Sex classification of 3D skull images using deep neural networks.
3D medical imaging
Deep learning
Plastic surgery
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
14 Jun 2024
14 Jun 2024
Historique:
received:
08
04
2023
accepted:
10
05
2024
medline:
15
6
2024
pubmed:
15
6
2024
entrez:
14
6
2024
Statut:
epublish
Résumé
Determining the fundamental characteristics that define a face as "feminine" or "masculine" has long fascinated anatomists and plastic surgeons, particularly those involved in aesthetic and gender-affirming surgery. Previous studies in this area have relied on manual measurements, comparative anatomy, and heuristic landmark-based feature extraction. In this study, we collected retrospectively at Cedars Sinai Medical Center (CSMC) a dataset of 98 skull samples, which is the first dataset of this kind of 3D medical imaging. We then evaluated the accuracy of multiple deep learning neural network architectures on sex classification with this dataset. Specifically, we evaluated methods representing three different 3D data modeling approaches: Resnet3D, PointNet++, and MeshNet. Despite the limited number of imaging samples, our testing results show that all three approaches achieve AUC scores above 0.9 after convergence. PointNet++ exhibits the highest accuracy, while MeshNet has the lowest. Our findings suggest that accuracy is not solely dependent on the sparsity of data representation but also on the architecture design, with MeshNet's lower accuracy likely due to the lack of a hierarchical structure for progressive data abstraction. Furthermore, we studied a problem related to sex determination, which is the analysis of the various morphological features that affect sex classification. We proposed and developed a new method based on morphological gradients to visualize features that influence model decision making. The method based on morphological gradients is an alternative to the standard saliency map, and the new method provides better visualization of feature importance. Our study is the first to develop and evaluate deep learning models for analyzing 3D facial skull images to identify imaging feature differences between individuals assigned male or female at birth. These findings may be useful for planning and evaluating craniofacial surgery, particularly gender-affirming procedures, such as facial feminization surgery.
Identifiants
pubmed: 38877045
doi: 10.1038/s41598-024-61879-6
pii: 10.1038/s41598-024-61879-6
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
13707Informations de copyright
© 2024. The Author(s).
Références
Secord, P. F. & Muthard, J. E. Individual differences in the perception of women’s faces. J. Abnorm. Psychol. 50(2), 238–242. https://doi.org/10.1037/h0044014 (1955).
doi: 10.1037/h0044014
pubmed: 14366888
Li, Z., Lei, X., Yan, X., Hu, Z. & Liu, H. Attractiveness evaluation and identity of self-face: The effect of sexual dimorphism. Iperception. 12(6), 20416695211058800. https://doi.org/10.1177/20416695211058799 (2021).
doi: 10.1177/20416695211058799
pubmed: 34881018
pmcid: 8646797
Rodman, R. Cosmetic bone contouring. Curr. Opin. Otolaryngol. Head Neck Surg. 25(4), 337–340. https://doi.org/10.1097/MOO.0000000000000370 (2017).
doi: 10.1097/MOO.0000000000000370
pubmed: 28504986
Cekdemir, Y. E., Mutlu, U., Karaman, G. & Balci, A. Estimation of sex using morphometric measurements performed on cranial computerized tomography scans. Radiol. Med. 126(2), 306–315. https://doi.org/10.1007/s11547-020-01233-8 (2021) (Epub 2020 Jun 12).
doi: 10.1007/s11547-020-01233-8
pubmed: 32533549
Morrison, S. D. et al. Facial feminization: Systematic review of the literature. Plast. Reconstr. Surg. 137(6), 1759–1770. https://doi.org/10.1097/PRS.0000000000002171 (2016).
doi: 10.1097/PRS.0000000000002171
pubmed: 27219232
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770–78 (2016).
Qi, C. R., Yi, L., Su, H. & Guibas, L. J. PointNet++: Deep hierarchical feature learning on point sets in a metric space. (2017.) arXiv [cs.CV]. arXiv. https://proceedings.neurips.cc/paper/2017/hash/d8bf84be3800d12f74d8b05e9b89836f-Abstract.html .
