ORIGINAL ARTICLE
Year : 2015 | Volume
: 3 | Issue : 1 | Page : 1--6
A comparison of fractal dimension values of peri-implant bone and healthy contralateral side using panoramic radiographs
Betül Ilhan1, Pelin Güneri1, Ahmet Saraçoglu2, Hüseyin Koca3, Hayal Boyacioglu4,
1 Department of Oral and Maxillofacial Radiology, School of Dentistry, Ege University, Bornova, Izmir, Turkey
2 Department of Prosthodontics, School of Dentistry, Ege University, Bornova, Izmir, Turkey
3 Department of Oral and Maxillofacial Surgery, School of Dentistry, Ege University, Bornova, Izmir, Turkey
4 Department of Statistics, Faculty of Science, Ege University, Bornova, Izmir, Turkey
Correspondence Address:
Dr. Betül Ilhan
Department of Oral and Maxillofacial Radiology, School of Dentistry, Ege University, 35100, Bornova, Izmir
Turkey
Abstract
Context: Trabecular bone around dental implants has been rarely evaluated by means of fractal analysis. Aims: The aim was to assess fractal dimension (FD) values of peri-implant bone 12 months after implant installation and to compare these FD values with those of healthy control sites on the contralateral side of the implant area. Materials and Methods: Panoramic radiographs from 29 patients with 29 mandibular implants functioning for 12 months were analyzed. Radiographs were digitized at 300 dpi with 256 grey levels. Healthy trabecular bone from the corresponding contralateral implant-free area was referred as «DQ»control site.«DQ» Three region of interests (ROIs) from mesial, distal and apical areas were selected for each implant and each control site. The FD from peri-implant and control ROIs were calculated by means of box-counting method. Statistical analysis was performed using two-way ANOVA with repeated measure on one factor. The independent variable included a between-subjects variable, the site and within-subject variable, repeated measures of ROI. An alpha level of 0.05 was utilized. Results: Mean FD for apical, mesial and distal peri-implant areas were 1.220, 1.222 and 1.226 respectively while the corresponding results were 1.198, 1.183 and 1.202 for control sites. There was not a significant main effect between the FD values of implant and control sides (P = 0.332). Similarly, result of the main effect of ROI (apical, mesial, distal) on FD values was not statistically significant (P = 0.632). Conclusions: Fractal dimension values from trabecular bone around implants as calculated from panoramic radiographs at 12 months revealed comparable results with implant-free healthy alveolar bone.
How to cite this article:
Ilhan B, Güneri P, Saraçoglu A, Koca H, Boyacioglu H. A comparison of fractal dimension values of peri-implant bone and healthy contralateral side using panoramic radiographs.J Oral Maxillofac Radiol 2015;3:1-6
|
How to cite this URL:
Ilhan B, Güneri P, Saraçoglu A, Koca H, Boyacioglu H. A comparison of fractal dimension values of peri-implant bone and healthy contralateral side using panoramic radiographs. J Oral Maxillofac Radiol [serial online] 2015 [cited 2023 Apr 2 ];3:1-6
Available from: https://www.joomr.org/text.asp?2015/3/1/1/151636 |
Full Text
Introduction
Digital subtraction radiography, quantitative computed tomography, magnetic resonance imaging (MRI) and dual energy X-ray absorptiometry are among the common methods, which are employed to quantify osseous repair and regeneration. [1],[2],[3] Unfortunately, various drawbacks of these techniques including relatively high cost and inaccessibility, increased radiation burden, the need for stabilization of the subject for longer periods of time and difficulties to reproduce projection geometry have limited their use in clinical practice. [1],[4] Thus, fractal analysis (FA) was promoted as a reliable, economical and easily available method for bone quality determination. [5],[6]
Fractal analysis is based on fractal mathematics for describing complex shapes and structural patterns. [5],[6] It indicates a figure's complexity and is expressed numerically as "fractal dimension" (FD), which measures self-similarity. Trabecular bone has a branching pattern that exhibits fractal properties, such as self-similarity and lack of well-defined scale. [6],[7],[8] Therefore, when applied to trabecular bone images on radiographs, this method can be considered as a reflection of trabecular bone microarchitecture. [7],[9],[10] In addition, it is reported that FA is not affected by variations in exposure, alignment and choice of region of interest (ROI), which may be another advantage of this method in clinical settings. [11],[12]
In dental field, FD values have been mostly used to study bone quality in osteoporotic patients and to determine the impact of periodontal disease on surrounding bone. [12],[13],[14],[15],[16],[17],[18] It has been reported that FD analysis is highly successful in distinguishing between radiographs of osteoporotic and nonosteoporotic bone; the higher the FD, the more the morphological complexity at ultra-structural level. [6],[19],[20] Some researchers have performed FAs on dental radiographs to compare the trabecular pattern differences between healthy people and moderate periodontitis patients. [15],[17] It has been shown that FD values from clinical radiographs of mandibular alveolar bone of patients with moderate or severe periodontitis were significantly lower than FDs calculated from healthy controls. [17] In endodontic field, Chen et al. [1] reported an increase in FD values measured from the periapical area on intraoral radiographs after root successful canal treatment. Additionally, Heo et al. [21] also showed a gradual increase in FD during the bony healing process after orthognathic surgery.
