J Forensic Sci, September 2014, Vol. 59, No. 5 doi: 10.1111/1556-4029.12467 Available online at: onlinelibrary.wiley.com
TECHNICAL NOTE ANTHROPOLOGY Nestor A. Perlaza,1 M.Sc.
Sex Determination from the Frontal Bone: A Geometric Morphometric Study
ABSTRACT: Sex estimation in human skeletal remains when using the cranium through traditional methods is a fundamental pillar in human identification; however, it may be possible to incur in a margin of error due because of the state of preservation in incomplete or fragmented remains. The aim of this investigation was sex estimation through the geometric morphometric analysis of the frontal bone. The sample employed 60 lateral radiographs of adult subjects of both sexes (30 males and 30 females), aged between 18 and 40 years, with mean age for males of 28 4 and 30 6 years for females. Thin-plate splines evidenced strong expansion of the glabellar region in males and contraction in females. No significant differences were found between sexes with respect to size. The findings suggest differences in shape and size in the glabellar region, besides reaffirming the use of geometric morphometrics as a quantitative method in sex estimation.
KEYWORDS: forensic science, geometric morphometrics, sex estimation, forensic, frontal bone, thin-plate spline, sexual dimorphism
Sex determination in human skeletal remains is one of the initial bioanthropological characteristics investigated by forensic anthropologists, whose purpose is the construction of a biologic profile that permits identifying the individual (1). Most of the axial and appendicular skeleton has been addressed for sex determination, with varying levels of precision (2). These methods may have no practical value when skeletal remains are incomplete or fragmented. Thereby, analysis must be made on the bones available and is dependent upon their state of preservation (3). The skull is one area of the skeleton most often used in sex determination. Its analyses yield different degrees of precision through morphological traits (4,5), discriminant functions (6,7), and through geometric morphometrics (8–10). Landmark-based geometric morphometrics permit differentiation of anatomic traits in a quantitative manner; Morphometrics relate the shape and size of anatomic traits using the location of landmarks on the structures (11–13). In general, there has been a global approach to selecting landmarks of the skull, which encompasses anatomic references to bones of the neurocranium and viscerocranium (8–10,14–16). The aim of this research was sex estimation through geometric analysis of the frontal bone. Materials and Methods The sample was comprised of 60 lateral radiographs of adult subjects of both sexes (30 males and 30 females), ranging in age between 18 and 40 years, with mean ages for males of 28 4 and 30 6 years for females. The radiographs were obtained 1 Department of Basic Sciences, Faculty of Health, Universidad Libre, Cali 760043, Colombia. Received 27 Feb. 2013; and in revised form 5 July 2013; accepted 29 July 2013.
1330
from the diagnostic imaging center, DIAGRAMAX, specializing in oral and maxillofacial radiology in the city of Cali (Colombia). The parameters for taking the radiographs were as follows: magnification 1:1; 7 mA; 76 KV; total time of exposition: 8 sec. The images were stored in JPG format. The subjects were classified to avoid duplication, check for radiograph error, and safeguard confidentiality. Additionally, any of the subjects with a history of craniofacial trauma were excluded. All ethical and legal requirements were followed. The landmarks selected on the image were digitized in the squamous (landmarks 1 and 2) and nasal part (landmarks 3 and 4) of the frontal bone (Fig. 1). For this study, the thin-plate spline (TPS) series of programs was used (17,18). According to Bookstein (11), the landmarks are classified as follows: type I permits identifying changes on the structure, and these are tissue transition zones (e.g., intersection between cranial sutures); type II are points of great biomechanical tension (e.g., bony attachment); and type III are defined as curvature endpoints. The landmarks selected are described in Table 1. Geometric Morphometric Analysis A generalized Procrustes analysis (GPA) was developed, which offered a size and confirmation variables matrix upon minimizing the sum of squared distances between landmarks. Visualization in TPS was applied to graphically represent the relative deformations. Likewise, a multivariate analysis was performed through principal component analysis (PCA) to describe the shape. The prior analyses were carried out with the TPS series of programs (17,18). To analyze size differences, centroid size (CS) was used through the PAST software (http://folk.uio. no/ohammer/past); in addition, a discriminant analysis was employed to evidence the differences among the groups when maximizing the differences among themselves. © 2014 American Academy of Forensic Sciences
PERLAZA
.