Feng, Y., Feng, Y., You, H., Zhao, X. & Gao, Y. MeshNet: Mesh neural network for 3D shape representation. In Proceedings of the AAAI Conference on Artificial Intelligence Vol. 33, no. 01, 8279–8286 (2019).
Kadir, T. & Brady, M. Saliency, scale and image description. Int. J. Comput. Vis. 45(2), 83–105 (2001).
doi: 10.1023/A:1012460413855
Kleisner, K. et al. How and why patterns of sexual dimorphism in human faces vary across the world. Sci. Rep. 11, 5978. https://doi.org/10.1038/s41598-021-85402-3 (2021).
doi: 10.1038/s41598-021-85402-3
pubmed: 33727579
pmcid: 7966798
Wen, Y. F., Wong, H. M., Lin, R., Yin, G. & McGrath, C. Inter-ethnic/racial facial variations: A systematic review and Bayesian meta-analysis of photogrammetric studies. PLoS ONE 10(8), e0134525. https://doi.org/10.1371/journal.pone.0134525 (2015).
doi: 10.1371/journal.pone.0134525
pubmed: 26247212
pmcid: 4527668
Thayer, Z. M. & Dobson, S. D. Geographic variation in chin shape challenges the universal facial attractiveness hypothesis. PLoS ONE 8(4), e60681. https://doi.org/10.1371/journal.pone.0060681 (2013) (Epub 2013 Apr 3).
doi: 10.1371/journal.pone.0060681
pubmed: 23560102
pmcid: 3616164
Butovskaya, M. L., Rostovtseva, V. V. & Mezentseva, A. A. Facial and body sexual dimorphism are not interconnected in the Maasai. J. Physiol. Anthropol. 41, 3. https://doi.org/10.1186/s40101-021-00276-8 (2022).
doi: 10.1186/s40101-021-00276-8
pubmed: 34996526
pmcid: 8740871
Mydlová, M., Dupej, J., Koudelová, J. & Velemínská, J. Sexual dimorphism of facial appearance in ageing human adults: A cross-sectional study. Forensic Sci. Int. 257, 519.e1-519.e9. https://doi.org/10.1016/j.forsciint.2015.09.008 (2015) (Epub 2015 Oct 22).
doi: 10.1016/j.forsciint.2015.09.008
pubmed: 26548377
Ferrario, V. F., Sforza, C., Pizzini, G., Vogel, G. & Miani, A. Sexual dimorphism in the human face assessed by Euclidean distance matrix analysis. J. Anat. 183(3), 593–600 (1993).
pubmed: 8300436
pmcid: 1259885
Nakamura, K. & Watanabe, K. A new data-driven mathematical model dissociates attractiveness from sexual dimorphism of human faces. Sci. Rep. 10(1), 16588. https://doi.org/10.1038/s41598-020-73472-8 (2020).
doi: 10.1038/s41598-020-73472-8
pubmed: 33024137
pmcid: 7538911
Shah, S. M. & Joshi, M. R. An assessment of asymmetry in the normal craniofacial complex. Angle Orthod. 48, 141–148 (1978).
pubmed: 277077
Vig, P. S. & Hewitt, A. B. Asymmetry of the human facial skeleton. Angle Orthod. 45, 125–129 (1975).
pubmed: 1054939
Woo, T. L. On the asymmetry of the human skull. Biometrika 22, 324–352 (1931).
doi: 10.1093/biomet/22.3-4.324
Harnádková, K., Kočandrlová, K., Kožejová Jaklová, L., Dupej, J. & Velemínská, J. The effect of sex and age on facial shape directional asymmetry in adults: A 3D landmarks-based method study. PLoS ONE. 18(8), e0288702. https://doi.org/10.1371/journal.pone.0288702 (2023).
doi: 10.1371/journal.pone.0288702
pubmed: 37535545
pmcid: 10399867
Claes, P. et al. Sexual dimorphism in multiple aspects of 3D facial symmetry and asymmetry defined by spatially dense geometric morphometrics. J Anat. 221(2), 97–114. https://doi.org/10.1111/j.1469-7580.2012.01528.x (2012) (Epub 2012 Jun 18).