Even though, previous studies showed the potential application of FD in analyzing trabecular structure on periapical or panoramic radiographs, trabecular bone around dental implants has been rarely evaluated by means of FA. [22],[23] The aim of this retrospective study was to assess the FD values of trabecular peri-implant bone 12 months after implant installation on panoramic radiographs, and to compare these FD values with those of healthy control sites on the contralateral side of the implant area.
Materials and Methods
In this retrospective study, panoramic radiographs from mandibular molar/premolar implants functioning for around 12 months were screened from the patient records of Ege University, School of Dentistry, Izmir, Turkiye. The patients with clinically asymptomatic implants, high quality and correctly positioned panoramic radiographs and symmetrical natural teeth were included into the study. Radiographs from patients who had conditions or were under medications affecting bone metabolism or bone turnover were excluded. Informed consents were obtained from all patients prior to their enrollment. Panoramic radiographs from 29 healthy patients (14 male, 15 female, mean age: 38.4 years) with 29 mandibular molar/premolar implants functioning for 12 months ±2 weeks were selected for the study. All implants were from the same manufacturer (FRIALIT/Xive Dentsply/Friadent, Germany). The implants had been inserted by the same experienced oral surgeon (H.K.) using conventional methods, and were loaded 4 months ±2 weeks after insertion. All prosthodontic rehabilitations were performed by the same specialist (A.S.).
The panoramic images were achieved using the same panoramic radiography unit (Orthopantomograph OP100, Instrumentarium Imaging, Tuusula, Finland), with a 2.5-mm Al equivalent total filtration, Lanex Medium Screen cassette (Eastman Kodak Co., Rochester, NY), and Ceadent DG films (15 cm × 30 cm, SE-64523, Strängnäs, Sweden) at 70 kVp, 16 mA, for 17.6 s. The films were processed with fresh developing solutions (Hacettepe Defiks, Ankara, Turkey) in an automatic processor (Velopex Extra-X, Medivance Instruments, X/41348, London, UK) at 28°C for 5 min, 15 s. The analog film panoramic radiographs were digitized using a flatbed scanner that has a transparency adapter (Epson EXP 1680Pro, Seiko Epson Corp., Nagano, Japan) with 8-bit gray-scale acquisition depth and 300 dpi spatial resolution. The image size was 1420 × 810 pixels with 256 gray levels. After digitization, all digital images were saved in Tagged Image File Format into a personal computer.
The procedures for calculation of the FD were performed using ImageJ 1.38x software, a version of the National Institutes of Health Image (US, NIH http://www.rsb.info.nih.gov/nih-image). NIH Image is a public domain program that can be downloaded from the World Wide Web ( http://www.rsb.info.nih.gov/ij/download.html (Accessed on 04 Dec 2013, Java 1.3.1_13). Healthy trabecular bone from the contralateral implant-free dentate area was referred as the control site. Three ROIs from mesial, distal and apical areas were selected for each implant (close to the neck and apex) and contralateral healthy control site. ROIs were selected arbitrarily within each image to provide maximum available area for measurement. The influence of transient alterations within alveolar bone was avoided and lamina dura, periodontal ligament and root apices were not included within ROIs.
