Sex determination from cephalometric radiography using data mining

Öz

Context—The skull is a crucial reference structure in skeletal analysis due to its hard and durable nature, which enables it to resist environmental degradation and preserve its morphological characteristics over extended periods. Because of these properties, craniofacial structures are widely used in forensic science, anthropology, and medical studies for identification purposes, including sex determination.

Objective—This study aims to perform sex determination using cephalometric radiography (CR) and to evaluate the effectiveness of statistical and correlation-based feature selection methods in improving classification performance. Additionally, the study investigates whether a reduced set of informative cephalometric features (CFs) can achieve comparable accuracy to the full feature set, thereby increasing model efficiency and interpretability.

Method—A dataset was created by extracting the and coordinates of 19 cephalometric landmarks (CLs) from CR images of 300 patients (150 females and 150 males). Based on these landmarks, 8 Euclidean distance (8-ED), 3 angular (3-A), and 3 frontal sinus (3-FS) measurements were calculated, resulting in a total of 14 CFs. Initially, statistical analyses were conducted to determine the relationship between each feature and the sex variable, identifying the most discriminative parameters. Following this, correlation analysis was performed to examine inter-feature relationships and to detect redundant information. Feature combinations with high correlation and strong association with the target variable were generated using predefined threshold values. To evaluate the effectiveness of these feature sets, comparative classification experiments were carried out using machine learning (ML) algorithms, with a particular focus on performance differences between full and reduced feature subsets.

Results—The experimental results showed that using the Support Vector Machine (SVM) classifier with 4 CFs yielded an accuracy of 89.33%. Notably, a comparable level of classification performance was achieved using only 4 CFs selected based on strong statistical significance and high correlation with the sex class. This finding suggests that many of the original CFs contain redundant or less informative data, and that careful feature selection can simplify the model without compromising its predictive capability.

Conclusion—The results demonstrate that effective feature selection based on statistical and correlation analyses can significantly reduce the number of CFs required while maintaining high classification performance in sex determination from cephalometric radiographs. This reduction not only improves computational efficiency but also enhances model interpretability and applicability in real-world scenarios. The proposed approach provides a robust framework for optimizing feature sets in similar biomedical classification problems.

Anahtar Kelimeler

• Cephalometric analysis
• Correlation analysis
• Data mining
• Machine learning
• Statistical analysis

Etik Beyan

The study was approved by the Ethics Committee of Selcuk University Faculty of Dentistry (protocol code 2023/22, approval date: 4 May 2023).

Veri madenciliği ile sefalometrik radyografiden cinsiyet belirleme

Öz

Arka Plan—Kafatası, sert ve dayanıklı yapısı sayesinde çevresel bozulmaya direnç göstermesi ve morfolojik özelliklerini uzun süre koruyabilmesi nedeniyle iskelet analizinde önemli bir referans yapıdır. Bu özellikleri sayesinde kraniyofasiyal yapılar; adli bilimler, antropoloji ve tıp çalışmalarında cinsiyet belirleme de dâhil olmak üzere kimliklendirme amaçlarıyla yaygın olarak kullanılmaktadır.

Amaç—Bu çalışmanın amacı, sefalometrik radyografi (CR) kullanılarak cinsiyet belirlemesi gerçekleştirmek ve istatistiksel ve korelasyon tabanlı özellik seçimi yöntemlerinin sınıflandırma performansını iyileştirmedeki etkinliğini değerlendirmektir. Ayrıca çalışma, bilgilendirici sefalometrik özelliklerden (CF'ler) oluşan daha küçük bir özellik kümesinin, tüm özellik kümesine kıyasla benzer doğruluk düzeyine ulaşıp ulaşamayacağını araştırarak modelin verimliliğini ve yorumlanabilirliğini artırmayı amaçlamaktadır.