A GEOMETRIC MORPHOMETRIC STUDY TO SEX DETERMINATION
FIG. 3––Scatter plot showing scores of first two PCs. Males are indicated by crosses and females by squares.
FIG. 1––Lateral radiograph with digitizing of the landmarks.
TABLE 1––List of landmarks. No 1.
Landmark
2.
Supratoral sulcos Glabella
3.
Nasion
4.
B. point
1331
Description
Type
Turning point where the protrusion of the glabellar zone starts The most anterior point of the squamous part midline The intersection between the frontonasal and the internasal suture The union between the frontal bone, nasal, and the maxilla
III III I I
Results Figure 2 shows the deformation in thin-plate splines, which details a contraction of the grid in the image corresponding to females and an expansion in the grid corresponding to males. Glabella and nasion are the landmarks where changes predominate in the magnitude of their directions for both sexes.
FIG. 2––Deformation of TPS showing changes in shape according to sex, in addition to mean deformation.
The PCA encompasses around 82% of the total variance (Fig. 3). The first principal component (PC 1) contains 58.8% of the total variance, and PC2 explains 23.34% of the variance. Regarding CS (P = 0.12), no significant differences were found (P < 0.05). Discriminant analysis offered a correctly classified 84.31% value. The discriminant formula is given below, and its section point is 0.0122, where X, Y, UNIX, and UNIY are values of partial warps of the unknown individual whose sex we seek to identify. Xð7:6715Þ þ Yð21:248Þ þ UNIXð8:1411Þ þ UNIYð47:184Þ
Discussion Sex estimation within human identification processes is supported by the most dimorphic characteristics of certain osseous zones. On given occasions, however, morphological traits serve to support subjective determinations by the investigator when applying qualitative methods. This current research found that the squamous and nasal part of the frontal bone is a region of sexual determination when viewed using geometric morphometrics, similar to the research conducted by Shearer and coworkers (19), which analyzed the browridge as quantitative estimator through morphometric analysis. Similar findings, reported by Garvin and Ruff (20), show significant volume and surface differences for the browridge. Rosas and Bastir (2002) also reported that a difference was found in the shape of the nasoglabellar region, with a marked projection in males. These authors suggested that said difference may be due to an “increment of the frontal sinus and upward displacement of the rhinion.” Kimmerle et al. (9) and Gonzales et al. (10) reported significant size differences between males and females for the skull region. These two studies did not report size differences, although it may be because they included different bones of the neurocranium and the viscerocranium or possibly because of population differentials. The estimation percentage via discriminant analysis on the shape in the current research (84.31%) was higher than that
1332
JOURNAL OF FORENSIC SCIENCES
reported by Gonzales et al. (10) at 65% and lower than that reported by Rodrıguez (21) at 94.7% for males and 100% for females. However, the latter study used the traditional craniometry method with multiple measurements and requires excellent preservation of the skull, which is a difficult condition in the forensic context. According to that reported by Spradley and Jantz (2), sexual determination through multivariate analysis of the postcranial skeleton provides better estimation than the study on the skull. Nevertheless, these types of approaches become difficult when the skeleton is incomplete or fragmented, which supports the use of the findings from this research. In summary, the present results indicate the importance of using landmarks and multivariate analysis through the geometric morphometrics on morphoscopic traits and traditional craniometry. Besides, it reaffirms that using few landmarks, with correct anatomic location in the skull, permits detecting differences as predictors of sexual dimorphism even when only relatively small portions of the skeleton are available for study. Acknowledgments The author thanks the Research Group Applied Biomechanics. Gratitude is also expressed to the Faculty of Health Sciences at Universidad Libre for its valuable support and to the center for diagnostic imaging (DIAGRAMAX) for permitting the collection of the sample. References 1. Sheuer L. Application of osteology to forensic medicine. Clin Anat 2002;15:297–312. 2. Spradley MK, Jantz R. Sex estimation in forensic anthropology: skull versus postcranial elements. J Forensic Sci 2011;56(2):289–96. 3. Bethard J, Sheet B. Sex determination from the second cervical vertebra: a test of Wescott’s method on a modern American sample. J Forensic Sci 2013;58(1):101–3. 4. Walrath DE, Turner P, Bruzek J. Reliability test of the visual assessment of cranial traits for sex determination. Am J Phys Anthropol 2004;125:132–7. 5. Williams BA, Rogers T. Evaluating the accuracy and precision of cranial morphological traits for sex determination. J Forensic Sci 2006;51 (4):729–35. 6. Walker PL. Sexing skulls using discriminant function analysis of visually assessed traits. Am J Phys Anthropol 2008;136(1):39–50.