The ROIs were outlined manually in each radiograph using Adobe Photoshop CS2 (Adobe Systems Inc., San Jose, CA) [Figure 1]. Thereafter, the ROIs were cropped and transferred to ImageJ 1.38x with the program menu. The saved images were processed using the method designed by White and Rudolph. [24] The cropped ROI was duplicated and the duplicated image was blurred with a Gaussian filter (kernel size = 30) to remove the fine and medium scale variations in image brightness. The blurred image was subtracted from the original image and 128 was added at each pixel location. The resultant image was converted to binary with threshold at the grey value of 128. As a result, the segmented objects approximated the bony trabecular pattern. The binary image was eroded and dilated once to reduce the noise before skeletonization. Finally, the image was skeletonized and used for FA [Figure 2]. On the skeletal binary image the skeletal structure indicated the bone pattern, whereas the nonskeletal structure represented the bone marrow. All digital manipulations and measurements were made within the ROIs. The FD of the skeletonized image was calculated with ImageJ 1.38x by using the box counting function from the "analyze" menu. Initially, the image was covered by a square grid of equally sized tiles, and the number of tiles referring to the trabecular bone was counted. The widths of the square boxes were 2, 3, 4, 6, 8, 12, 16, 32 and 64 pixels. Then, the number of the counted tiles was plotted against the total number of tiles in a double logarithmic scale. Finally, FD was calculated from the slope of the line fitted on the data points.{Figure 1}{Figure 2}
Kolmogorov-Smirnov statistics was used for the normality test. A two-way ANOVA with repeated measure on one factor was conducted to determine whether there was a statistical significance between implant and control sites. The independent variable included a between-subjects variable, the site and within-subject variable, repeated measures of ROI (mesial-distal-apikal). An alpha level of 0.05 was utilized for this analysis. All statistical analyses were performed using the Statistical Package for Social Sciences (Version 10.0, SPSS Inc., Chicago, Illinois, USA).
Results
Results for model assumptions of normality, homogeneity of covariance (P = 0.725 > 0.05) and linearity were satisfactory. The mean FD values and standard deviations for implant and healthy control sites are presented in [Table 1].{Table 1}
The mean FD values for implant sites were 1.220 ± 0.014, 1.222 ± 0.019 and 1.226 ± 0.019 for apical, mesial and distal areas respectively. The corresponding mean FD results were calculated as 1.198 ± 0.016, 1.183 ± 0.019 and 1.202 ± 0.018 for the healthy control sites.
There was not a significant main effect between the FD values of implant and control sides (P = 0.332). Likewise, when the groups were analyzed according to ROI (mesial, distal, apical), the result of the main effect of ROI on FD values was not statistically significant (P = 0.632).
Discussion
Our results indicated that FD values of trabecular bone around implants as calculated from panoramic radiographs at 12 months revealed comparable results with implant-free healthy alveolar bone, suggesting complete bone healing around implants.
Trabecular bone around dental implants has been rarely evaluated by means of FA before. In a recent study, Corpas Ldos et al. [25] have evaluated peri-implant bone tissue 3 months after implant placement by using FA on conventional intraoral, cone beam computed tomography and histological images. They stated that, even though, the bone level measurements correlated in all methods, the FA method did not detect the changes and failed to correlate with true histological results. Recently, Veltri et al. [26] used an animal model to investigate the correlations between FA results and insertion torque and resonance frequency analysis (RFA), which are initial markers of biomechanical competence of an implant. They reported that FD results significantly correlated with final insertion torques, but not with resonance frequencies. Their results also revealed that FD value of 1.83 was defined as the breakpoint of soft bone quality. Wilding et al. [23] measured FD values on serial panoramic radiographs from patients with fixed implant-supported prosthesis. They selected the bone adjacent and distal to the posterior implant as the ROI and found that FD increased with trabecular remodeling occurring up to 2 years after implantation, mostly in the region of bone around the neck of the implant. In a previous study by Veltri et al. [22] the relationship between damping values which were measured with the Osstell™ equipment and FD values at peri-implant bone were investigated and no significant correlation was observed. In that study, FD was measured by box-counting method on intraoral radiographs from maxillary