Tooth shape and sex estimation: A 3D geometric morphometric landmark-based comparative analysis of artificial neural networks, support vector machines, and random forest models

Puntos clave

• This analysis aims to improve sex estimation accuracy from dental morphology using advanced machine learning models.
• 3D scans of dental casts from 120 individuals (60 males and 60 females)
• Landmarking nine tooth types for detailed analysis
• Employed Procrustes superimposition and Principal Component Analysis (PCA)
• Trained Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Random Forest (RF) models
• Assessed performance using 5-fold cross-validation metrics like accuracy and F1 score
• Random Forest achieved the highest accuracy of 97.95% for estimating sex from mandibular second premolars
• SVM yielded moderate accuracy ranging from 70% to 88%
• ANN showed the lowest performance with accuracy between 58% and 70%, particularly struggling in female classification recall
• Mandibular premolars displayed the most dimorphic characteristics for sex estimation

Resumen