7. Gaper R, Black S, Last J. Sex determination from the foramen magnum: discriminant function analysis in an eighteenth and nineteenth century British sample. Int J Legal Med 2009;123(1):25–33. 8. Rosas A, Bastir M. Thin-plate spline analysis of allometry and sexual dimorphism in the human craniofacial complex. Am J Phys Anthropol 2002;117(3):236–45. 9. Kimmerle E, Ross A, Slice D. Sexual dimorphism in America: geometric morphometric analysis of the craniofacial region. J Forensic Sci 2008;53 (1):54–7. 10. Gonzales P, Bernal V, Perez S. Analysis of sexual dimorphism of craniofacial traits using geometric morphometric techniques. Int J Osteoarchaeol 2011;21:82–91. 11. Bookstein FL. Morphometric tools for landmark data: Geometry and biology. Cambridge, U.K.: Cambridge University Press, 1991. 12. Toro M, Manriquez G, Suazo I. Morfometrıa geometrica y el estudio de las formas biologicas: de la morfologıa descriptiva a la morfologıa cuantitativa. Int J Morphol 2010;28(4):977–90. 13. Potgieter W, Meiring J, Boon J, Petrorius E, Petrorius J, Becker P. Mandibular landmarks as an aid in minimizing injury to the marginal mandibular branch: a metric and geometric anatomical study. Clin Anat 2005;18(3):171–8. 14. Viðarsdottir U, O’Higgins P, Stringer C. A geometric morphometric study of regional differences in the ontogeny of the modern human facial skeleton J Anat 2002;201(3):211–29. 15. McIntyre G, Mossey P. Size and shape measurement in contemporary cephalometrics. Eur J Orthod 2003;25(3):231–42. 16. McDowell J, L’Abbe E, Kenyhercz M. Nasal aperture shape evaluation between black and white South Africans. Forensic Sci Int 2012; 222(1–3):397.e1–6. 17. Rohlf FJ. tpsDig; Version 1.4. Department of Ecology and Evolution, State University of New York at Stony Brook.tpsSplin, 2004; http://life. bio.sunysb.edu/morph/ (accessed December 10, 2012). 18. Rohlf FJ. tpsSpline; Version 1.20. Department of Ecology and Evolution, State University of New York at Stony Brook.tpsSplin. 2004; http://life. bio.sunysb.edu/morph/ (accessed December 12, 2012). 19. Shearer B, Sholts S, Garvin H, W€arml€ander S. Sexual dimorphism in human browridge volume measured from 3D models of dry crania: a new digital morphometrics approach. Forensic Sci Int 2012;222(1–3):400.e1–5. 20. Garvin H, Ruff C. Sexual dimorphism in skeletal browridge and chin morphologies determined using a new quantitative method. Am J Phys Anthropol 2012;147(4):661–70. 21. Rodrıguez JV. La diversidad poblacional de Colombia en el tiempo y espacio: Estudio craneometrico. Revista Academia colombiana de Ciencias 2007;31:321–46. Additional information and reprint requests: Nestor A. Perlaza, M.Sc. Department of Basic Sciences Faculty of Health, Universidad Libre Cali 760043 Colombia E-mail: [email protected]
TECHNICAL NOTE ANTHROPOLOGY Nestor A. Perlaza,1 M.Sc.