implants functioning for 3 years. Mesial and distal aspects of each implant were selected as ROIs (47 × 47 pixels) and the mean FD was reported as 1.47. In a recent study Lee et al. [27] investigated, whether the FD from panoramic radiograph is related to the primary stability of the implant as represented by RFA. Panoramic images of 22 patients with 52 implants were taken before surgery and just after installation. The fixed size for ROIs was 60 × 90 pixels around the implant installation areas, and FD was measured by tile-counting method. Their results revealed a statistically significant correlation between FD values and implant stability quotient of RFA, with mandibular implants presenting a higher correlation. They also suggested that FD acquired from panoramic radiographs may be a useful predictor of the initial stability of dental implants. Unfortunately, FD values from healthy control sites were not measured in previous investigations. In the present study, FD values for implant sites were 1.220 ± 0.014, 1.222 ± 0.019 and 1.226 ± 0.019 for apical, mesial and distal areas respectively with no significant differences from corresponding control sites. FD was calculated using box counting method, which is the most widely used technique to quantify the trabecular pattern.[17],[28],[29],[30],[31] Our results concerning FD are lower than those reported by Veltri et al. [22] which may be influenced by the differences in selected jaw areas and discrepancies in the selection of ROIs. In contrast to their study design, only mandibular premolar/molar implants were included in the present study in order to eliminate the superimposition of anatomical entities such as maxillary sinus and hard plate. The difference between the results can also be considered as to reflect the influence of physiological function and mechanical loading on the trabecular bone architecture. [32],[33]
In our study, it was difficult to select ROIs with a fixed size due to the variations in individual anatomy and amount of available bone of each patient. Therefore, in order to retain as much useful information as possible from the radiographs, selection of ROIs was performed arbitrarily rather than cropping to the smallest common size. [17] Veltri et al. [22] have utilized 47 × 47 pixels ROI in contact with the implants in their early investigation but have enlarged the size of ROIs to 90 × 168 pixels in their recent study. [26] Lee et al. used standard 60 × 90 pixels ROI area around implant installation sites. [27] On the other hand, Corpas Ldos et al. [25] have preferred 4 mm × 1 mm ROIs, which corresponded to the implant from the first to last thread. We attempted to use ROIs as large as the size of the implants and provide larger areas for fractal measurements. In FA, which is actually a mathematical process, using small ROIs would prevent to establish the empty and full boxes (in other words, FDs) in separate images and thus, would not accurately reveal the fractal differences between the interested areas. [17]
Analyses of FD have been mostly correlated with the density and quality of bone and the presence of osteoporosis in previous studies. [12],[13],[14],[15],[16],[17],[18] Southard et al. have stated that there is a positive relationship between the FD and the density of alveolar bone and as the bone density increases so does the FD. [34] Ruttimann et al. [6] have reported an increase in FD of the mandible after experimental demineralization, while Southard et al. [35] showed a decrease in mean FD in radiographs of decalcified human alveolar bone. Nair et al. [4] and Heo et al. [21] reported that the FD increased during the bone recovery process. Similarly, Wojtowicz et al. [36] showed the increasing complexity of trabecular patterns as the bone grew by using the FD of a section from infants' maxillas. In an early study; Bollen et al. [7] measured higher FD values on the panoramic radiographs of patients with a thinner, severely eroded mandibular cortex and a history of osteoporotic fractures. Yasar and Akgünlü [32] and Ergün et al. [29] observed that differences in occlusal forces generated in the dentate and edentulous regions during mastication cause alterations in trabecular bone structure, which lead to lower FDs in dentate regions. Comparing FD values from periapical radiographs of patients with gingivitis and periodontitis, Shrout et al. [15] reported that the periodontitis group has significantly lower mean FD value, which suggested that periodontal health seemed to be positively correlated with the fractal index and as periodontal health deteriorated, fractal index decreased. In endodontic field, while Chen et al. [1] reported an increase in FD values measured from the periapical area on intraoral radiographs, Yu et al. [31] observed a decrease of FD in reactive bone regions, both after clinically successful endodontic treatments.