Yöntem—300 hastanın (150 kadın ve 150 erkek) CR görüntülerinden 19 sefalometrik referans noktasının (CL'ler) x ve y koordinatlarının çıkarılmasıyla bir veri seti oluşturulmuştur. Bu referans noktalarına dayalı olarak 8 Öklid mesafesi (8-ED), 3 açısal (3-A) ve 3 frontal sinüs (3-FS) ölçümü hesaplanmış ve toplam 14 sefalometrik özellik (CF) elde edilmiştir. İlk olarak, her bir özellik ile cinsiyet değişkeni arasındaki ilişkiyi belirlemek ve en ayırt edici parametreleri tespit etmek amacıyla istatistiksel analizler gerçekleştirilmiştir. Bunun ardından, özellikler arasındaki ilişkileri incelemek ve tekrarlayan bilgileri belirlemek amacıyla korelasyon analizi yapılmıştır. Hedef değişkenle yüksek korelasyona ve güçlü ilişkiye sahip özellik kombinasyonları, önceden belirlenmiş eşik değerleri kullanılarak oluşturulmuştur. Bu özellik kümelerinin etkinliğini değerlendirmek amacıyla, makine öğrenmesi (ML) algoritmaları kullanılarak karşılaştırmalı sınıflandırma deneyleri gerçekleştirilmiş ve özellikle tüm özellik kümesi ile azaltılmış özellik alt kümeleri arasındaki performans farklılıklarına odaklanılmıştır.

Bulgular—4 sefalometrik özellik (CF) kullanılarak Destek Vektör Makinesi (SVM) sınıflandırıcısı ile %89,33 doğruluk elde edildiğini göstermiştir. Dikkat çekici olarak, cinsiyet sınıfıyla güçlü istatistiksel anlamlılığa ve yüksek korelasyona dayalı olarak seçilen yalnızca 4 CF kullanıldığında da karşılaştırılabilir bir sınıflandırma performansı elde edilmiştir. Bu bulgu, başlangıçta kullanılan CF'lerin birçoğunun tekrarlayan veya daha az bilgilendirici veriler içerdiğini ve dikkatli bir özellik seçiminin tahmin yeteneğinden ödün vermeden modeli basitleştirebileceğini göstermektedir.

Sonuç—İstatistiksel ve korelasyon analizlerine dayalı etkili özellik seçiminin, sefalometrik radyografilerden cinsiyet belirleme işleminde yüksek sınıflandırma performansını korurken gerekli CF sayısını önemli ölçüde azaltabileceğini göstermektedir. Bu azaltma yalnızca hesaplama verimliliğini artırmakla kalmamakta, aynı zamanda modelin yorumlanabilirliğini ve gerçek dünya senaryolarındaki uygulanabilirliğini de geliştirmektedir. Önerilen yaklaşım, benzer biyomedikal sınıflandırma problemlerinde özellik kümelerinin optimize edilmesi için sağlam bir çerçeve sunmaktadır.

Anahtar Kelimeler

• Sefalometrik analiz
• Korelasyon analizi
• Veri madenciliği
• Makine öğrenimi
• İstatistiksel analiz

Etik Beyan

Çalışma, Selçuk Üniversitesi Diş Hekimliği Fakültesi Etik Kurulu tarafından onaylanmıştır (protokol kodu: 2023/22, onay tarihi: 4 Mayıs 2023).

