Acharya, Ashith B. BDS, GDFO
From the Department of Forensic Odontology, S.D.M. College of Dental Sciences and Hospital, Dharwad, Karnataka, India.
Manuscript received January 30, 2008; accepted June 30, 2008.
Reprints: Ashith B. Acharya, BDS, GDFO, Department of Forensic Odontology, S.D.M. College of Dental Sciences and Hospital, Sattur, Dharwad 580009, Karnataka, India. E-mail: email@example.com.
Estimating age is an important step in establishing identity of postmortem remains. Teeth are one of the strongest structures in the human body and usually survive postmortem destruction. Hence, they play a major role both in comparative as well as reconstructive identification. Scientific methods to evaluate dental age changes have developed over the last century, with Gustafson's morpho-histologic approach1 occupying a prominent position. More recently, biochemical methods have been developed and claim to give precise age estimates.2,3 However, these are time- and technique-intensive and may not easily be accomplished in the average forensic/dental set-up. Consequently, some of the parameters suggested by Gustafson1 still find favor with odontologists, both in terms of research and casework application. Studies on Gustafson's 6 variables found that dentin translucency was best suited for age estimation when used alone.4 Miles believed that measuring the regressive changes was a better approach for age assessment than visually grading them, and attempted age estimation from measuring translucency.5 Bang and Ramm further developed this approach and considered it to be a simple yet relatively objective method for age assessment.6 Traditionally, translucency has been quantified with the aid of vernier calipers.6,7 However, attempts to quantify translucency using digital aids have been proposed over the last 2 decades,8,9 with one group of researchers concluding that computer-based translucency measurements contributed best to age estimation.10 These methods require capturing tooth images on a video camera, converting the analog signal to digital and subsequent image processing using customized software programs. With advances in computing technology, digital evaluation of translucency can be more easily accomplished today. The aim of this article is to describe a new and comparatively simple method for capturing and quantifying dentinal translucency using standard computer hardware and commercially available computer software. Also, to ascertain the method's effectiveness in forensic age estimation, the author compares digital measurements to conventionally obtained translucency measurements.
MATERIALS AND METHODS
Eighty-one permanent teeth from as many individuals in the age group 19 to 82 years (mean age, 51.6 years), extracted for valid clinical reasons such as malocclusion/orthodontic treatment, periodontal disease and caries, were obtained from the Department of Oral and Maxillofacial Surgery of this institution and from private practices of the region. Carious teeth were included in the sample contingent to the roots being unaffected macroscopically by disease. The extracted teeth were thoroughly cleaned and soft tissue remnants removed from the root surface with a scalpel. Teeth were kept in 10% formalin and, following fixation, mounted in autopolymerizing acrylic for sectioning by hard-tissue microtome (Leica SP 1600, Leica Microsystems, Germany). Mounted teeth were sectioned longitudinally to 250 μm in the buccolingual plane as close as was possible to the central axis of the tooth.
Conventional Translucency Measurement
Tooth sections were placed in front of a constant light source and the maximum extent of dentin translucency determined to the nearest 0.1 mm using a caliper with digital reading (Altraco Inc., Sausalito, CA). The measurements were taken between the apical limit and the most coronal extent of translucency within the root. Measurements were made by the author who had no prior information about personal data of the subjects.
Digital Translucency Measurement
The computer hardware used in the method included a Compaq Presario 2.8 GHz CPU with 512 MB RAM and 17 inch CRT monitor (Hewlett-Packard Co., Palo Alto, CA) and a HP Scanjet G3010 flat-bed scanner (Hewlett-Packard Co., Palo Alto, CA). Each tooth section was placed next to an ABFO No. 2 scale (Lightning Powder Co. Inc., Jacksonville, FL) on the scanner platen. The long axis of the section was aligned parallel to the y-axis of the scale. Prior to scanning, the scanner setting was verified to be 100% of the original to ensure life-size scanned images. Subsequently an image of 600 dpi resolution of the section with scale was obtained (Fig. 1). The scanner lid was kept open while scanning and ambient light conditions kept to a minimum (note: keeping the lid closed obstructs passage of the optical scanning light through the translucent zone, rendering the entire tooth section opaque). Scanned images were imported to Adobe Photoshop version 7.0.1 image-editing software (Adobe Systems Inc., Mountain View, CA) for viewing and measuring the extent of translucency. The different dental tissues are generally appreciable on the image and dentinal translucency, in particular, appears as a dark region on the section (Fig. 1). Translucency was measured using a number of tools available on Adobe Photoshop. The method for measuring translucency that follows has been adapted from different steps described by Johansen and Bowers11 for digital analysis of bitemark evidence.
