Wang Xindi, Wei Shanshan, Zhao Zibo, Luo Xiang, Song Feng, Li Yuan
Department of Forensic Genetics, West China School of Basic Medical Sciences & Forensic Medicine, Sichuan University, Chengdu, Sichuan 610041, PR China.
Department of Artificial Intelligence, College of Blockchain Industry, Chengdu University of Information Technology, Chengdu, Sichuan 610225, PR China.
Forensic Sci Int. 2024 Dec;365:112271. doi: 10.1016/j.forsciint.2024.112271. Epub 2024 Oct 23.
The 3D-3D superimposition technique offers a comprehensive comparison and quantification of the similarity between two 3D models, making it a promising approach for forensic science. This systematic review aims to explore scanning techniques, 3D analysis software, and 3D-3D superimposition methods, along with providing an overview of their applications and performance in personal identification. We searched four databases (PubMed, Scopus, Web of Science, and Embase) for articles published between January 2014 and May 2024. QUADAS-2 was adopted to assess the risk of bias in the included studies. After applying the eligibility criteria, a total of 25 articles were selected for our review. From these articles, we identified six scanning techniques and eight software programs. These diverse scanning technologies and 3D software significantly streamline the process of 3D data registration and similarity calculations. The surface-based registration was the most frequently utilized approach and was considered more suitable for forensic research. Studies have shown that 3D-3D superimposition has emerged as a valuable tool in various fields, including the antemortem (AM) and postmortem (PM) comparison, pair-matching, re-association, craniofacial superimposition, and face identification. It is anticipated that further research involving a broader range of sample types and standardized protocols will further enhance the applicability of 3D-3D superimposition technology in forensic practice.
3D-3D叠加技术可对两个3D模型之间的相似性进行全面比较和量化，使其成为法医学中一种很有前景的方法。本系统综述旨在探索扫描技术、3D分析软件和3D-3D叠加方法，并概述它们在个人识别中的应用和性能。我们在四个数据库（PubMed、Scopus、Web of Science和Embase）中搜索了2014年1月至2024年5月发表的文章。采用QUADAS-2评估纳入研究的偏倚风险。应用纳入标准后，共选择了25篇文章进行综述。从这些文章中，我们确定了六种扫描技术和八个软件程序。这些多样的扫描技术和3D软件显著简化了3D数据配准和相似度计算的过程。基于表面的配准是最常用的方法，被认为更适合法医研究。研究表明，3D-3D叠加已成为各个领域的重要工具，包括生前（AM）和死后（PM）比较、配对匹配、重新关联、颅面叠加和面部识别。预计进一步开展涉及更广泛样本类型和标准化方案的研究将进一步提高3D-3D叠加技术在法医实践中的适用性。
