Forensic anthropology population data Analyses of sexual dimorphism of reconstructed pelvic computed tomography images of contemporary Japanese using curvature of the greater sciatic notch, pubic arch and greater pelvis Hitoshi Biwasaka a,b , Yasuhiro Aoki c, *, Kei Sato a , Toyohisa Tanijiri d , Sachiko Fujita b , Koji Dewa b , Kunihiro Yoshioka e , Makiko Tomabechi e a Forensic Science Laboratory, Iwate Prefectural Police Headquarters, Morioka 020-8540, Japan b Department of Legal Medicine, Iwate Medical University School of Medicine, Morioka 020-8505, Japan c Department of Forensic Medicine, Nagoya City University Graduate School of Medical Sciences, Nagoya 467-8601, Japan d Medic Engineering Co., Kyoto 606-8107, Japan e Department of Radiology, Iwate Medical University School of Medicine, Morioka 020-8505, Japan 1. Introduction New imaging techniques have been recently introduced in various forensic fields [1–5]. In particular, 3-dimensional (3D) images reconstructed from volume data of multi-slice computed tomography (CT) boast high reproducibility, and several attempts utilizing CT reconstruction images of bones have been reported in the field of forensic and physical anthropology [6–11]. We also have confirmed the reproducibility of 3D-CT images of a skeletonized pelvis and their applicability to anthropological measurements [12]. 3D-CT images of the pelvis reproduce well complex curved features, and the stored data format facilitates computerized geometrical analyses. Moreover, 3D measurements using these data could potentially elucidate both subtle and intuitional differences of morphological features which have not been demonstrated clearly by conventional 2-dimensional (2D) methods. However, application of this to analysis of the pelvis has not been extensively investigated and standardized. Regarding sexing of the pelvis, various anatomical features and discriminant functions have been proposed, especially on the pubis and greater sciatic notch (GSN) [13–18]. Among them, sexing from the subpubic arch has been performed mainly by non- metric methods because curved and twisted features create problems with the simple setting of angle measurements [13,14,16]. On the other hand, linear metric methods such as the ratio between the length and depth of the GSN are not always available as they require anatomical landmarks which are often missing in forensic and archeological samples. To overcome these difficulties, some methods using geometric morphometric tech- niques based on variance analysis have recently been reported [19–21]. These methods would be independent from the experience of the observer, although some of the proposed methods still require a comparatively large number of landmarks and semilandmarks [19]. Forensic Science International 219 (2012) 288.e1–288.e8 A R T I C L E I N F O Article history: Received 29 October 2010 Received in revised form 28 June 2011 Accepted 30 November 2011 Available online 23 December 2011 Keywords: Forensic anthropology Multi-slice CT Pelvis Curvature Sex determination Image analysis Forensic anthropology population data A B S T R A C T Three-dimensional pelvic images were reconstructed from multi-slice CT data of contemporary Japanese (males: 124; females: 104, 25–92 years old), and curvature analysis to examine sexual dimorphism was carried out in the great sciatic notch (GSN), the pubic arch and the greater pelvis in the images. Reconstructed pelvic CT images were visualized fairly well and anatomical landmarks were easily recognizable. When calculating the radii (curvature radii) of the best-fit circles for the spline curve lines set along the edges of the GSNs and of the pubic arches, sexes from these regions were correctly identified in 89.1% (males: 93.8%; females: 83.7%) and 94.7% (males: 97.3%; females: 91.8%) of cases, respectively, by setting an appropriate cut-off value. Furthermore, sexing was possible even in deeper regions of the GSN which are relatively resistant to postmortem damage. Curvature radii of the best-fit spheres of greater pelves showed no significant difference between sexes. However, curvature of the best-fit sphere for the left iliac fossa was significantly larger than that of the right one (p < 10 À24 ) in males, and the ratios were >1.0 in 88% of all male specimens analyzed. Meanwhile, no significant difference was observed among female samples. Although some left-sided dominancy has been reported in 2-dimensional measurements of the human pelvis, this 3-dimensional laterality in males was much more significant, and is a potential index of sex difference. ß 2011 Elsevier Ireland Ltd. All rights reserved. * Corresponding author. Tel.: +81 52 853 8179; fax: +81 52 842 7701. E-mail address: aokiy@med.nagoya-cu.ac.jp (Y. Aoki). Contents lists available at SciVerse ScienceDirect Forensic Science International jou r nal h o mep age: w ww.els evier .co m/lo c ate/fo r sc iin t 0379-0738/$ – see front matter ß 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2011.11.032