Sex Prediction Based on Mandibular Morphometry Using Machine Learning Algorithms

Year 2025, Volume: 26 Issue: 4, 481 - 489, 25.12.2025

Hakan Öztürk , Eda Duygu İpek , Göksel Tuzcu

https://doi.org/10.69601/meandrosmdj.1789340

Abstract

Objective: The aim of this study is to investigate the potential of mandibular symphysis and mental trigone morphometry in determining sex in the Turkish population and to evaluate the performance of machine learning algorithms in sex prediction.
Methods: Computed tomography (CT) scans of 350 adult individuals (191 males, 159 females) were retrospectively analyzed. Eleven morphometric parameters were measured from midsagittal and three-dimensional reconstructions of the mandible. Sex differences were assessed using independent t-test or Mann–Whitney U test, with ROC curve analysis performed to determine discriminatory accuracy and optimal cut-off points. ML models (SVM, KNN, Naïve Bayes, Logistic Regression, Random Forest, and XGBoost) were implemented in R. The dataset was split into 70% training and 30% testing sets, with 5-fold cross-validation and grid search applied for model optimization. Performance metrics included accuracy, sensitivity, specificity, F1-score, and AUC.
Results: Ten of the eleven morphometric variables differed significantly between sexes (p<0.001), with males exhibiting larger dimensions. Symphyseal height (MSH, AUC=0.727), thickness (MST, AUC=0.726), and cross-sectional area (MSA, AUC=0.725) showed the highest discriminatory power. ML algorithms achieved strong classification performance (AUC range=0.821–0.851). Logistic Regression (F1-score=0.817, AUC=0.845) and XGBoost (F1-score=0.803, AUC=0.851) outperformed other models, while SHAP analysis identified PM_TML and MSH as the most influential predictors.
Conclusion: Mandibular morphometry, particularly symphysis parameters, provides reliable indicators for determining sexual dimorphism. Analyzing these parameters using Logistic Regression and XGBoost methods offers a robust and interpretable methodological framework for forensic sex determination in the Turkish population.

Keywords

Machine learning , Sex prediction , Sexual dimorphism , Mandibular morphometrics , Forensic anthropology

Makine Öğrenimi Algoritmaları ile Mandibular Morfometriye Dayalı Cinsiyet Tahmini

Abstract

Amaç: Bu çalışmanın amacı, mandibular simfiz ve trigonum mentale morfometrisinin Türk toplumunda cinsiyet belirlemedeki potansiyelini araştırmak ve cinsiyet tahmininde makine öğrenmesi algoritmalarının performanslarını değerlendirmektir.
Yöntemler: 350 yetişkin bireyin (191 erkek, 159 kadın) bilgisayarlı tomografi (BT) taramaları retrospektif olarak analiz edildi. Mandibulanın orta sagital ve üç boyutlu rekonstrüksiyonlarından on bir morfometrik parametre ölçüldü. Cinsiyet farklılıkları bağımsız t-testi veya Mann-Whitney U testi kullanılarak değerlendirildi ve ayırt edici doğruluğu ve optimal kesme noktalarını belirlemek için ROC eğrisi analizi yapıldı. ML modelleri (SVM, KNN, Naïve Bayes, Lojistik Regresyon, Rastgele Orman ve XGBoost) R'de uygulandı. Veri seti %70 eğitim ve %30 test setlerine bölündü ve model optimizasyonu için 5 kat çapraz doğrulama ve ızgara arama uygulandı. Performans ölçütleri arasında doğruluk, duyarlılık, özgüllük, F1-skoru ve AUC yer aldı.
Bulgular: On bir morfometrik değişkenin on tanesi cinsiyetler arasında önemli ölçüde farklılık gösterdi (p<0,001) ve erkekler daha büyük boyutlara sahipti. Symphyseal yükseklik (MSH, AUC=0,727), kalınlık (MST, AUC=0,726) ve kesit alanı (MSA, AUC=0,725) en yüksek ayırt edici gücü gösterdi. ML algoritmaları güçlü sınıflandırma performansı elde etti (AUC aralığı=0,821–0,851). Lojistik Regresyon (F1=0,817, AUC=0,845) ve XGBoost (F1=0,803, AUC=0,851) diğer modellerden daha iyi performans gösterirken, SHAP analizi PM_TML ve MSH'yi en etkili belirleyiciler olarak tanımladı.
Sonuç: Mandibular morfometri, özellikle simfiz parametreleri, cinsel dimorfizmin belirlenmesinde güvenilir göstergeler sağlamaktadır. Bu parametrelerin Lojistik Regresyon ve XGBoost yöntemleri kullanılarak analiz edilmesi, Türk popülasyonunda adli cinsiyet tayini açısından sağlam ve yorumlanabilir bir metodolojik çerçeve sunmaktadır

Keywords

Makine öğrenmesi , Cinsiyet tahmini , Cinsel dimorfizm , Mandibular morfometri , Adli antropoloji

