Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 5, 2026

Image Rendering Techniques in Postmortem Computed Tomography: Evaluation of Biological Health and Profile in Stranded Cetaceans

Published on: September 27, 2020

Deep Learning-Based Panoramic Radiograph Retrieval from Antemortem Images for Forensic Identification.

Abdülkadir İzci¹, Uğur Kayhan², Adem Pekince³

¹Faculty of Medicine, Department of Forensic Medicine, Afyonkarahisar Health Sciences University, Afyonkarahisar, Turkey. drkadirizci@gmail.com.

International Journal of Legal Medicine

|July 3, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Medical malpractice in pediatric surgery: an analysis of Supreme Court decisions involving physicians.

Annals of Saudi medicine·2026

Same author

Attitudes of emergency healthcare workers on electric vehicle accidents: a descriptive study.

BMC emergency medicine·2025

Same author

The relationship between hypostasis and gas detected in postmortem computed tomography and the postmortem interval.

Diagnostic and interventional radiology (Ankara, Turkey)·2025

Same author

Relationships between meaning and purpose in life, life satisfaction, and occupational burnout among forensic specialists and autopsy technicians, and the impact of autopsy practice.

Journal of forensic sciences·2025

Same author

The Impact of the Covid-19 Pandemic on Child Sexual Abuse: A Comparative Analysis.

Journal of interpersonal violence·2025

Same author

Improving TMJ Diagnosis: A Deep Learning Approach for Detecting Mandibular Condyle Bone Changes.

Diagnostics (Basel, Switzerland)·2025

This study shows a deep learning framework using panoramic dental X-rays is effective for forensic identification. The expert-assisted system achieved high accuracy in matching individuals from large datasets.

Area of Science:

Forensic Science
Artificial Intelligence
Radiology

Background:

Forensic identification relies on accurate methods for matching individuals.
Panoramic dental radiographs offer unique anatomical features for identification.
Automated systems can enhance the efficiency and scale of forensic analysis.

Purpose of the Study:

To assess the effectiveness of a deep learning framework for forensic identification using panoramic dental radiographs.
To evaluate an expert-assisted, semi-automated Convolutional Neural Network (CNN) approach.
To determine applicability in large-scale, retrieval-based identification scenarios.

Main Methods:

Developed a CNN-based deep learning framework using 2,440 panoramic radiographs from 842 individuals.

Keywords:

Artificial intelligence Deep learning Dental identification Forensic dentistry Forensic odontology

More Related Videos

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Related Experiment Videos

Last Updated: Jul 5, 2026

Image Rendering Techniques in Postmortem Computed Tomography: Evaluation of Biological Health and Profile in Stranded Cetaceans

Published on: September 27, 2020

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Preprocessed images via masking, cropping, and resizing.

Employed subject-level 5-fold cross-validation with four deep learning backbones, ensuring data separation between training and validation sets.

Main Results:

The ConvNeXt-Tiny model demonstrated the best performance among the evaluated backbones.
Achieved Top-1 accuracy of 75.00±6.51%, Top-3 accuracy of 82.22±3.16%, and Top-5 accuracy of 83.33±1.96%.
The framework maintained class balance and prevented data leakage between training and validation sets.

Conclusions:

The proposed expert-assisted, semi-automated deep learning framework is effective for forensic identification.
The method shows significant potential for application in large-scale forensic identification practices.
Convolutional Neural Network models, particularly ConvNeXt-Tiny, show promise in analyzing dental radiographs for identification.