Automated Identification from Dental Data (AutoIDD): A new development in digital forensics

Abstract

There has been a significant expansion in the use of 3-dimensional (3D) dental images in recent years. In the field of forensic odontology, an automated 3D dental identification system could enhance the identification process. This study presents a novel method for automated human dental identification using 3D digital dental data by utilising a dental identification scenario. The total study sample was divided into two groups: Group A (120 dental models) and Group B (120 Intra-oral scans-IOS). Group A data was composed of 3D scanned dental models of post-orthodontic treated patients (30 maxillary and 30 mandibular). This data was considered as AM digital data. To generate an identical sample, the dental casts (60) of the same patients were retrieved and laser scanned. These models were considered as PM digital data. Group B data (IOS) was obtained from 30 study participants. To reconstruct a dental identification scenario 30 maxillary and 30 mandibular IOS were obtained from 30 participants and were considered as IOS-AM. After one year, another set of IOS (60) were acquired from the same participants and were considered as IOS-PM. The results showed that the AutoIDD (Automated Identification from Dental Data) software was consistent in accuracy; capable of differentiating "correct matches" (high match percentage) from "non-matches" (very low percentage) by 3D image superimposition. The match percentage of the maxillary and mandibular IOS ranged from 64-100% and 81-100%, with a mean percentage of 96.7 and 96.4 respectively. This study demonstrated the feasibility of using 3D scans through a new automated software-AutoIDD in digital forensics to assist the forensic expert in confirming the identity of a deceased individual from the available AM dental records.

Exploring the use of selfies in human identification

“…Difficulties that may interfere in uniqueness, such as orthodontic treatment, are discussed in these studies as well. 6,10,24 Several participants mentioned the assessment of anatomical landmarks on the skull such as the nasion, glabella and gnathion in acquiring a possible correspondent, and comparing them on AM photographs. This may serve as a useful tool in establishing the identity of the individual as illustrated in studies of anthropometrical landmarks.…”

“…[3][4][5] Specialised two-(2-D) and threedimensional (3-D) software encompassing skull-photo superimposition methods are utilised for this approach, with recently developed techniques such as GrinLine Identification technique (GLID) and Automated Identification from Dental Data (AutoIDD). [6][7][8][9][10] In the mortuary, Forensic Odontologists are recommended to document their cases by photographing and labelling the decedent's dentition from the frontal, left and right lateral profile, maxillary and mandibular occlusal views, together with the palatal/lingual views of the teeth to observe possible dental treatment and other distinctive traits on those surfaces. 1,11 Considering AM data, smile photographs and selfies of the individual taken from social media (SM), family keepsakes and video imaging may be used to assist with the identification of an individual.…”

O uso de fotografias do sorriso como um método de identificação humana: relatos em série

“…Reesu et al (2020) 37 demonstraram a viabilidade da sobreposição de imagens por meio da comparação visual de fotografia AM e PM do modelo de gesso (fase I) com a sobreposição de imagem utilizada no modelo de gesso 3D (fase II), e concluíram que a sobreposição de imagem 3D reduz as limitações dos métodos 2D. As fotografias são projeções 2D de objetos 3D, desta forma, várias tomadas fotográficas na ossada são necessárias para aumentar as chances de se reproduzir a perspectiva encontrada na fotografia AM.…”

3D superimposition of human dentition contours in personal identification: A preliminary study

“…Reesu [25] proposed a novel technique for automatic human teeth identification using 3D scans and 3D superimposition processed by ICP algorithms.…”