Kaynakça

1. K. R. Patil, R. N. Mody, “Determination of sex by discriminant function analysis and stature by regression analysis: a lateral cephalometric study”, Forensic Science International, 147(2–3), 175–180, 2005. https://doi.org/10.1016/j.forsciint.2004.09.071.
2. A. Mathur, A. Sande, M. Risbud, P. Ramdurg, S. R. Ashwinirani, “Determination of sex by discriminant function analysis of lateral radiographic cephalometry using angular, linear and proportional cephalometric variables in Western Maharashtrian population”, Journal of Oral Medicine, Oral Surgery, Oral Pathology and Oral Radiology, 3(3), 153–157, 2017, https://joooo.org/archive/volume/3/issue/3/article/16223/pdf.
3. M. Ülgen, Ortodonti: anomaliler, sefalometri, etyoloji, büyüme ve gelişim, tanı, Ankara Üniversitesi Diş Hekimliği Fakültesi Yayınları, Ankara, Türkiye, 2000.
4. A. E. Athanasiou, Orthodontic Cephalometry, Mosby-Wolfe, London, 1995.
5. B. Phulari, An atlas on cephalometric landmarks, JB Medical Publishers Ltd, 2013.
6. T. Rakosi, An atlas and manual of cephalometric radiography, Wolfe Medical Publications, 1982.
7. T. C. Niño-Sandoval, S. V. G. Pérez, F. A. González, R. A. Jaque, C. Infante-Contreras, “Use of automated learning techniques for predicting mandibular morphology in skeletal class I, II and III”, Forensic Science International, 281, 187.e1, 2017. https://doi.org/10.1016/j.forsciint.2017.10.004.
8. A. K. Subramanian, Y. Chen, A. Almalki, G. Sivamurthy, D. Kafle, “Cephalometric analysis in orthodontics using artificial intelligence—A comprehensive review”, BioMed Research International, 2022, 1880113, 2022, https://doi.org/10.1155/2022/1880113.