Sex Determination from the Frontal Bone: A Geometric Morphometric Study
ABSTRACT: Sex estimation in human skeletal remains when using the cranium through traditional methods is a fundamental pillar in human identification; however, it may be possible to incur in a margin of error due because of the state of preservation in incomplete or fragmented remains. The aim of this investigation was sex estimation through the geometric morphometric analysis of the frontal bone. The sample employed 60 lateral radiographs of adult subjects of both sexes (30 males and 30 females), aged between 18 and 40 years, with mean age for males of 28 4 and 30 6 years for females. Thin-plate splines evidenced strong expansion of the glabellar region in males and contraction in females. No significant differences were found between sexes with respect to size. The findings suggest differences in shape and size in the glabellar region, besides reaffirming the use of geometric morphometrics as a quantitative method in sex estimation.
KEYWORDS: forensic science, geometric morphometrics, sex estimation, forensic, frontal bone, thin-plate spline, sexual dimorphism
Sex determination in human skeletal remains is one of the initial bioanthropological characteristics investigated by forensic anthropologists, whose purpose is the construction of a biologic profile that permits identifying the individual (1). Most of the axial and appendicular skeleton has been addressed for sex determination, with varying levels of precision (2). These methods may have no practical value when skeletal remains are incomplete or fragmented. Thereby, analysis must be made on the bones available and is dependent upon their state of preservation (3). The skull is one area of the skeleton most often used in sex determination. Its analyses yield different degrees of precision through morphological traits (4,5), discriminant functions (6,7), and through geometric morphometrics (8–10). Landmark-based geometric morphometrics permit differentiation of anatomic traits in a quantitative manner; Morphometrics relate the shape and size of anatomic traits using the location of landmarks on the structures (11–13). In general, there has been a global approach to selecting landmarks of the skull, which encompasses anatomic references to bones of the neurocranium and viscerocranium (8–10,14–16). The aim of this research was sex estimation through geometric analysis of the frontal bone. Materials and Methods The sample was comprised of 60 lateral radiographs of adult subjects of both sexes (30 males and 30 females), ranging in age between 18 and 40 years, with mean ages for males of 28 4 and 30 6 years for females. The radiographs were obtained 1 Department of Basic Sciences, Faculty of Health, Universidad Libre, Cali 760043, Colombia. Received 27 Feb. 2013; and in revised form 5 July 2013; accepted 29 July 2013.
1330
from the diagnostic imaging center, DIAGRAMAX, specializing in oral and maxillofacial radiology in the city of Cali (Colombia). The parameters for taking the radiographs were as follows: magnification 1:1; 7 mA; 76 KV; total time of exposition: 8 sec. The images were stored in JPG format. The subjects were classified to avoid duplication, check for radiograph error, and safeguard confidentiality. Additionally, any of the subjects with a history of craniofacial trauma were excluded. All ethical and legal requirements were followed. The landmarks selected on the image were digitized in the squamous (landmarks 1 and 2) and nasal part (landmarks 3 and 4) of the frontal bone (Fig. 1). For this study, the thin-plate spline (TPS) series of programs was used (17,18). According to Bookstein (11), the landmarks are classified as follows: type I permits identifying changes on the structure, and these are tissue transition zones (e.g., intersection between cranial sutures); type II are points of great biomechanical tension (e.g., bony attachment); and type III are defined as curvature endpoints. The landmarks selected are described in Table 1. Geometric Morphometric Analysis A generalized Procrustes analysis (GPA) was developed, which offered a size and confirmation variables matrix upon minimizing the sum of squared distances between landmarks. Visualization in TPS was applied to graphically represent the relative deformations. Likewise, a multivariate analysis was performed through principal component analysis (PCA) to describe the shape. The prior analyses were carried out with the TPS series of programs (17,18). To analyze size differences, centroid size (CS) was used through the PAST software (http://folk.uio. no/ohammer/past); in addition, a discriminant analysis was employed to evidence the differences among the groups when maximizing the differences among themselves. © 2014 American Academy of Forensic Sciences
PERLAZA
.