doi: 10.1111/j.1469-7580.2012.01528.x
pubmed: 22702244
pmcid: 3406357
Kimbel, W. H., White, T. D. & Johanson, D. C. Cranial morphology of Australopithecus afarensis: A comparative study based on a composite reconstruction of the adult skull. Am. J. Phys. Anthropol. 64(4), 337–388. https://doi.org/10.1002/ajpa.1330640403 (1984).
doi: 10.1002/ajpa.1330640403
pubmed: 6435455
Ceballos, J. L. & Rentschler, E. H. Roentgen diagnosis of sex based on adult skull characteristics; comparison study of cephalometry of male and female skull films (frontal pro-projection). Radiology. 70(1), 55–61. https://doi.org/10.1148/70.1.55 (1958).
doi: 10.1148/70.1.55
pubmed: 13506060
Lamy, M., La Rosa, S., May, L. & Broome, M. Surgical margins in 3D planned mandibular resections for squamous cell carcinomas of the oral cavity. J. Craniofac. Surg. https://doi.org/10.1097/SCS.0000000000009068 (2022) (Epub ahead of print).
doi: 10.1097/SCS.0000000000009068
pubmed: 36730970
Toma, A. M. et al. A three-dimensional look for facial differences between males and females in a British-Caucasian sample aged 151/2 years old. Orthod. Craniofac. Res 11(3), 180–185 (2008).
doi: 10.1111/j.1601-6343.2008.00428.x
pubmed: 18713155
Kočnar, T., Saribay, S. A. & Kleisner, K. Perceived attractiveness of Czech faces across 10 cultures: Associations with sexual shape dimorphism, averageness, fluctuating asymmetry, and eye color. PLoS ONE. 14(11), e0225549. https://doi.org/10.1371/journal.pone.0225549 (2019).
doi: 10.1371/journal.pone.0225549
pubmed: 31751432
pmcid: 6872208
Bannister, J. J. et al. Sex differences in adult facial three-dimensional morphology: Application to gender-affirming facial surgery. Facial Plastic Surg. Aesthet. Med. 24(S2), S24-30 (2022).
doi: 10.1089/fpsam.2021.0301
Li, J. et al. MedShapeNet—A Large-Scale Dataset of 3D Medical Shapes for Computer Vision. arXiv preprint arXiv:2308.16139 (2023).
Kodym, O., Španěl, M. & Herout, A. Skull shape reconstruction using cascaded convolutional networks. Comput. Biol. Med. 123, 103886. https://doi.org/10.1016/j.compbiomed.2020.103886 (2020) (Epub 2020 Jun 27).
doi: 10.1016/j.compbiomed.2020.103886
pubmed: 32658793
Xiong, Y.-T. et al. Virtual reconstruction of midfacial bone defect based on generative adversarial network. Head Face Med. 18(1), 19 (2022).
doi: 10.1186/s13005-022-00325-2
pubmed: 35761334
pmcid: 9235085
Li, J. et al. AutoImplant 2020-first MICCAI challenge on automatic cranial implant design. IEEE Trans. Med. Imaging 40(9), 2329–2342 (2021).
doi: 10.1109/TMI.2021.3077047
pubmed: 33939608
Li, J. et al. Synthetic skull bone defects for automatic patient-specific craniofacial implant design. Sci. Data 8(1), 36 (2021).
doi: 10.1038/s41597-021-00806-0
pubmed: 33514740
pmcid: 7846796
Semper-Hogg, W. et al. Virtual reconstruction of midface defects using statistical shape models. J. Cranio-Maxillo-Facial Surg. 45(4), 461–466 (2017).
doi: 10.1016/j.jcms.2016.12.020
Fuessinger, M. A. et al. Virtual reconstruction of bilateral midfacial defects by using statistical shape modeling. J. Cranio-Maxillo-Facial Surg. 47(7), 1054–1059 (2019).
doi: 10.1016/j.jcms.2019.03.027