9. V. G. Naikmasur, R. Shrivastava, S. Mutalik, “Determination of sex in South Indians and immigrant Tibetans from cephalometric analysis and discriminant functions”, Forensic Science International, 197(1-3), 2010. https://doi.org/10.1016/j.forsciint.2009.12.052.
10. T. Hsiao, S. Tsai, S. Chou, J. Pan, Y. Tseng, H. Chang, H. Chen, “Sex determination using discriminant function analysis in children and adolescents: a lateral cephalometric study,” International Journal of Legal Medicine, 124(2), 155–160, 2010. https://doi.org/10.1007/s00414-009-0412-1.
11. A. Binnal, B. K. Devi, “Identification of sex using lateral cephalogram: role of cephalofacial parameters”, Journal of Indian Academy of Oral Medicine and Radiology, 24(4), 280–283, 2012.
12. M. IP, P. M. David, “Determination of sexual dimorphism and stature using lateral cephalogram”, Journal of Indian Academy of Oral Medicine and Radiology, 25(2), 1–5, 2013.
13. A. Yağcı, S. K. Büyük, “Dengeli yüz oranlarına ve normal oklüzyona sahip genç Türk erişkinlerin McNamara sefalometrik normları”, Journal of Health Sciences, 22(1), 1–6, 2013. https://izlik.org/JA25PP46KG.
14. R. U. Mathur, A. M. Mahajan, R. C. Dandekar, R. B. Patil, “Determination of sex using discriminant function analysis in young adults of Nashik: a lateral cephalometric study”, Journal of Advanced Medical and Dental Sciences Research, 2(1), 21–25, 2014. https://jamdsr.com/pdf1/DeterminationofSexusingDiscriminantFunction.pdf
15. D. D. Divakar, J. John, A. A. Al Kheraif, S. Mavinapalla, R. Ramakrishnaiah, S. Vellappally, M. I. Hashem, M.H. N. Dalati, B. H. Durgesh, R. A. Safadi, S. Anil, “Sex determination using discriminant function analysis in Indigenous (Kurubas) children and adolescents of Coorg, Karnataka, India: a lateral cephalometric study”, Saudi Journal of Biological Sciences, 23(6), 782–788, 2016, https://doi.org/10.1016/j.sjbs.2016.05.008.
16. R. Qaq, S. Mânica, G. Revie, “Sex estimation using lateral cephalograms: a statistical analysis”, Forensic Science International: Reports, 1, 100034, 2019. https://doi.org/10.1016/j.fsir.2019.100034.
17. A. Bagherpour, N. Anbiaee, S. Motaghi, A. Jahanbin, “Gender determination using digital lateral cephalograms: a discriminant function analysis”, Journal of Dental Materials and Techniques, 9(4), 221–230, 2020. https://doi.org/10.22038/jdmt.2021.48358.1371.
18. M. Farhadian, F. Salemi, A. Shokri, A. Safi, S. Rahimpanah, “Comparison of data mining algorithms for sex determination based on mastoid process measurements using cone-beam computed tomography”, Imaging Science in Dentistry, 50(4), 323–330, 2020. https://doi.org/10.5624/isd.2020.50.4.323.
19. C. P. Alves, C. Costa, E. Michel-Crosato, M. G. H. Biazevic, “Use of the frontal sinus to evaluate sexual dimorphism in a Brazilian sample”, Forensic Imaging, 33, 200548, 2023. https://doi.org/10.1016/j.fri.2023.200548.
20. S. Kenawy, M. Ellabban, A. Fadel, “A novel frontal sinus index protocol for gender determination: a retrospective lateral cephalometric study for Egyptians”, Egyptian Dental Journal, 67(3), 2199–2204, 2021. https://doi.org/10.21608/edj.2021.76583.1639.
21. B. Zheng, Y. Zhong, N. A. Al-Worafi, Y. Liu, “The dimensional and morphological assessment of the frontal sinus in sex estimation among different populations”, Head & Face Medicine, 19(1), 8, 2023, https://doi.org/10.1186/s13005-023-00355-4.
22. J. McDonnell, R. McNeill, R. West, “Advancement genioplasty: a retrospective cephalometric analysis of osseous and soft tissue changes”, Journal of Oral Surgery, 35(8), 640–647, 1977. https://europepmc.org/article/med/267188.
23. W. Hochban, U. Brandenburg, “Morphology of the viscerocranium in obstructive sleep apnoea syndrome cephalometric evaluation of 400 patients”, Journal of Cranio-Maxillofacial Surgery, 22(4), 205–213, 1994. https://doi.org/10.1016/S1010-5182(05)80559-1.
24. R. S. Nanda, R. M. Merill, “Cephalometric assessment of sagittal relationship between maxilla and mandible”, American Journal of Orthodontics and Dentofacial Orthopedics, 105(4), 328–344, 1994. https://doi.org/10.1016/S0889-5406(94)70127-X.
25. M. Y. Hajeer, A. F. Ayoub, D. T. Millett, “Three-dimensional assessment of facial soft-tissue asymmetry before and after orthognathic surgery”, British Journal of Oral and Maxillofacial Surgery, 42(5), 396–404, 2004. https://doi.org/10.1016/j.bjoms.2004.05.006.