A GEOMETRIC MORPHOMETRIC STUDY TO SEX DETERMINATION
FIG. 3––Scatter plot showing scores of first two PCs. Males are indicated by crosses and females by squares.
FIG. 1––Lateral radiograph with digitizing of the landmarks.
TABLE 1––List of landmarks. No 1.
Landmark
2.
Supratoral sulcos Glabella
3.
Nasion
4.
B. point
1331
Description
Type
Turning point where the protrusion of the glabellar zone starts The most anterior point of the squamous part midline The intersection between the frontonasal and the internasal suture The union between the frontal bone, nasal, and the maxilla
III III I I
Results Figure 2 shows the deformation in thin-plate splines, which details a contraction of the grid in the image corresponding to females and an expansion in the grid corresponding to males. Glabella and nasion are the landmarks where changes predominate in the magnitude of their directions for both sexes.
FIG. 2––Deformation of TPS showing changes in shape according to sex, in addition to mean deformation.
The PCA encompasses around 82% of the total variance (Fig. 3). The first principal component (PC 1) contains 58.8% of the total variance, and PC2 explains 23.34% of the variance. Regarding CS (P = 0.12), no significant differences were found (P < 0.05). Discriminant analysis offered a correctly classified 84.31% value. The discriminant formula is given below, and its section point is 0.0122, where X, Y, UNIX, and UNIY are values of partial warps of the unknown individual whose sex we seek to identify. Xð7:6715Þ þ Yð21:248Þ þ UNIXð8:1411Þ þ UNIYð47:184Þ
Discussion Sex estimation within human identification processes is supported by the most dimorphic characteristics of certain osseous zones. On given occasions, however, morphological traits serve to support subjective determinations by the investigator when applying qualitative methods. This current research found that the squamous and nasal part of the frontal bone is a region of sexual determination when viewed using geometric morphometrics, similar to the research conducted by Shearer and coworkers (19), which analyzed the browridge as quantitative estimator through morphometric analysis. Similar findings, reported by Garvin and Ruff (20), show significant volume and surface differences for the browridge. Rosas and Bastir (2002) also reported that a difference was found in the shape of the nasoglabellar region, with a marked projection in males. These authors suggested that said difference may be due to an “increment of the frontal sinus and upward displacement of the rhinion.” Kimmerle et al. (9) and Gonzales et al. (10) reported significant size differences between males and females for the skull region. These two studies did not report size differences, although it may be because they included different bones of the neurocranium and the viscerocranium or possibly because of population differentials. The estimation percentage via discriminant analysis on the shape in the current research (84.31%) was higher than that
1332
JOURNAL OF FORENSIC SCIENCES
reported by Gonzales et al. (10) at 65% and lower than that reported by Rodrıguez (21) at 94.7% for males and 100% for females. However, the latter study used the traditional craniometry method with multiple measurements and requires excellent preservation of the skull, which is a difficult condition in the forensic context. According to that reported by Spradley and Jantz (2), sexual determination through multivariate analysis of the postcranial skeleton provides better estimation than the study on the skull. Nevertheless, these types of approaches become difficult when the skeleton is incomplete or fragmented, which supports the use of the findings from this research. In summary, the present results indicate the importance of using landmarks and multivariate analysis through the geometric morphometrics on morphoscopic traits and traditional craniometry. Besides, it reaffirms that using few landmarks, with correct anatomic location in the skull, permits detecting differences as predictors of sexual dimorphism even when only relatively small portions of the skeleton are available for study. Acknowledgments The author thanks the Research Group Applied Biomechanics. Gratitude is also expressed to the Faculty of Health Sciences at Universidad Libre for its valuable support and to the center for diagnostic imaging (DIAGRAMAX) for permitting the collection of the sample. References 1. Sheuer L. Application of osteology to forensic medicine. Clin Anat 2002;15:297–312. 2. Spradley MK, Jantz R. Sex estimation in forensic anthropology: skull versus postcranial elements. J Forensic Sci 2011;56(2):289–96. 3. Bethard J, Sheet B. Sex determination from the second cervical vertebra: a test of Wescott’s method on a modern American sample. J Forensic Sci 2013;58(1):101–3. 4. Walrath DE, Turner P, Bruzek J. Reliability test of the visual assessment of cranial traits for sex determination. Am J Phys Anthropol 2004;125:132–7. 5. Williams BA, Rogers T. Evaluating the accuracy and precision of cranial morphological traits for sex determination. J Forensic Sci 2006;51 (4):729–35. 6. Walker PL. Sexing skulls using discriminant function analysis of visually assessed traits. Am J Phys Anthropol 2008;136(1):39–50.