The contradiction between the results from these studies indicates that; at this time there is no consensus on the relationship between FD and trabecular bone complexity. Some findings support the idea that FD increases in the diseased and osteoporotic state, while others support that the diseased state reduces trabecular complexity and decreases FD. It is known that trabecular bone shows directional anisotropy of its mechanical properties and architecture, depending on its physiological function and mechanical loading on the skeleton. [33] Therefore, it may be logical to suggest that both assumptions may be correct, depending on the disease that affects trabecular bone and how it destroys the fine trabeculae in different parts of the body. Additionally, the possible influence of anatomical variations, the technique used to measure FD and the differences in selected areas should be noted.
It is necessary to examine the peri-implant bone at intervals in order to confirm the presence of adequate bone and detect signs of failing integration at an early stage. It has been shown that the conditions surrounding radiography, such as irradiation angle or irradiation quantity, do not affect the FD of the trabecular pattern. [11],[12] Therefore, FA held the promise as an economical and easily available method to assess the changes in trabecular bone. In our investigation, FD values of trabecular bone around implants as calculated from panoramic radiographs at 12 months revealed comparable results with those of implant-free healthy alveolar bone, suggesting a complete healing around the implants.
Until date, studies that investigated the applicability of FA and FD measurements in order to establish the dental implant stability and bone healing around the implants have led to both approving and opposing conclusions. As Lee et al. [27] have suggested a unified method to calculate the FD shall be provided. After application of this standardized process to larger groups of subjects, and comparison of this method both with histological parameters and with other measures that define the bone quality, the value of FD would be more precisely assessed.
References
| 1 | Chen SK, Oviir T, Lin CH, Leu LJ, Cho BH, Hollender L. Digital imaging analysis with mathematical morphology and fractal dimension for evaluation of periapical lesions following endodontic treatment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005;100: 467-72. |
| 2 | Drage NA, Palmer RM, Blake G, Wilson R, Crane F, Fogelman I. A comparison of bone mineral density in the spine, hip and jaws of edentulous subjects. Clin Oral Implants Res 2007;18:496-500. |
| 3 | Simmons CA, Hipp JA. Method-based differences in the automated analysis of the three-dimensional morphology of trabecular bone. J Bone Miner Res 1997;12:942-7. |
| 4 | Nair MK, Seyedain A, Webber RL, Nair UP, Piesco NP, Agarwal S, et al. Fractal analyses of osseous healing using tuned aperture computed tomography images. Eur Radiol 2001;11:1510-5. |
| 5 | Fazzalari NL, Parkinson IH. Fractal dimension and architecture of trabecular bone. J Pathol 1996;178:100-5. |
| 6 | Ruttimann UE, Webber RL, Hazelrig JB. Fractal dimension from radiographs of peridental alveolar bone. A possible diagnostic indicator of osteoporosis. Oral Surg Oral Med Oral Pathol 1992;74:98-110. |
| 7 | Bollen AM, Taguchi A, Hujoel PP, Hollender LG. Fractal dimension on dental radiographs. Dentomaxillofac Radiol 2001;30:270-5. |
| 8 | Geraets WG, van der Stelt PF. Fractal properties of bone. Dentomaxillofac Radiol 2000;29:144-53. |
| 9 | Parkinson IH, Fazzalari NL. Methodological principles for fractal analysis of trabecular bone. J Microsc 2000;198:134-42. |
| 10 | 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;102:235-41. |
| 11 | Jolley L, Majumdar S, Kapila S. Technical factors in fractal analysis of periapical radiographs. Dentomaxillofac Radiol 2006;35:393-7. |
| 12 | Shrout MK, Potter BJ, Hildebolt CF. The effect of image variations on fractal dimension calculations. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997;84:96-100. |
| 13 | Law AN, Bollen AM, Chen SK. Detecting osteoporosis using dental radiographs: A comparison of four methods. J Am Dent Assoc 1996;127:1734-42. |
| 14 | Prouteau S, Ducher G, Nanyan P, Lemineur G, Benhamou L, Courteix D. Fractal analysis of bone texture: A screening tool for stress fracture risk? Eur J Clin Invest 2004;34:137-42. |
| 15 | Shrout MK, Roberson B, Potter BJ, Mailhot JM, Hildebolt CF. A comparison of 2 patient populations using fractal analysis. J Periodontol 1998;69:9-13. |
| 16 | Southard TE, Southard KA, Krizan KE, Hillis SL, Haller JW, Keller J, et al. Mandibular bone density and fractal dimension in rabbits with induced osteoporosis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000;89:244-9. |
| 17 | Updike SX, Nowzari H. Fractal analysis of dental radiographs to detect periodontitis-induced trabecular changes. J Periodontal Res 2008;43:658-64. |
| 18 | Yasar F, Akgünlü F. The differences in panoramic mandibular indices and fractal dimension between patients with and without spinal osteoporosis. Dentomaxillofac Radiol 2006;35:1-9. |
| 19 | Caligiuri P, Giger ML, Favus M. Multifractal radiographic analysis of osteoporosis. Med Phys 1994;21:503-8. |
| 20 | Webber RL, Underhill TE, Horton RA, Dixon RL, Pope TL Jr. Predicting osseous changes in ankle fractures. A fractal-based radiographic analysis. IEEE Eng Med Biol 1993;12:103-10. |
| 21 | Heo MS, Park KS, Lee SS, Choi SC, Koak JY, Heo SJ, et al. Fractal analysis of mandibular bony healing after orthognathic surgery. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002;94:763-7. |
| 22 | Veltri M, Balleri P, Ferrari M. Damping factor for monitoring the bone interface at dental implants. Clin Oral Implants Res 2007;18:738-42. |
| 23 | Wilding RJ, Slabbert JC, Kathree H, Owen CP, Crombie K, Delport P. The use of fractal analysis to reveal remodelling in human alveolar bone following the placement of dental implants. Arch Oral Biol 1995;40:61-72. |
| 24 | White SC, Rudolph DJ. Alterations of the trabecular pattern of the jaws in patients with osteoporosis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1999;88:628-35. |
| 25 | Corpas Ldos S, Jacobs R, Quirynen M, Huang Y, Naert I, Duyck J. Peri-implant bone tissue assessment by comparing the outcome of intra-oral radiograph and cone beam computed tomography analyses to the histological standard. Clin Oral Implants Res 2011; 22:492-9. |
| 26 | Veltri M, Ferrari M, Balleri P. Correlation of radiographic fractal analysis with implant insertion torque in a rabbit trabecular bone model. Int J Oral Maxillofac Implants 2011;26:108-14. |
| 27 | Lee DH, Ku Y, Rhyu IC, Hong JU, Lee CW, Heo MS, et al. A clinical study of alveolar bone quality using the fractal dimension and the implant stability quotient. J Periodontal Implant Sci 2010;40:19-24. |
| 28 | Demirbas AK, Ergün S, Güneri P, Aktener BO, Boyacioglu H. Mandibular bone changes in sickle cell anemia: Fractal analysis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;106:e41-8. |
| 29 | Ergün S, Saraçoglu A, Güneri P, Ozpinar B. Application of fractal analysis in hyperparathyroidism. Dentomaxillofac Radiol 2009;38: 281-8. |
| 30 | Hua Y, Nackaerts O, Duyck J, Maes F, Jacobs R. Bone quality assessment based on cone beam computed tomography imaging. Clin Oral Implants Res 2009;20:767-71. |
| 31 | Yu YY, Chen H, Lin CH, Chen CM, Oviir T, Chen SK, et al. Fractal dimension analysis of periapical reactive bone in response to root canal treatment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009;107:283-8. |
| 32 | Yasar F, Akgünlü F. Fractal dimension and lacunarity analysis of dental radiographs. Dentomaxillofac Radiol 2005;34:261-7. |
| 33 | Yi WJ, Heo MS, Lee SS, Choi SC, Huh KH, Lee SP. Direct measurement of trabecular bone anisotropy using directional fractal dimension and principal axes of inertia. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:110-6. |
| 34 | Southard TE, Southard KA, Lee A. Alveolar process fractal dimension and postcranial bone density. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001;91:486-91. |
| 35 | Southard TE, Southard KA, Jakobsen JR, Hillis SL, Najim CA. Fractal dimension in radiographic analysis of alveolar process bone. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996;82:569-76. |
| 36 | Wojtowicz A, Chaberek S, Pirino A, Montella A, Bandiera P, Kinsner A, et al. The trabecular structure of developing human mandible. Clin Orthod Res 2001;4:161-71. |
|
