Small Candice, Schepartz Lynne, Hemingway Jason, Brits Desiré
School of Anatomical Sciences, Faculty of Health Sciences, University of the Witwatersrand, Human Variation and Identification Research Unit (HVIRU), Johannesburg 2193, South Africa.
Forensic Sci Int. 2018 Jun;287:127-135. doi: 10.1016/j.forsciint.2018.02.012. Epub 2018 Feb 21.
The skull is the element most frequently presented to forensic anthropologists for analysis yet weathering, corpse maiming, and scavenger activity often result in damage and fragmentation. This fragmentation results in a reduction in the number of traditional calliper derived measurements that can be obtained and subjected to discriminant based analyses for sex estimation. In this investigation, we employed three-dimensional geometric morphometric methods to derive novel interlandmark distance measures across six regions of the cranium including the basicranium, basipalate, zygoma, orbits and the cranium globally to create functions to discriminate sex with high efficacy, even in the event of fragmentation. Forty-five homologous landmarks were digitised across each of 227 (114 males and 113 females) South African crania of European descent (white) sampled from the Raymond A Dart Collection of Human Skeletons, housed in the School of Anatomical Sciences, University of the Witwatersrand, South Africa. A total of 990 interlandmark distances (ILDs) were mathematically derived using Pythagorean geometry. These ILDs were then filtered by region and subjected to both direct and stepwise discriminant function analyses. Discriminant equations where derived for each region and achieved the following average cross-validated sex estimation accuracies: basicranium-74%; basipalate-80.2%; zygomatic-82.4; orbits-71.8%; nasomaxilla-83.7%; global cranium-88.2%. A large number of the ILDs used to derive the discriminant functions are novel, demonstrating the efficacy of geometric morphometric methods and illustrating the need to reassess old methods of data collection using modern methods to determine whether they best capture biological differences. The results of this study provide an invaluable contribution to forensic anthropology in South Africa as it provides an accurate, practical means of assessing sex using fragmentary material that may otherwise have been disregarded. These will undeniable aid in accurate sex estimation and ultimately, victim identification.
颅骨是法医人类学家最常拿到手进行分析的人体部位，但风化、尸体残害和食腐动物活动常常导致颅骨受损和破碎。这种破碎使得能够获取并用于基于判别分析进行性别估计的传统卡尺测量数量减少。在本研究中，我们采用三维几何形态测量方法，在颅骨的六个区域（包括颅底、腭基、颧骨、眼眶以及整个颅骨）得出新的地标间距离测量值，以创建即使在颅骨破碎情况下也能高效判别性别的函数。从南非金山大学解剖科学学院收藏的雷蒙德·A·达特人类骨骼收藏中，选取了227具（114例男性和113例女性）欧洲血统（白人）的南非颅骨，对每个颅骨的45个同源地标进行数字化处理。利用毕达哥拉斯几何原理在数学上推导出总共990个地标间距离（ILDs）。然后按区域对这些ILDs进行筛选，并进行直接判别函数分析和逐步判别函数分析。为每个区域推导判别方程，得到以下平均交叉验证性别估计准确率：颅底 - 74%；腭基 - 80.2%；颧骨 - 82.4%；眼眶 - 71.8%；鼻上颌骨 - 83.7%；整个颅骨 - 88.2%。用于推导判别函数的大量ILDs是新的，这证明了几何形态测量方法的有效性，并说明需要用现代方法重新评估旧的数据收集方法，以确定它们是否能最好地捕捉生物差异。本研究结果为南非法医人类学做出了宝贵贡献，因为它提供了一种准确、实用的方法，可利用原本可能被忽视的碎片材料评估性别。这些无疑将有助于准确的性别估计，并最终有助于受害者身份识别。
