Reprodutibilidade de evidências forenses em mandíbulas: comparação entre reconstrução tridimensional e impressão 3D

Abstract

Introduction: 3D printing can be used for many applications, including document imaging, human identification, comparative dentistry, bite mark and pattern analysis, ballistics, facial reconstruction, crime scene reconstruction, as well as disaster victim identification (DVI), forensic anthropology, and archaeology. The present study was designed to evaluate the reproducibility of artificial mandibles containing lesions of forensic interest through three dimensional reconstruction obtained by cone beam computed tomography (CBCT) and 3D printing. Methods: In this study, a quantitative and descriptive approach was used, dispensing ethical assessment for the use of 14 artificial jaws in resin (polyurethane). Five stages were performed, from the preparation of nine simulated lesions of forensic interest (bullet entrance hole, bullet exit hole, multiple bullet holes, sharp force trauma, tooth loss, coronary tooth fractures, mandibular fracture and complete and incomplete condyle fractures) to the tomographic scans of the 14 jaws using CBCT with voxel sizes of 0. 25, 0.3 and 0.4mm, followed by the analysis of 10 landmarks in the 3D volumes generated (Article 1) and subsequent three-dimensional printing using the FDM technique of volumes generated with voxel 0.3, to compare the two reproduction techniques used (tomographic = AT and printed = AC). The data were analyzed using T-Student and ICC tests and presented in Bland-Altman plots (article 2). Results: In article 1, the results showed that the protocols with voxels of 0.3 mm should be preferred for the evaluation of linear and angular measurements. Statistical analysis showed that there was no statistically significant difference in the intra-examiner A (p=0.920) and B (p=0.424) and inter-examiner (p=0.664) scores. This demonstrated the reliability of the used method. The study showed that it is possible to create reference points in 3D models, which is an important step in consolidating the use of prototypes in forensic studies. In article 2, the statistical analysis with the ICC test showed that most of the variables (simulated lesions) were well reproduced (ICC above 0.8) in both techniques. It was also observed that the pairs AT and AC had measurements close to the mean difference and remained within the confidence interval, although the printing technique showed less variation in the measurements performed compared to the original ones (AC). Discussion: According to the results of the two articles, there was success in the use of 3D technologies (CBCT and printing) to evaluate the reliability and reproducibility of these techniques in forensic contexts. Article 1 showed, in its results, that the error found in the inter-examiner comparison (using the same measurement method) was small and remained stable, regardless of the evaluated variables. Regarding the comparison of three-dimensional reproducibility by tomographic (CBCT) and printed methods, article 2 showed that both techniques are reliable. The result of the reproducibility of the simulated lesions selected for article 2 showed that all lesions were reproduced. The shape, number, and anatomical relationships were preserved. The simulated lesions caused by firearm projectile (bullet entrance and bullet exit hole), sharp force trauma and tooth loss were rated from good to excellent in tomographic and 3D printed reproductions. Articles 1 and 2 presented as a result of this thesis are complementary in the sense that the first has evaluated three dimensionals volumes digitally reproduced (well established craniometric points were observed), and the second allows a comparative evaluation between three-dimensional digital and printed volumes (analyzing the reproduction of specific forensic lesions). Conclusion: The implementation of this methodology of prototypes as conditions of clinical and forensic simulation allows the comparison of the human database in identification issues. The analog measurement technique applied in the 3D printed volume proved to be accurate and reproducible when compared with the computerized technique using 3D digital images. New studies are needed in the search for standardization of three-dimensional measurements in digitized and printed volumes.