For convenience of measuring apical and coronal extent of translucency, “guides” were placed on the image (Fig. 2). These guides can be activated by inserting Photoshop's in-built “rulers” along the edges of the image (on the Menu Bar choose View > Rulers, or Ctrl + R, or Command + R for Macintosh systems). Once the rulers are activated, guides are placed by clicking the cursor within the x-axis (horizontal part) of the ruler and dragging onto the image. Click and drag as many guides as may be required onto the image but it is anticipated that 2 should suffice. To move a guide, the Move Tool is used; alternatively, the Ctrl key is held down (Command key for Macintosh systems) and the guide moved to the desired location. Once the respective guide has been placed at the apical and coronal extent of root dentin translucency, the distance between them can be obtained using the Measure Tool on the Toolbox (Fig. 2). Using this tool, a line is drawn between the guides; the distance (D1) is displayed in the Options Bar. If the Options Bar is not displayed, it can be activated by choosing Window > Options. Measurements obtained using the Measure Tool are sensitive to 0.1 mm. The measuring line drawn can be kept vertical by holding down the Shift key. The units were ensured to be in mm by comparing with the reference ABFO No. 2 scale. In the event units are not in mm, choose Edit > Preferences > Units and Rulers and select “mm” under Units and click OK.
Owing to the relatively small sample size, all tooth sections—irrespective of sex or tooth type—were pooled together; age was the only variable isolated in the statistical analysis. A paired t test was performed to evaluate potential differences between conventional and digital measurements. Paired t test was also performed to evaluate potential intraobserver variation for both conventional and digital measurements (translucency measurements were repeated using both methods on at least 38 tooth sections). In addition, translucency measurements obtained from both methods were correlated to known age using linear regression analysis. Statistical analyses were performed on an MS Office 2007 Excel spreadsheet (Microsoft Corp., Redmond, WA) and SPSS 10.0 statistical software program (SPSS Inc., Chicago, IL). Pearson's correlation coefficients obtained for both methods were noted and the regression equations derived used to calculate age on a control sample of 15 sections (obtained from 15 subjects whose ages ranged between 35 and 70 years). These sections were not used in deriving the regression formulas. The difference between estimated and known age for both methods were compared.
The mean of all digital measurements was 5.5 mm as against 5.7 mm for conventional measurements. However, paired t test to evaluate differences between digital and conventional measurements revealed no statistically significant differences (P = 0.71). The paired t test to assess intraobserver bias also revealed no statistical differences within both conventional and digital methods (P > 0.05).
Pearson's correlation coefficients (r) and linear regression equations are shown in Table 1. Figures 3A, B depicts diagrammatically the relationship between translucency measurements and known age for conventional and digital methods. The correlation coefficients, while statistically significant for both methods (P < 0.001), was higher for digital measurements (r = 0.49) vis-à-vis conventional measurements (r = 0.46). Application of linear regression equations on the control sample (n = 15) showed that the digital method could estimate age to within ±5 years in 9 of 15 cases (60%) as against 6 of 15 cases (40%) for the conventional method (Table 2).
Assessment of dentin translucency is one of the most simple to assess among Gustafson's 6 variables.1 According to Miles5 and Bang and Ramm,6 an advantage of translucency measurements is that a relatively inexperienced examiner can use it to estimate age. Indeed, translucency can be assessed macroscopically on intact teeth, although tooth sections provide better detail.6 Sectioned teeth were used in the present report for this particular reason. Moreover, scanning unsectioned teeth did not facilitate translucency observation or measurement.
Advantages of the Digital Method
Digital recordings of translucency length were first made on life-size images and then verified at ×3 magnification using the Zoom Tool within Photoshop (desired magnifications can also be selected on the Navigator Palette by choosing Window > Navigator). This allowed better visualization of the junction between translucent and nontranslucent zones, giving scope for “fine-tuning” the measurements. Consequently, minor differences in measurements between the magnified and unmagnified images were generally observed, and it is plausible that these refined measurements contributed to better correlation coefficients for the digital method (Table 1). Hence, one may infer that some form of magnification is essential, which is easily accomplished on the software program.
Conventional measurements were made using a caliper without magnification aids and use of a magnifier could have enhanced clarity of translucent and opaque zones. However, irrespective of magnification, an impediment to caliper-based measurements is that the caliper beaks could not always be stabilized on the 250 μm tooth sections since there was risk of damage to the thin sections from the pointed beaks. Therefore, calipers are probably better suited for measuring translucency on intact teeth. On the other hand, the “touch-free” or “noninvasive” digital evaluation prevents potential damage to thin tooth sections.
