Gender Prediction Using Machine Learning Algorithms with Parameters Obtained from Calcaneus

Öz

Background: Partial or total disruption of body integrity may occur in cases such as war, natural disasters and accidents. In such cases, the importance of the calcaneus bone, which has a hard and minimal structure, increases for identification. With this hypothesis, the aim of this study is to estimate gender from the calcaneus by utilising the current approach of machine learning (ML) algorithm. Materials and Methods: The study was performed on X-Ray images of 200 female and 200 male subjects aged 18-65 years. Maximum length, facies articularis cuboidea height, maximum width, body width, minimum length, anteroposterior length of the calcaneus, posterior facet angle, anterior angle of the cuboid facet of the calcaneus, facet height, posterior facet length, anterior proces length, calcaneus inclination angle, talocalcaneal angle, Böhler angle, Gissane angle and calcaneus tuber angle were measured. Then the obtained data were used in the input of ML algorithms. Results: As a result, a highly accurate and reliable sex prediction rate between 0.86-0.91 was obtained with ML algorithms. In addition, it was found that the maximum width of the calcaneus parameter made the highest contribution to sex prediction among the parameters with SHapley Additive exPlanations. Conclusions: As a result of our study, it was found that calcaneus with minimal and rigid structure provided high accuracy in terms of gender prediction using ML algorithms. In this respect, we think that this study will be a reference for forensic and morphometric studies.

Anahtar Kelimeler

• Calcaneus
• X-Ray
• machine learning algorithms
• gender prediction

Etik Beyan

The study was performed with the decision of İzmir Bakırçay University Non-Interventional Local Ethics Committee dated 31.05.2023 and numbered 1064.

Teşekkür

I would like to thank all the authors for their contributions.

Calcaneus’tan Elde Edilen Parametrelerle Makine Öğrenme Algoritmaları Kullanılarak Cinsiyet Tahmini

Öz

Amaç: Savaş, doğal afetler ve kazalar gibi durumlarda vücut bütünlüğünün kısmen veya tamamen bozulması söz konusu olabilir. Bu gibi durumlarda sert ve minimal bir yapıya sahip olan kalkaneus kemiğinin kimlik tespiti için önemi artmaktadır. Bu hipotezle, bu çalışmanın amacı makine öğrenmesi (ML) algoritmasının güncel yaklaşımını kullanarak calcaneus’tan cinsiyet tahmini yapmaktır. Materyal ve Metot: Çalışma 18-65 yaş arası 200 kadın ve 200 erkek bireyin X-Ray görüntüleri üzerinde gerçekleştirildi. Maksimum uzunluk, facies articularis cuboidea yüksekliği, maksimum genişlik, gövde genişliği, minimum uzunluk, calcaneus’un ön-arka uzunluğu, arka faset açısı, calcaneus’un küboid fasetinin ön açısı, faset yüksekliği, arka faset uzunluğu, ön proces uzunluğu, calcaneus inklinasyon açısı, talokalkaneal açı, Böhler açısı, Gissane açısı ve kalkaneus tuber açısı ölçüldü. Daha sonra elde edilen veriler ML algoritmalarının girişinde kullanıldı. Bulgular: Sonuç olarak, ML algoritmaları ile 0,86-0,91 arasında oldukça doğru ve güvenilir bir cinsiyet tahmin oranı elde edilmiştir. Ayrıca SHapley Additive exPlanations ile parametreler arasında cinsiyet tahminine en yüksek katkıyı calcaneus’un maksimum genişliği parametresinin yaptığı tespit edildi. Sonuçlar: Çalışmamız sonucunda minimal ve rijit yapıya sahip calcaneus’un ML algoritmaları kullanılarak cinsiyet tahmini açısından yüksek doğruluk sağladığı tespit edilmiştir. Bu açıdan bu çalışmanın adli ve morfometrik çalışmalar için referans olacağını düşünmekteyiz.

Anahtar Kelimeler

• Calcaneus
• X-Ray
• makine öğrenmesi algoritmaları
• cinsiyet tahmini

Kaynakça

1. 1. Krishan K, Chatterjee PM, Kanchan T, Kaur S, Baryah N, Singh R. A review of sex estimation techniques during examination of skeletal remains in forensic anthropolo-gy casework. Forensic science international. 2016;261:165. e1-. e8.
2. 2. Singh R, Mishra SR, Passey J, Kumar P, Singh S, Sinha P, et al. Sexual dimorphism in adult human mandible of North Indian origin. Forensic medicine and anatomy re-search. 2015;3(03):82.
3. 3. Zeyfeoğlu Y, Hancı İH. İnsanlarda kimlik tespiti. Türk Tabipleri Birliği Sürekli Tıp Eğitimi Dergisi. 2001;375:4659-62.
4. 4. Case DT, Ross AH. Sex determination from hand and foot bone lengths. Journal of forensic sciences. 2007;52(2):264-70.
5. 5. Issa SY, Khanfour AA, Kharoshah M. A model for stature estimation and sex prediction using percutaneous ulnar and radial lengths in autopsied adult Egyptians. Egyptian journal of forensic sciences. 2016;6(2):84-9.
6. 6. Marlow EJ, Pastor RF. Sex determination using the second cervical vertebra—a test of the method. Journal of forensic sciences. 2011;56(1):165-9.
7. 7. Acar A. Yoncatepe toplumunda Calcaneus ve Talus kemiklerinden cinsiyet ve boy tahmini. Antropoloji. 2014(28):109-22.
8. 8. Curate F, Umbelino C, Perinha A, Nogueira C, Silva AM, Cunha E. Sex determination from the femur in Portu-guese populations with classical and machine-learning classifiers. Journal of Forensic and Legal Medicine. 2017;52:75-81.