7. Gaper R, Black S, Last J. Sex determination from the foramen magnum: discriminant function analysis in an eighteenth and nineteenth century British sample. Int J Legal Med 2009;123(1):25–33. 8. Rosas A, Bastir M. Thin-plate spline analysis of allometry and sexual dimorphism in the human craniofacial complex. Am J Phys Anthropol 2002;117(3):236–45. 9. Kimmerle E, Ross A, Slice D. Sexual dimorphism in America: geometric morphometric analysis of the craniofacial region. J Forensic Sci 2008;53 (1):54–7. 10. Gonzales P, Bernal V, Perez S. Analysis of sexual dimorphism of craniofacial traits using geometric morphometric techniques. Int J Osteoarchaeol 2011;21:82–91. 11. Bookstein FL. Morphometric tools for landmark data: Geometry and biology. Cambridge, U.K.: Cambridge University Press, 1991. 12. Toro M, Manriquez G, Suazo I. Morfometrıa geometrica y el estudio de las formas biologicas: de la morfologıa descriptiva a la morfologıa cuantitativa. Int J Morphol 2010;28(4):977–90. 13. Potgieter W, Meiring J, Boon J, Petrorius E, Petrorius J, Becker P. Mandibular landmarks as an aid in minimizing injury to the marginal mandibular branch: a metric and geometric anatomical study. Clin Anat 2005;18(3):171–8. 14. Viðarsdottir U, O’Higgins P, Stringer C. A geometric morphometric study of regional differences in the ontogeny of the modern human facial skeleton J Anat 2002;201(3):211–29. 15. McIntyre G, Mossey P. Size and shape measurement in contemporary cephalometrics. Eur J Orthod 2003;25(3):231–42. 16. McDowell J, L’Abbe E, Kenyhercz M. Nasal aperture shape evaluation between black and white South Africans. Forensic Sci Int 2012; 222(1–3):397.e1–6. 17. Rohlf FJ. tpsDig; Version 1.4. Department of Ecology and Evolution, State University of New York at Stony Brook.tpsSplin, 2004; http://life. bio.sunysb.edu/morph/ (accessed December 10, 2012). 18. Rohlf FJ. tpsSpline; Version 1.20. Department of Ecology and Evolution, State University of New York at Stony Brook.tpsSplin. 2004; http://life. bio.sunysb.edu/morph/ (accessed December 12, 2012). 19. Shearer B, Sholts S, Garvin H, W€arml€ander S. Sexual dimorphism in human browridge volume measured from 3D models of dry crania: a new digital morphometrics approach. Forensic Sci Int 2012;222(1–3):400.e1–5. 20. Garvin H, Ruff C. Sexual dimorphism in skeletal browridge and chin morphologies determined using a new quantitative method. Am J Phys Anthropol 2012;147(4):661–70. 21. Rodrıguez JV. La diversidad poblacional de Colombia en el tiempo y espacio: Estudio craneometrico. Revista Academia colombiana de Ciencias 2007;31:321–46. Additional information and reprint requests: Nestor A. Perlaza, M.Sc. Department of Basic Sciences Faculty of Health, Universidad Libre Cali 760043 Colombia E-mail: [email protected]