Applicability of Digital Translucency Measurement in Age Estimation
The absence of statistically significant differences between digital and conventional measurements (P = 0.71) indicates that digital quantification can substitute for conventional measurement. In fact, digital measurements are better correlated to age (Table 1; Figs. 3A, B), which is in contrast to previous results where conventional measurements had superior correlation.8,10 However, one of these studies observed that dentin translucency was the single-most useful parameter for assessing age using digital methods.10 In the present report, age calculation using linear regression equations on the control sample (n = 15) showed better ability for the digital method to assess age—age was estimated to within ±5 years in 60% of cases as against 40% for the conventional method (Table 2). Hence, although Pearson's correlation coefficients exhibit minimal variation between the 2 methods (Table 1), application of regression formulas derived from them on an independent sample reveals recognizable differences, with the digital method emerging superior to the conventional one. However, both methods have an equal tendency to either over- or underestimate age, and no apparent difference could be observed in their ability to reliably estimate age in younger or older age-groups (data not shown).
Comparison of Present and Previous Digital Methods
The digital approach used in the present study can be categorized as a semi-automatic analysis since the limits of translucency are designated manually. Previous studies8,9 have also reported semi-automatic analysis of translucency, with one of them developing a customized program for the purpose.9 In addition, these methods required capturing images of teeth using a video camera and converting the analog signal to digital.8,9 An advantage of the present method is that the software program used is commercially available and a widely used image-editing digital aid. Also, digitizing tooth sections is straightforward with current computer hardware and software. Furthermore, accuracy of age estimation using the present digital method (60%) is much higher to that of a previous one, where only 21.4% of cases were estimated to within ±5 years of known age.8 On the other hand, the conventional method's accuracy levels for both studies are comparable—40% of control cases estimated to within ±5 years in the present study as against 42.9% obtained earlier.8 This could imply that the digital method reported here is superior. Hence, contrary to the conclusion drawn previously that translucency can be assessed either by digital or conventional methods,8 this author recommends application of the digital method presented here. Indeed, López-Nicolas et al12 have been critical of image analysis systems used in earlier studies8,9 for digital quantification of translucency and recommend exploring alternatives. Using the computer software used in the present study or similar commercially available image-editing programs could be a viable alternative. The equipment required to analyze translucency is readily available and tooth sections can be digitized easily. The images can be stored and conveniently retrieved for future use, irrespective of the condition of the actual tooth section.
In conclusion, a new and relatively simple method for measuring dentin translucency has been developed using commercially available digital aids. The measurements obtained using this method is more refined, better correlated to age, and produce superior age estimates when compared with conventional caliper-based quantification. Considering that computer systems and software continue to develop rapidly, one can expect further improvements for translucency assessment in the near future.
The author thanks Dr. Vimi S. Mutalik for assistance in collecting the samples. The author is also grateful to Prof. C. Bhasker Rao, Director of the institution, for his continued support to research in forensic odontology and approving an institutional grant for the study.
1. Gustafson G. Age determination on teeth. J Am Dent Assoc
2. Ohtani S. Estimation of age from the teeth of unidentified corpses using the amino acid racemization method with reference to actual cases. Am J Forensic Med Pathol
3. Spalding KL, Buchholz BA, Bergman LE, et al. Age written in teeth by nuclear tests. Nature
4. Johanson G. Age determination from teeth. Odontologisk Revy
. 1971;22(suppl 21):90.
5. Miles AE. Dentition in the estimation of age. J Dent Res
. 1963;42(suppl 1):255–263.
6. Bang G, Ramm E. Determination of age in humans from root dentine transparency. Acta Odontol Scand
7. Lamendin H, Baccino E, Humbert JF, et al. A simple technique for age estimation in adult corpses: the two criteria dental method. J Forensic Sci
8. Drusini A, Calliari I, Volpe A. Root dentine transparency: age determination of human teeth using computerized densitometric analysis. Am J Phys Anthropol
9. Mandojana JM, Martin-de las Heras S, Valenzuela A, et al. Differences in morphological age-related dental changes depending on postmortem interval. J Forensic Sci.
10. Valenzuela A, Martin-de las Heras S, Mandojana JM, et al. Multiple regression models for age estimation by assessment of morphologic dental changes according to teeth source. Am J Forensic Med Pathol.
11. Johansen RJ, Bowers CM. Digital Analysis of Bite Mark Evidence Using Adobe Photoshop.
Santa Barbara, CA: Forensic Imaging Services; 2000.
12. López-Nicolas M, Morales A, Luna A. Morphometric study of teeth in age calculation. J Forensic Odontostomatol
© 2010 Lippincott Williams & Wilkins, Inc.