9. 9. Ekizoglu O, Inci E, Palabiyik FB, Can IO, Er A, Bozdag M, et al. Sex estimation in a contemporary Turkish popula-tion based on CT scans of the calcaneus. Forensic sci-ence international. 2017;279:310. e1-. e6.
10. 10. Giurazza F, Schena E, Del Vescovo R, Cazzato R, Mortato L, Saccomandi P, et al., editors. Sex estimation from scapular measurement by CT scans evaluations in a Caucasian population. Engineering in Medicine and Biol-ogy Society (EMBC), 2013 35th Annual International Conference of the IEEE; 2013: IEEEX.
11. 11. Bhardwaj R, Nambiar AR, Dutta D, editors. A study of machine learning in healthcare. 2017 IEEE 41st annual computer software and applications conference (COMP-SAC); 2017: IEEE.
12. 12. Ariu D, Giacinto G, Roli F, editors. Machine learning in computer forensics (and the lessons learned from ma-chine learning in computer security). Proceedings of the 4th ACM workshop on Security and artificial intelligence; 2011.
13. 13. Awais M, Naeem F, Rasool N, Mahmood S. Identifica-tion of sex from footprint dimensions using machine learning: a study on population of Punjab in Pakistan. Egyptian Journal of Forensic Sciences. 2018;8:1-9.
14. 14. Erkartal HŞ, Tatlı M, Secgin Y, Toy S, Duman BS. Gender estimation with parameters obtained from the upper dental arcade by using machine learning algorithms and artificial neural networks. European Journal of Thera-peutics. 2023;29(3):352-8.
15. 15. Torun Bİ, Çay N. Ayak Arkus Açısı ve Ayak Uzunluğu Arasındaki İlişki. Kafkas Journal of Medical Sciences. 2018;8(3):172-7.
16. 16. Doğruyol G, Çimen M. Medial ve lateral ark açıları ile ayak uzunluğunun yaş, cinsiyet ve taraf farklılığı açısın-dan radyolojik olarak incelenmesi. Turkish Journal of Science and Health. 2020;2(1):76-83.
17. 17. Cekdemir YE, Mutlu U, Karaman G, Balci A. Evaluation of computed tomography images of calcaneus for esti-mation of sex. La radiologia medica. 2021;126(8):1064-73.
18. 18. Ahmed KF, Wang G, Silander J, Wilson AM, Allen JM, Horton R, et al. Statistical downscaling and bias correc-tion of climate model outputs for climate change im-pact assessment in the US northeast. Global and Plane-tary Change. 2013;100:320-32.
19. 19. Senol D, Secgin Y, Duman BS, Toy S, Oner Z. Sex and age estimation with machine learning algorithms with pa-rameters obtained from cone beam computed tomog-raphy images of maxillary first molar and canine teeth. Egyptian Journal of Forensic Sciences. 2023;13(1):27.
20. 20. Introna Jr F, Di Vella G, Pietro Campobasso C, Dragone M. Sex determination by discriminant analysis of calca-nei measurements. Journal of Forensic Sciences. 1997;42(4):725-8.
21. 21. Peckmann TR, Orr K, Meek S, Manolis SK. Sex determi-nation from the calcaneus in a 20th century Greek popu-lation using discriminant function analysis. Science & Justice. 2015;55(6):377-82.
22. 22. Bidmos MA, Asala SA. Discriminant function sexing of the calcaneus of the South African whites. Journal of fo-rensic sciences. 2003;48(6):JFS2003104.
23. 23. Moore MK, DiGangi EA, Ruíz FPN, Davila OJH, Medina CS. Metric sex estimation from the postcranial skeleton for the Colombian population. Forensic science interna-tional. 2016;262:286. e1-. e8.
24. 24. Böhler L. Diagnosis, pathology, and treatment of frac-tures of the os calcis. JBJS. 1931;13(1):75-89.
25. 25. Seyahi A, Uludag S, Koyuncu L, Atalar A, Demirhan M. The calcaneal angles in the Turkish population. Acta or-thopaedica et traumatologica turcica. 2009;43(5):406-11.
26. 26. Jiao L, Li H, Liao T, Han Z, Wu H, Jiang L. Impact of percutaneous poking reduction combined with minimal-ly invasive plate internal fixation on foot function and complications of patients with Sanders type II and III calcaneal fractures. American Journal of Translational Research. 2021;13(5):5329.
27. 27. Riepert T, Drechsler T, Schild H, Nafe B, Mattern R. Estimation of sex on the basis of radiographs of the cal-caneus. Forensic science international. 1996;77(3):133-40.
28. 28. Agoada D. The relationship between linear osteological and radiographic measurements of the human calcane-us and talus. The Anatomical Record. 2018;301(1):21-33.
29. 29. Kumar A, Rastogi S, Haider Y, Kumar S, Chauhan S, Passey J. Morphometric variations of the lateral surface of calcaneus: Can standard plate sizes fit all? Journal of Clinical Orthopaedics and Trauma. 2021;13:156-62.