26. B. H. Broadbent, “A new X-ray technique and its application to orthodontia”, Angle Orthodontist, 51(2), 93–114, 1981.
27. A. J. Pacini, “Roentgen ray anthropometry of the skull”, Journal of Radiology, 3, 230–238, 1922.
28. A. Kamoen, L. Dermaut, R. Verbeeck, “The clinical significance of error measurement in the interpretation of treatment results”, European Journal of Orthodontics, 23(5), 569–578, 2001. https://doi.org/10.1093/ejo/23.5.569.
29. H. L. D. Da Silveira, H. E. D. Silveira, “Reproducibility of cephalometric measurements made by three radiology clinics”, Angle Orthodontist, 76(3), 394–399, 2006.
30. A. P. R. Durao, A. Morosolli, P. Pittayapat, N. Bolstad, A. P. Ferreira, R. Jacobs, “Cephalometric landmark variability among orthodontists and dentomaxillofacial radiologists: a comparative study”, Imaging Science in Dentistry, 45(4), 213–220, 2015. https://doi.org/10.5624/isd.2015.45.4.213.
31. C. Lindner, C. W. Wang, C. T. Huang, C. H. Li, S. W. Chang, T. F. Cootes, “Fully automatic system for accurate localization and analysis of cephalometric landmarks in lateral cephalograms”, Scientific Reports, 6, 33581, 2016. https://doi.org/10.1038/srep33581.
32. R. H. Biggerstaff, “Craniofacial characteristics as determinants of age, sex, and race in forensic dentistry”, Dental Clinics of North America, 21(1), 85–97, 1977. https://doi.org/10.1016/S0011-8532(22)00892-8.
33. S. Ö. Arık, B. Ibragimov, L. Xing, “Fully automated quantitative cephalometry using convolutional neural networks”, Journal of Medical Imaging, 4(1), 014501, 2017. https://doi.org/10.1117/1.JMI.4.1.014501.
34. R. Agrawal, T. Imielinski, A. Swami, “Mining association rules between sets of items in large databases”, ACM SIGMOD Record, 22(2), 207–216, 1993. https://doi.org/10.1145/170036.170072.
35. D. A. Freedman, Statistical models: theory and practice, New York, NY, USA, Cambridge University Press, 2009.
36. T. Hastie, R. Tibshirani, J. Friedman, The elements of statistical learning: data mining, inference, and prediction, Second Ed., New York, NY, USA, Springer, 2009. https://doi.org/10.1007/978-0-387-84858-7.
37. I. Guyon, A. Elisseeff, “An introduction to variable and feature selection”, The Journal of Machine Learning Research, 3, 1157–1182, 2003.
38. I. T. Jolliffe, Principal component analysis, Second Ed., New York, NY, USA, Springer-Verlag, 2002. https://doi.org/10.1007/b98835.
39. J. Han, M. Kamber, Data mining: concepts and techniques, 2nd ed., San Francisco, CA, USA, Morgan Kaufmann Publishers, Elsevier 2006.
40. A. Vellido, J. Martín-Guerrero, P. J. G. Lisboa, “Making machine learning models interpretable”, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, Belgium, 25-27 April 2012, 163-172.
41. E. Frank, L. Trigg, G. Holmes, I. H. Witten, “Technical note: naive Bayes for regression”, Machine Learning, 41, 5–25, 2000.
42. A. Verma, “Evaluation of classification algorithms with solutions to class imbalance problem on bank marketing dataset using WEKA”, International Research Journal of Engineering and Technology, 6(3), 54–61, 2019.
43. S. Bharati, M. A. Rahman, P. Podder, “Breast cancer prediction applying different classification algorithm with comparative analysis using WEKA”, Proceedings of the 4th International Conference on Electrical Engineering and Information and Communication Technology (iCEEiCT), Dhaka, Bangladesh, 13-15 September 2018, 580-583. https://doi.org/10.1109/CEEICT.2018.8628084.
44. V. N. Vapnik, “An overview of statistical learning theory”, IEEE Transactions on Neural Networks, 10(5), 988–999, 1999. https://doi.org/10.1109/72.788640.
45. Y. Huang, S. Meng, X. Li, W. Fan, “A classification method for wood vibration signals of Chinese musical instruments based on GMM and SVM”, Traitement du Signal, 35(2), 137-151, 2018. https://doi.org/10.3166/TS.35.137-151.
46. T. Cover; P. Hart, "Nearest neighbor pattern classification", IEEE Transactions on Information Theory, 13(1), 21–27, 1967. https://doi.org/10.1109/TIT.1967.1053964.
47. Y. Z. Ayık, A. Özdemir, U. Yavuz, “Lise türü ve lise mezuniyet başarısının, kazanılan fakülte ile ilişkisinin veri madenciliği tekniği ile analizi”, Atatürk Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 10(2), 441–454, 2010. https://izlik.org/JA32RJ25TZ.