References

• 1. Tunis TS, Sarig R, Cohen H, Medlej B, Peled N, May H. Sex estimation using computed tomography of the mandible. Int J Legal Med 2017; 131: 1691-700.
• 2. Mobin N, Vathsalya SK. Sexual dimorphism in adult human mandibles: a southern Indian study. Int J Anat Radiol Surg 2018; 7: 15-21.
• 3. Toneva D, Nikolova S, Agre G, Zlatareva D, Fileva N, Lazarov N. Sex estimation based on mandibular measurements. Anthropol Anz 2024; 81: 1-9.
• 4. Alias A, Ibrahim A, Bakar SNA, Shafie MS, Das S, Abdullah N, et al. Anthropometric analysis of mandible: an important step for sex determination. Clin Ter 2018; 169: e217-23.
• 5. Basso IB, Freitas PJF, Ferraz AX, Borkovski AJ, Borkovski AL, Santos RS, et al. Sex prediction through machine learning utilizing mandibular condyles, coronoid processes, and sigmoid notches features. PLoS One 2024; 19: e0312824.
• 6. Tokpınar A, Alkan Y. The role of mandibular morphological markers in determining sex and age: anatomical and anthropometric analysis. Folia Morphol (Warsz) 2025; in press.
• 7. Alarcón JA, Bastir M, Rosas A. Variation of mandibular sexual dimorphism across human facial patterns. Homo 2016; 67: 188-202.
• 8. Linjawi AI, Afify AR, Baeshen HA, Birkhed D, Zawawi KH. Mandibular symphysis dimensions in different sagittal and vertical skeletal relationships. Saudi J Biol Sci 2021; 28: 280-5.
• 9. Al-Khateeb SN, Al Maaitah EF, Alhaija ESA, Badran SA. Mandibular symphysis morphology and dimensions in different anteroposterior jaw relationships. Angle Orthod 2014; 84: 304-9.
• 10. Dobson SD, Trinkaus E. Cross-sectional geometry and morphology of the mandibular symphysis in Middle and Late Pleistocene Homo. J Hum Evol 2002; 43: 67-87.
• 11. Tunis TS, Hershkovitz I, May H, Vardimon AD, Sarig R, Shpack N. Variation in chin and mandibular symphysis size and shape in males and females: a CT-based study. Int J Environ Res Public Health 2020; 17: 4249.
• 12. Meneganzin A, Ramsey G, DiFrisco J. What is a trait? Lessons from the human chin. J Exp Zool B Mol Dev Evol 2024; 342: 65-75.
• 13. Schwartz JH, Tattersall I. The human chin revisited: what is it and who has it? J Hum Evol 2000; 38: 367-409.
• 14. Thayer ZM, Dobson SD. Sexual dimorphism in chin shape: implications for adaptive hypotheses. Am J Phys Anthropol 2010; 143: 417-25.
• 15. FIPAT. Terminologia Anatomica: International Anatomical Terminology. 2nd ed. Federative International Programme for Anatomical Terminology; 2019. Available from: https://fipat.library.dal.ca/TA2/
• 16. Garvin HM, Ruff CB. Sexual dimorphism in skeletal browridge and chin morphologies determined using a new quantitative method. Am J Phys Anthropol 2012; 147: 661-70.
• 17. Braun S, Ridel AF, L’Abbé EN, Theye CEG, Oettlé AC. Repeatability of a morphoscopic sex estimation technique for the mental eminence on micro-focus X-ray computed tomography models. Forensic Imaging 2022; 28: 200500.
• 18. Kuha A, Ackermann J, Junno JA, Oettlé A, Oura P. Deep learning in sex estimation from photographed human mandible using the Human Osteological Research Collection. Leg Med (Tokyo) 2024; 70: 102476.
• 19. Toy S, Secgin Y, Oner Z, Turan MK, Oner S, Senol D. A study on sex estimation by using machine learning algorithms with parameters obtained from computerized tomography images of the cranium. Sci Rep 2022; 12: 4278.
• 20. Prabha PS, Ganesan A, Lakshmi KC, Murugan AJ. Sex determination through analysis of mandibular indices using lateral cephalogram: an artificial intelligence diagnostics. Discov Artif Intell 2025; 5: 108.
• 21. Yang W, Liu X, Wang K, Hu J, Geng G, Feng J. Sex determination of three-dimensional skull based on improved backpropagation neural network. Comput Math Methods Med 2019; 2019: 9163547.
• 22. Bewes J, Low A, Morphett A, Pate FD, Henneberg M. Artificial intelligence for sex determination of skeletal remains: application of a deep learning artificial neural network to human skulls. J Forensic Leg Med 2019; 62: 40-3.
• 23. Toneva DH, Nikolova SY, Fileva NF, Zlatareva DK. Size and shape of human mandible: sex differences and influence of age on sex estimation accuracy. Leg Med (Tokyo) 2023; 65: 102322.
• 24. Mendes LC, Delrieu J, Gillet C, Telmon N, Maret D, Savall F. Sexual dimorphism of the mandibular conformational changes in aging human adults: a multislice computed tomographic study by geometric morphometrics. PLoS One 2021; 16: e0253564.
• 25. Okumura Y, Koizumi S, Suginouchi Y, Hikita Y, Kim YI, Adel M, et al. Chin morphology in relation to the skeletal pattern, age, gender, and ethnicity. Appl Sci 2022; 12: 12717.