Automatic identification of individuals using deep learning method on panoramic radiographs
Abstract Background/purpose: The dentition shows individual characteristics and dental structures are stable with respect to postmortem decomposition, allowing the dentition to be used as an effective tool in forensic dentistry. We developed an automatic identification system using panoramic radiogr...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-04-01
|
| Series: | Journal of Dental Sciences |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1991790222002732 |
| _version_ | 1797862189978091520 |
|---|---|
| author | Akifumi Enomoto Atsushi-Doksa Lee Miho Sukedai Takeshi Shimoide Ryuichi Katada Kana Sugimoto Hiroshi Matsumoto |
| author_facet | Akifumi Enomoto Atsushi-Doksa Lee Miho Sukedai Takeshi Shimoide Ryuichi Katada Kana Sugimoto Hiroshi Matsumoto |
| author_sort | Akifumi Enomoto |
| collection | DOAJ |
| description | Abstract Background/purpose: The dentition shows individual characteristics and dental structures are stable with respect to postmortem decomposition, allowing the dentition to be used as an effective tool in forensic dentistry. We developed an automatic identification system using panoramic radiographs (PRs) with a deep learning method. Materials and methods: In total, 4966 PRs from 1663 individuals with various changes in image characteristics due to various dental treatments were collected. In total, 3303 images were included in the data set used for model training. Vgg16, Vgg19, ResNet50, ResNet101, and EfficientNet models were applied for identification. The precision curves were evaluated. Results: The matching precision rates of all models (Vgg16, Vgg19, ResNet50, ResNet101, and EfficientNet) were examined. Vgg16 was the best model, with a precision of around 80–90% on 200 epochs, using the Top-N metrics concept with 5–15 candidate labels. The model can successfully identify the individual even with low quantities of dental features in 5–10 s. Conclusion: This identification system with PRs using a deep learning method appears useful. This identification system could prove useful not only for unidentified bodies, but also for unidentified wandering elderly people. This project will be beneficial for police departments and government offices and support disaster responses. |
| first_indexed | 2024-04-09T22:15:13Z |
| format | Article |
| id | doaj.art-8d0e57c3da3e4073836351c848bfdcf3 |
| institution | Directory Open Access Journal |
| issn | 1991-7902 |
| language | English |
| last_indexed | 2024-04-09T22:15:13Z |
| publishDate | 2023-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Dental Sciences |
| spelling | doaj.art-8d0e57c3da3e4073836351c848bfdcf32023-03-23T04:35:09ZengElsevierJournal of Dental Sciences1991-79022023-04-01182696701Automatic identification of individuals using deep learning method on panoramic radiographsAkifumi Enomoto0Atsushi-Doksa Lee1Miho Sukedai2Takeshi Shimoide3Ryuichi Katada4Kana Sugimoto5Hiroshi Matsumoto6Department of Oral and Maxillofacial Surgery, Kindai University, Faculty of Medicine, Osaka-Sayama, Osaka, Japan; Corresponding author.Department of Oral and Maxillofacial Surgery, Kindai University, Faculty of Medicine, Osaka-Sayama, Osaka, JapanDepartment of Oral and Maxillofacial Surgery, Kindai University, Faculty of Medicine, Osaka-Sayama, Osaka, JapanDepartment of Oral and Maxillofacial Surgery, Kindai University, Faculty of Medicine, Osaka-Sayama, Osaka, JapanDepartment of Legal Medicine, Osaka University Graduate School of Medicine, Suita, Osaka, JapanDepartment of Legal Medicine, Osaka University Graduate School of Medicine, Suita, Osaka, JapanDepartment of Legal Medicine, Osaka University Graduate School of Medicine, Suita, Osaka, JapanAbstract Background/purpose: The dentition shows individual characteristics and dental structures are stable with respect to postmortem decomposition, allowing the dentition to be used as an effective tool in forensic dentistry. We developed an automatic identification system using panoramic radiographs (PRs) with a deep learning method. Materials and methods: In total, 4966 PRs from 1663 individuals with various changes in image characteristics due to various dental treatments were collected. In total, 3303 images were included in the data set used for model training. Vgg16, Vgg19, ResNet50, ResNet101, and EfficientNet models were applied for identification. The precision curves were evaluated. Results: The matching precision rates of all models (Vgg16, Vgg19, ResNet50, ResNet101, and EfficientNet) were examined. Vgg16 was the best model, with a precision of around 80–90% on 200 epochs, using the Top-N metrics concept with 5–15 candidate labels. The model can successfully identify the individual even with low quantities of dental features in 5–10 s. Conclusion: This identification system with PRs using a deep learning method appears useful. This identification system could prove useful not only for unidentified bodies, but also for unidentified wandering elderly people. This project will be beneficial for police departments and government offices and support disaster responses.http://www.sciencedirect.com/science/article/pii/S1991790222002732Automatic identificationDeep learningPanoramic radiograph |
| spellingShingle | Akifumi Enomoto Atsushi-Doksa Lee Miho Sukedai Takeshi Shimoide Ryuichi Katada Kana Sugimoto Hiroshi Matsumoto Automatic identification of individuals using deep learning method on panoramic radiographs Journal of Dental Sciences Automatic identification Deep learning Panoramic radiograph |
| title | Automatic identification of individuals using deep learning method on panoramic radiographs |
| title_full | Automatic identification of individuals using deep learning method on panoramic radiographs |
| title_fullStr | Automatic identification of individuals using deep learning method on panoramic radiographs |
| title_full_unstemmed | Automatic identification of individuals using deep learning method on panoramic radiographs |
| title_short | Automatic identification of individuals using deep learning method on panoramic radiographs |
| title_sort | automatic identification of individuals using deep learning method on panoramic radiographs |
| topic | Automatic identification Deep learning Panoramic radiograph |
| url | http://www.sciencedirect.com/science/article/pii/S1991790222002732 |
| work_keys_str_mv | AT akifumienomoto automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs AT atsushidoksalee automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs AT mihosukedai automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs AT takeshishimoide automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs AT ryuichikatada automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs AT kanasugimoto automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs AT hiroshimatsumoto automaticidentificationofindividualsusingdeeplearningmethodonpanoramicradiographs |