Appl Radiat Isot 1995; 46: 783–790. 19. Tosoni GM, Lurie AG, Cowan AE, Burleson JA. Pixel intensity and fractal analyses: detecting osteoporosis in perimenopausal and postmenopausal women by using digital panoramic images. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006; 102: 235–241. doi: 10.1016/j.tripleo.2005.08.020 20. de Molon RS, Sakakura CE, Morais-Camillo JA, de Almeida Junior PC, de Castro Monteiro Loffredo L, Scaf G. Comparison between embossed digital imaging and unprocessed film-based radiography in detecting periodontal bone defects: an in vitro study. Oral Radiol 2012; 28: 95–100. Available from: http://dx.doi. org/10.1007/s11282-012-0088-y 21. Christgau M, Hiller KA, Schmalz G, Kolbeck C, Wenzel A. Quantitative digital subtraction radiography for the determination of small changes in bone thickness: an in vitro study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998; 85: 462–472. 22. Ricketts DN, Ekstrand KR, Martignon S, Ellwood R, Alatsaris M, Nugent Z. Accuracy and reproducibility of conventional ra- diographic assessment and subtraction radiography in detecting demineralization in occlusal surfaces. Caries Res 2007; 41: 121–128. doi: 10.1159/000098045 23. Sakakura CE, Neto RS, Bellucci M, Wenzel A, Scaf G, Marcantonio E Jr. Influence of homeopathic treatment with comfrey on bone density around titanium implants: a digital subtraction radiography study in rats. Clin Oral Implants Res 2008; 19: 624–628. doi: 10.1111/j.1600-0501.2007.01514.x 24. Grondahl HG, Grondahl K, Webber RL. A digital subtraction technique for dental radiography. Oral Surg Oral Med Oral Pathol 1983; 55: 96–102. 25. Gröndahl HG, Gröndahl K. Subtraction radiography for the di- agnosis of periodontal bone lesions. Oral Surg Oral Med Oral Pathol 1983; 55: 208–213. 26. Wenzel A, Sewerin I. Sources of noise in digital subtraction ra- diography. Oral Surg Oral Med Oral Pathol 1991; 71: 503–508. 27. Attaelmanan A, Borg E, Gröndahl HG. Digitisation and display of intra-oral films. Dentomaxillofac Radiol 2000; 29: 97–102. doi: 10.1038/sj/dmfr/4600511 28. Wenzel A, Borg E, Hintze H, Gröndahl HG. Accuracy of caries diag- nosis in digital images from charge-coupled device and storage phosphor systems: an in vitro study. Dentomaxillofac Radiol 1995; 24: 250–254. 29. Davis M, Allen KM, Hausmann E. Effects of small angle dis- crepancies on interpretations of subtraction images. Oral Surg Oral Med Oral Pathol 1994; 78: 397–400. 30. Gröndahl K, Gröndahl HG, Webber RL. Influence of variations in projection geometry on the detectability of periodontal bone lesions. A comparison between subtraction radiography and conventional radiographic technique. J Clin Periodontol 1984; 11: 411–420. 31. Sogur E, Baksı BG, Orhan K, Paksoy SC, Dogan S, Erdal YS, et al. Effect of tube potential and image receptor on the detection of natural proximal caries in primary teeth. Clin Oral Investig 2011; 15: 901–907. doi: 10.1007/s00784-010-0461-3 32. Farman AG, Farman TT. A comparison of 18 different x-ray detec- tors currently used in dentistry. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005; 99: 485–489. doi: 10.1016/j.tripleo.2004.04.002 Influence of soft-tissue materials in bone and dental tissue R S de Molon et al 7 of 7 Dentomaxillofac Radiol, 42, 20130235 http://dx.doi.org/10.5624/isd.2012.42.4.243 http://dx.doi.org/10.5624/isd.2012.42.4.243 http://dx.doi.org/10.1259/dmfr/33467269 http://dx.doi.org/10.1259/dmfr/22255081 http://dx.doi.org/10.1016/j.tripleo.2006.05.016 http://dx.doi.org/10.1259/dmfr/15558921 http://dx.doi.org/10.1016/j.tripleo.2008.07.008 http://dx.doi.org/10.1016/j.tripleo.2008.07.008 http://dx.doi.org/10.1016/j.tripleo.2006.09.009 http://dx.doi.org/10.1259/dmfr/15182314 http://dx.doi.org/10.1259/dmfr/25386411 http://dx.doi.org/10.1259/dmfr/31981949 http://dx.doi.org/10.1259/dmfr/31981949 http://dx.doi.org/10.1259/dmfr/61300663 http://dx.doi.org//S1517-74912004000100008 http://dx.doi.org//S1517-74912004000100008 http://dx.doi.org/10.1016/j.tripleo.2005.08.020 http://dx.doi.org/10.1007/s11282-012-0088-y http://dx.doi.org/10.1007/s11282-012-0088-y http://dx.doi.org/10.1159/000098045 http://dx.doi.org/10.1111/j.1600-0501.2007.01514.x http://dx.doi.org/10.1038/sj/dmfr/4600511 http://dx.doi.org/10.1038/sj/dmfr/4600511 http://dx.doi.org/10.1007/s00784-010-0461-3 http://dx.doi.org/10.1016/j.tripleo.2004.04.002