Biological Gender Estimation from Panoramic Dental X-ray Images Based on Multiple Feature Fusion Model
- PDF / 1,116,378 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 55 Downloads / 210 Views
Biological Gender Estimation from Panoramic Dental X‑ray Images Based on Multiple Feature Fusion Model Wenchi Ke1 · Fei Fan1 · Peixi Liao1 · Yancun Lai1 · Qingsong Wu1 · Wenchao Du1 · Hu Chen1 · Zhenhua Deng1 · Yi Zhang1 Received: 13 June 2020 / Revised: 20 September 2020 / Accepted: 12 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In a catastrophe, the conventional biological characteristics of the victims will be destroyed. Forensic odontology is the main method to identify the victims. Estimating the gender of the victims has a significant meaning and can greatly help identify the victims. In this paper, we propose a new automatic method to the gender estimation from panoramic dental X-ray images based on improved convolutional neural network with multiple feature fusion module. Our dataset includes 19,976 panoramic dental X-ray images from Chinese patients. The method we propose can estimate 142 images per second on the conventional computing equipment and it achieves state-of-the-art performance, accuracy of 94.6% ± 0.58%, in our dataset. Our model is interpreted by perturbation-based forward propagation approaches, and the results show that focus of our method on the area of mandible and teeth is reliable which is in accordance with forensic practice. Keywords Gender estimation · Panoramic dental X-ray · Convolutional neural network · Feature fusion
1 Introduction Identification of human remains is the most important step for further investigation in forensic cases [1]. However, in some cases (such as disaster, suicide, murder, etc.), the conventional biological characteristics of the victims are often unavailable and the method of identity used under normal circumstances will be fail. The 2014 INTERPOL Disaster Victim Identification Guide states that“the primary and most reliable means of identification are friction ridge analysis, forensic odontology, and DNA analysis” [2]. Among them, friction ridge analysis and DNA analysis may be fail due to the unavailable biological characteristics of victims. Since teeth are the * Hu Chen [email protected] 1
Sichuan University, Chengdu, China
13
Vol.:(0123456789)
54
Page 2 of 11
Sensing and Imaging
(2020) 21:54
hardest parts of the human body and extremely resistant to decomposition [3], and the mandible is the largest and strongest bone in the face, forensic odontology is the key method to determine their identification. One of the most important elements for the identification is the gender and determining the gender of the victims firstly is of great help to determine the identity. Generally, Sex can be determined up to 100% accuracy with the entire adult bone analysis [1]. However, when the skull is missing, the accuracy will decline very much and the mandible plays a very important role [4–6]. Since panoramic dental X-ray (PDR) images contains the areas of teeth and mandible as shown in Fig. 1, they are suitable for gender inference. Most methods for estimating the gender of a person based on skelet
Data Loading...
