Short Laura J, Khambay Balvinder, Ayoub Ashraf, Erolin Caroline, Rynn Chris, Wilkinson Caroline
Orthodontics, Medical School, MVLS College, University of Glasgow, United Kingdom.
Paediatric Dentistry and Orthodontics, Faculty of Dentistry, University of Hong Kong, Hong Kong.
Forensic Sci Int. 2014 Apr;237:147.e1-147.e8. doi: 10.1016/j.forsciint.2013.12.042. Epub 2014 Jan 10.
Human forensic facial soft tissue reconstructions are used when post-mortem deterioration makes identification difficult by usual means. The aim is to trigger recognition of the in vivo countenance of the individual by a friend or family member. A further use is in the field of archaeology. There are a number of different methods that can be applied to complete the facial reconstruction, ranging from two dimensional drawings, three dimensional clay models and now, with the advances of three dimensional technology, three dimensional computerised modelling. Studies carried out to assess the accuracy of facial reconstructions have produced variable results over the years. Advances in three dimensional imaging techniques in the field of oral and maxillofacial surgery, particularly cone beam computed tomography (CBCT), now provides an opportunity to utilise the data of live subjects and assess the accuracy of the three dimensional computerised facial reconstruction technique.
The aim of this study was to assess the accuracy of a computer modelled facial reconstruction technique using CBCT data from live subjects.
This retrospective pilot study was carried out at the Glasgow Dental Hospital Orthodontic Department and the Centre of Anatomy and Human Identification, Dundee University School of Life Sciences. Ten patients (5 male and 5 female; mean age 23 years) with mild skeletal discrepancies with pre-surgical cone beam CT data (CBCT) were included in this study. The actual and forensic reconstruction soft tissues were analysed using 3D software to look at differences between landmarks, linear and angular measurements and surface meshes.
There were no statistical differences for 18 out of the 23 linear and 7 out of 8 angular measurements between the reconstruction and the target (p<0.05). The use of Procrustes superimposition has highlighted potential problems with soft tissue depth and anatomical landmarks' position. Surface mesh analysis showed that this virtual sculpture technique can be objectively assessed using the distance between the meshes. This study found that the percentage of faces with less than ±2.5mm error ranged from 56% to 90%. This may be improved if Procrustes superimposition could be applied to all the mesh points rather than specific landmarks.
当尸体面部软组织出现死后退变，导致难以通过常规方法进行身份识别时，就会采用人类法医面部软组织重建技术。其目的是促使朋友或家庭成员认出个体生前的面容。该技术在考古领域也有进一步应用。有多种不同方法可用于完成面部重建，从二维绘图、三维黏土模型，到如今随着三维技术的进步，出现了三维计算机建模。多年来，为评估面部重建准确性所开展的研究结果各不相同。口腔颌面外科领域三维成像技术的进步，尤其是锥形束计算机断层扫描（CBCT），现在为利用活体受试者数据并评估三维计算机面部重建技术的准确性提供了契机。
本研究旨在利用活体受试者的CBCT数据评估计算机建模面部重建技术的准确性。
这项回顾性试点研究在格拉斯哥牙科医院正畸科以及邓迪大学生命科学学院解剖学与人类身份识别中心开展。本研究纳入了10名患者（5名男性和5名女性；平均年龄23岁），他们存在轻度骨骼差异且术前有锥形束CT数据（CBCT）。使用三维软件对实际和法医重建软组织进行分析，以观察标志点、线性和角度测量以及表面网格之间的差异。
重建与目标之间的23项线性测量中的18项以及8项角度测量中的7项，不存在统计学差异（p<0.05）。普洛透斯叠加法的使用凸显了软组织深度和解剖标志点位置方面的潜在问题。表面网格分析表明，这种虚拟雕塑技术可通过网格之间的距离进行客观评估。本研究发现，误差小于±2.5毫米的面部比例在56%至90%之间。如果能将普洛透斯叠加法应用于所有网格点而非特定标志点，这一比例可能会得到改善。
