Related Experiment Video
Updated: Jun 3, 2025
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
764
Systematic Review and Meta-Analysis on the Accuracy of Artificial Intelligence Algorithms in Individuals Gender
Mahmood Dashti1, Tara Azimi2, Farshad Khosraviani3
1Researcher, Dentofacial Deformities Research Center, Research Institute of Dental Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
International Dental Journal
|January 11, 2025
Summary
Artificial intelligence (AI) can determine gender from dental panoramic radiographs (OPGs) with 88.66% accuracy. While AI shows high specificity, further research is needed to improve sensitivity and consistency across diverse populations.
Area of Science:
- Dental Radiology
- Artificial Intelligence
- Forensic Odontology
Background:
- Artificial intelligence (AI) is advancing dental imaging analysis, particularly for panoramic radiographs (orthopantomograms or OPGs).
- AI offers a novel approach to gender determination from OPGs, traditionally a forensic task.
- Evaluating AI's accuracy in this application is crucial for its clinical and forensic utility.
Purpose of the Study:
- To systematically review and meta-analyze the accuracy of AI algorithms for gender determination using OPGs.
- To assess the reliability, sensitivity, and specificity of AI models in this task.
- To identify potential clinical and forensic applications of AI in OPG analysis.
Main Methods:
- Systematic review and meta-analysis adhering to PRISMA guidelines.
- Inclusion of studies utilizing AI for gender detection from OPG images, sourced from five major databases.
- Statistical analysis using R software, including forest and funnel plots, to assess diagnostic performance and publication bias.
Main Results:
- The meta-analysis included 13 studies, reporting a pooled accuracy of 88.66% for AI-based gender determination from OPGs.
- AI models demonstrated high specificity, with some reaching 99.20% accuracy.
- Variability in sensitivity was observed, suggesting differences in model performance based on datasets and features.
Conclusions:
- AI algorithms show significant potential for accurate gender determination from OPG images.
- Near-perfect accuracy has been achieved by some AI models.
- Further research is necessary to enhance model consistency and applicability across diverse demographic groups.