48. J. Han, M. Kamber, J. Pei, Data Mining: Concepts and Techniques, 4th ed., Morgan Kaufmann, 2022.
49. L. Breiman, “Bagging predictors”, Machine Learning, 24(2), 123–140, 1996. https://doi.org/10.1007/bf00058655.
50. S. A. A. Shah, W. Aziz, M. Arif, M. S. A. Nadeem, “Decision trees based classification of cardiotocograms using bagging approach”, Proceedings of the 13th International Conference on Frontiers of Information Technology (FIT), Islamabad, Pakistan, 14-16 December 2015. https://doi.org/10.1109/FIT.2015.14.
51. Y. Freund, R. E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting”, Journal of Computer and System Sciences, 55(1), 119–139, 1997. https://doi.org/10.1006/jcss.1997.1504.
52. O. Sagi, L. Rokach, “Ensemble learning: a survey”, Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery, 8(4), e1249, 2018. https://doi.org/10.1002/widm.1249.
53. V. A. Dev, M. R. Eden, “Formation lithology classification using scalable gradient boosted decision trees”, Computers & Chemical Engineering, 128, 392–404, 2019. https://doi.org/10.1016/j.compchemeng.2019.06.001.
54. L. Breiman, “Random forests”, Machine Learning, 45, 5–32, 2001.
55. C. Avcı, A. Akbaş, “Sleep apnea classification based on respiration signals by using ensemble methods”, Biomedical Materials and Engineering, 26, S1703–S1710, 2015. https://doi.org/10.3233/BME-151470.
56. B. Akkoç, A. Arslan, H. Kök, “Automatic gender determination from 3D digital maxillary tooth plaster models based on the random forest algorithm and discrete cosine transform”, Computer Methods and Programs in Biomedicine, 143, 59–65, 2017. https://doi.org/10.1016/j.cmpb.2017.03.001.
57. W. H. Chen, S. H. Hsu, H. P. Shen, “Application of SVM and ANN for intrusion detection”, Computers & Operations Research, 32(10), 2617–2634, 2005. https://doi.org/10.1016/j.cor.2004.03.019.
58. A. K. Dwivedi, “Artificial neural network model for effective cancer classification using microarray gene expression data”, Neural Computing and Applications, 29(12), 1545–1554, 2018. https://doi.org/10.1007/s00521-016-2701-1.
59. A. Ahad, A. Fayyaz, T. Mehmood, “Speech recognition using multilayer perceptron”, Proceedings of the IEEE Students Conference (ISCON’02), Lahore, Pakistan, 16-17 August 2002. https://doi.org/10.1109/ISCON.2002.1215948.
60. F. Z. Solak, “COVID-19 detection using variational mode decomposition of cough sounds”, Konya Journal of Engineering Sciences, 11(2), 354–369, 2023. https://doi.org/10.36306/konjes.1110235.
61. P. Mishra, C.M. Pandey, U.Singh, A. Gupta, C. Sahu, A. Keshri, "Descriptive Statistics and Normality Tests for Statistical Data", Annals of Cardiac Anaesthesia., 22(1), 67–72, 2019. https://doi.org/10.4103/aca.ACA_157_18.
62. N. Smirnov, “On the estimation of the discrepancy between empirical curves of distribution for two independent samples”, Moscow University Mathematics Bulletin, 2, 3–26, 1939.
63. H. B. Mann, D. R. Whitney, “On a test of whether one of two random variables is stochastically larger than the other”, Annals of Mathematical Statistics, 18(1), 50–60, 1947. https://doi.org/10.1214/aoms/1177730491.
64. H. C. Jan, W. H. Yang, C. H. Ou, “Combination of the preoperative systemic immune-inflammation index and monocyte-lymphocyte ratio as a novel prognostic factor in patients with upper-tract urothelial carcinoma”, Annals of Surgical Oncology, 26(2), 669–684, 2019. https://doi.org/10.1245/s10434-018-6942-3.
65. R. K. Badam, M. Manjunath, M. S. Rani, “Determination of sex by discriminant function analysis of lateral radiographic cephalometry”, Journal of Indian Academy of Oral Medicine and Radiology, 23(3), 179–183, 2011, https://www.researchgate.net/publication/284833858_Determination_of_Sex_by_Discriminant_Function_Analysis_of_Lateral_Radiographic_Cephalometry.
66. S. A. Veyre-Goulet, C. Mercier, O. Robin, C. Guérin, “Recent human sexual dimorphism study using cephalometric plots on lateral teleradiography and discriminant function analysis”, Journal of Forensic Sciences, 53(4), 786–789, 2008. https://doi.org/10.1111/j.1556-4029.2008.00759.x.
67. A. Mitsea, I. Pouliezou, N. Christoloukas, I. Sifakakis, “Is lateral cephalometric analysis useful to sex estimation in forensic science? A cross-sectional study”, Journal of Forensic and Legal Medicine, 118, 103077, 2026. https://doi.org/10.1016/j.jflm.2026.103077.