Sex Prediction Using Machine Learning Algorithms and Artificial Neural Networks from the Temporomandibular Joint

Purpose: The aim of our study is to determine sex using parameters obtained from the temporomandibular joint (TMJ) region of the mandible, in conjunction with machine learning (ML) algorithms and artificial neural networks (ANN).Methods: This study was conducted using computed tomography (CT) images obtained from 200 individuals (100 men and 100 women) aged 18–65. Parameters related to the temporomandibular joint were measured from the acquired CT images. The values obtained from these measurements were used to determine sex using machine learning algorithms and artificial neural network methods. The models were evaluated using a 20% test data split and a 10-fold crossvalidation method. Permutation importance and SHAP analyses were used for model interpretation.Results: Among the ML models, logistic regression (LR) achieved the highest test accuracy (82.5%) on the independent 20% test set. The multilayer perceptron (MLPClassifier), an ANN-based model, achieved a test accuracy of 80.0%. In the 10-fold cross-validation analysis, the support vector machine (SVM) achieved the highest average accuracy among the ML models (0.8438±0.1224), while the MLPClassifier reached an average accuracy of 0.6438±0.1384. The SHAP analysis of the LR model identified the left Fpost–AEtop variable, with an average absolute SHAP value of 0.2108, as the most influential predictor.Conclusions: Three-dimensional TMJ morphometry provides important and highly accurate information for sex classification. The Fpost–AEtop left measurement was consistently identified as the most informative predictor based on model interpretability analyses. This finding further supports the importance of regional morphology in forensic and anatomical classification studies.

Authors

Institutions

Publication Details

Journal
Acibadem Universitesi Saglik Bilimleri Dergisi
Published
2026-11-19
DOI
https://doi.org/10.31067/acusaglik.1906771
Primary Topic
Forensic Anthropology and Bioarchaeology Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Sex Prediction Using Machine Learning Algorithms and Artificial Neural Networks from the Temporomandibular Joint

Şeyma Toy, Nesibe Yılmaz, Yusuf Seçgin, Halil Şaban Erkartal et al.
Acibadem Universitesi Saglik Bilimleri Dergisi
Forensic Anthropology and Bioarchaeology Studies
article

Sex Prediction Using Machine Learning Algorithms and Artificial Neural Networks from the Temporomandibular Joint

Şeyma Toy, Nesibe Yılmaz, Yusuf Seçgin, Halil Şaban Erkartal, Nevin Köremezli Keskin, Yasemin KÜÇÜKTEPE
article en

Abstract

Purpose: The aim of our study is to determine sex using parameters obtained from the temporomandibular joint (TMJ) region of the mandible, in conjunction with machine learning (ML) algorithms and artificial neural networks (ANN).Methods: This study was conducted using computed tomography (CT) images obtained from 200 individuals (100 men and 100 women) aged 18–65. Parameters related to the temporomandibular joint were measured from the acquired CT images. The values obtained from these measurements were used to determine sex using machine learning algorithms and artificial neural network methods. The models were evaluated using a 20% test data split and a 10-fold crossvalidation method. Permutation importance and SHAP analyses were used for model interpretation.Results: Among the ML models, logistic regression (LR) achieved the highest test accuracy (82.5%) on the independent 20% test set. The multilayer perceptron (MLPClassifier), an ANN-based model, achieved a test accuracy of 80.0%. In the 10-fold cross-validation analysis, the support vector machine (SVM) achieved the highest average accuracy among the ML models (0.8438±0.1224), while the MLPClassifier reached an average accuracy of 0.6438±0.1384. The SHAP analysis of the LR model identified the left Fpost–AEtop variable, with an average absolute SHAP value of 0.2108, as the most influential predictor.Conclusions: Three-dimensional TMJ morphometry provides important and highly accurate information for sex classification. The Fpost–AEtop left measurement was consistently identified as the most informative predictor based on model interpretability analyses. This finding further supports the importance of regional morphology in forensic and anatomical classification studies.

Acibadem Universitesi Saglik Bilimleri DergisiVol. 17(October, November, December 2026)
Karabük University (TR), Okan University (TR)
Forensic Anthropology and Bioarchaeology Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon