Sex Estimation from 3D Analysis of Paranasal Sinuses: A Multicenter Study Using Deep Learning and Machine Learning

• Maria Fernanda A. F. Scarcela UFC
• Antonio Everton C. Teixeira UFC
• Diego S. de Mendonça Centro de Ortodontia Paulo Picanço
• Saulo A. F. de Oliveira IFCE
• Bruno R. dos Santos Silva UFC
• Carlos Caminha UFC
• Fábio W. G. Costa UFC
• Wellington Franco UFC

DOI: https://doi.org/10.5753/eniac.2025.14530

Abstract

The paranasal sinuses are used in forensic sex identification, typically through manual measurements on computed tomography (CT) images. This study proposes to automate this classification using 3D convolutional neural networks (ResNet-50, DenseNet-201) integrated with supervised algorithms (SVM, Random Forest, MLP, KNN). Eight combinations were tested, with ResNet-50 + RF achieving the best performance, with an accuracy of 92.60%, precision of 91.18%, recall of 94.55%, and F1-score of 92.64%. The results indicate high potential for the combined use of deep learning and supervised learning to determine sex from segmented masks of the paranasal sinuses, with relevant application in the forensic field.

References

Bewes, J., Low, A., Morphett, A., Pate, F. D., and Henneberg, M. (2019). Artificial intelligence for sex determination of skeletal remains: Application of a deep learning artificial neural network to human skulls. Journal of forensic and legal medicine, 62:40–43.

Cao, Y., Ma, Y., Vieira, D. N., Guo, Y., Wang, Y., Deng, K., Chen, Y., Zhang, J., Qin, Z., Chen, F., et al. (2021). A potential method for sex estimation of human skeletons using deep learning and three-dimensional surface scanning. International Journal of Legal Medicine, 135(6):2409–2421.

Cavaignac, E., Savall, F., Faruch, M., Reina, N., Chiron, P., and Telmon, N. (2016). Geometric morphometric analysis reveals sexual dimorphism in the distal femur. Forensic science international, 259:246–e1.

Cohen, O., Warman, M., Fried, M., Shoffel-Havakuk, H., Adi, M., Halperin, D., and Lahav, Y. (2018). Volumetric analysis of the maxillary, sphenoid and frontal sinuses: A comparative computerized tomography based study. Auris Nasus Larynx, 45(1):96–102.

Cox, M., Malcolm, M., and Fairgrieve, S. I. (2009). A new digital method for the objective comparison of frontal sinuses for identification. Journal of forensic sciences, 54(4):761–772.

da Silva, R. L. B., Yang, S., Kim, D., Kim, J. H., Lim, S.-H., Han, J., Kim, J.-M., Kim, J.-E., Huh, K.-H., Lee, S.-S., et al. (2024). Automatic segmentation and classification of frontal sinuses for sex determination from cbct scans using a two-stage anatomy-guided attention network. Scientific Reports, 14(1):11750.

de Barros, F., da Silva Fernandes, C. M., Kuhnen, B., Scarso Filho, J., Gonçalves, M., and da Costa Serra, M. (2021). Paranasal sinuses and human identification. Research, Society and Development, 10(9):e48710918161–e48710918161.

de Mendonça, D. S., Kurita, L. M., Carvalho, F. S. R., Tuji, F. M., Silva, P. G. d. B., Bezerra, T. P., de Aguiar, A. S. W., and Gurgel Costa, F. W. (2021). Development and validation of a new formula for sex estimation based on multislice computed tomographic measurements of maxillary and frontal sinuses among brazilian adults. Dentomaxillofacial Radiology, 50(6):20200490.

de Oliveira Gamba, T., Alves, M. C., and Haiter-Neto, F. (2016). Mandibular sexual dimorphism analysis in cbct scans. Journal of forensic and legal medicine, 38:106–110.

de Oliveira Gamba, T., Yamasaki, M. C., Groppo, F. C., da Silveira, H. L. D., de Almeida Boscolo, S. M., Sanderink, G. C., and Berkhout, W. E. R. (2017). Validation study of a new method for sexual prediction based on cbct analysis of maxillary sinus and mandibular canal. Archives of oral biology, 83:118–123.

Hamidi, O., Afrasiabi, M., and Namaki, M. (2024). Gadnn: a revolutionary hybrid deep learning neural network for age and sex determination utilizing cone beam computed tomography images of maxillary and frontal sinuses. BMC Medical Research Methodology, 24(1):50.

Jaitley, M., Phulambrikar, T., Kode, M., Gupta, A., and Singh, S. K. (2016). Foramen magnum as a tool for sexual dimorphism: A cone beam computed tomography study. Indian Journal of Dental Research, 27(5):458–462.

Jasim, H. H. and Al-Taei, J. A. (2013). Computed tomographic measurement of maxillary sinus volume and dimension in correlation to the age and gender (comparative study among individuals with dentate and edentulous maxilla). Journal of Baghdad College of Dentistry. Accessed: 2025-06-30.

Kim, D.-K., Cho, B.-J., Lee, M.-J., and Kim, J. H. (2021). Prediction of age and sex from paranasal sinus images using a deep learning network. Medicine, 100(7):e24756.

Kondou, H., Morohashi, R., Kimura, S., Idota, N., Matsunari, R., Ichioka, H., Bandou, R., Kawamoto, M., Ting, D., and Ikegaya, H. (2023). Artificial intelligence-based forensic sex determination of east asian cadavers from skull morphology. Scientific Reports, 13(1):21026.

Liu, X., Song, L., Liu, S., and Zhang, Y. (2021). A review of deep-learning-based medical image segmentation methods. Sustainability, 13(3):1224.

Martins Filho, I. E., Lopez-Capp, T. T., Biazevic, M. G. H., and Michel-Crosato, E. (2016). Sexual dimorphism using odontometric indexes: Analysis of three statistical techniques. Journal of Forensic and Legal Medicine, 44:37–42.

Moosa, S. S., Shaikh, M. H. R., Khwaja, M., Shaikh, S. A. H., Siddiqui, F. B., Daimi, S. R. H., Hiware, S. D., Ismail, E. E., and Begum, Y. (2021). Sexual dimorphic parameters of femur: a clinical guide in orthopedics and forensic studies. Journal of Medicine and Life, 14(6):762.

Nagare, S. P., Chaudhari, R. S., Birangane, R. S., and Parkarwar, P. C. (2018). Sex determination in forensic identification, a review. Journal of forensic dental sciences, 10(2):61–66.

Pereira, J. G. D., Santos, J. B. S., Sousa, S. P. d., Franco, A., and Silva, R. H. A. (2021). Frontal sinuses as tools for human identification: a systematic review of imaging methods. Dentomaxillofacial Radiology, 50(5):20200599.

Piraianu, A.-I., Fulga, A., Musat, C. L., Ciobotaru, O.-R., Poalelungi, D. G., Stamate, E., Ciobotaru, O., and Fulga, I. (2023). Enhancing the evidence with algorithms: how artificial intelligence is transforming forensic medicine. Diagnostics, 13(18):2992.

Senol, D., Secgin, Y., Duman, B. S., Toy, S., and Oner, Z. (2023). Sex and age estimation with machine learning algorithms with parameters obtained from cone beam computed tomography images of maxillary first molar and canine teeth. Egyptian Journal of Forensic Sciences, 13(1):27.

Sidhu, R., Chandra, S., Devi, P., Taneja, N., Sah, K., and Kaur, N. (2014). Forensic importance of maxillary sinus in gender determination: A morphometric analysis from western uttar pradesh, india. European Journal of General Dentistry, 3(01):53–56.

Spradley, M. K. (2016). Metric methods for the biological profile in forensic anthropology: sex, ancestry, and stature. Academic forensic pathology, 6(3):391–399.

Teke, H. Y., Duran, S., Canturk, N., and Canturk, G. (2007). Determination of gender by measuring the size of the maxillary sinuses in computerized tomography scans. Surgical and radiologic anatomy, 29:9–13.

Thurzo, A., Kosnáčová, H. S., Kurilová, V., Kosmel’, S., Beňuš, R., Moravanskỳ, N., Kováč, P., Kuracinová, K. M., Palkovič, M., and Varga, I. (2021). Use of advanced artificial intelligence in forensic medicine, forensic anthropology and clinical anatomy. In Healthcare, volume 9, page 1545. MDPI.

Tiwari, S., Jain, G., Shetty, D. K., Sudhi, M., Balakrishnan, J. M., and Bhatta, S. R. (2023). A comprehensive review on the application of 3d convolutional neural networks in medical imaging. Engineering Proceedings, 59(1):3.

Zheng, B., Zhong, Y., Al-Worafi, N. A., and Liu, Y. (2023). The dimensional and morphological assessment of the frontal sinus in sex estimation among different populations. Head & Face Medicine, 19(1):8.

Zorba, E., Moraitis, K., and Manolis, S. K. (2011). Sexual dimorphism in permanent teeth of modern greeks. Forensic science international, 210(1-3):74–81.