Li et al The Journal of Craniofacial Surgery.Volume 33,Number 1,January/February 2022 realize high-precision segmentation of the orbit contour,and with 14.Kim J.Park SW.Choi J.et al.Ageing of the bony orbit is a major cause the CNN-based orbit contour sorting algorithm,the aging degree of of age-related intraorbital fat hemiation.J Plast Reconstr Aesthet Surg the bony orbit can be identified precisely.It is preliminarily 2018:71:658-664 validated that the aging mode of Mongolian bony orbit contour 15.Kahn D,Shawir R.Aging of the bony orbit:a three-dimensional is that the bone resorption of the superior orbital rim is more computed tomographic study.Aesthet Surg 2008:28:258-264 obvious than that of the inferior orbital rim,and the change of 16.Pessa JE,Chen Y.Curve analysis of the aging orbital aperture.Plast the orbit area,perimeter,height and circularity is not obvious in the Recon.str Surg2002:109:751-755 aging process.In our future work,we will collect more data. 17.Ching JA,Ford JM,Decker SJ.Aging of the adult bony orbit./ Craniofac Surg 2020:31:1082-1085 validate such data with the method put forward in this article, 18.Jeon A,Lee U,Kwak D,et al.Aging of the bony orbit in East Asians:a and quantize changes of aging features of orbit contours with three-dimensional computed tomographic study.Surg Radiol Anar automatic methods. 2020:42:617-626 19.Wang S,Kang B,Ma J,et al.A deep learning algorithm using CTimages REFERENCES to screen for Corona virus disease (COVID-19).Eur Radiol 2021:31:6096-6104 1.Lambros V.Observations on periorbital and midface aging.Plast 20.Tan X,Li K,Zhang J,et al.Automatic model for cervical cancer Reconstr Surg2007:120:1367-1376 screening based on convolutional neural network:a retrospective. 2.Varani J,Spearman D,Perone P.et al.Inhibition of type I procollagen multicohort,multicenter study.Cancer Cell Int 2021:21:35 synthesis by damaged collagen in photoaged skin and by collagenase- 21.Li L,Song X.Guo Y,et al.Deep convolutional neural networks for degraded collagen in vitro.Am J Pathol 2001:158:931-942 automatic detection of orbital blowout fractures.J Craniofac Surg 3.Gosain AK,Klein MH,Sudhakar PV,et al.A volumetric analysis of 2020:31:400-403 soft-tissue changes in the aging midface using high-resolution MRI: 22.Ronneberger O,Fischer P.Brox TU.U-Net:Convolutional Networks implications for facial rejuvenation.Plast Reconstr Surg for Biomedical Image Segmentation.International Conference on 2005:115:1143-1152 Medical Image Computing and Computer-Assisted Intervention 4.Farkas JP.Pessa JE.Hubbard B.et al.The science and theory behind 2015:234-241 facial aging.Plast Reconstr Surg Glob Open 2013:1:8-15 23.Byra M.Jarosik P,Szubert A,et al.Breast mass segmentation in 5.Matros E,Momoh A,Yaremchuk MJ.The aging midfacial skeleton: ultrasound with selective kernel U-Net convolutional neural network. implications for rejuvenation and reconstruction using implants.Facial Biomed Signal Process Control 2020:61:102027 Plast Surg2009:25:252-259 24.Long J,Ma G,Liu H,et al.Cascaded hybrid residual U-Net for glioma 6.Shaw JRB,Katzel EB,Koltz PF,et al.Facial bone density:effects of segmentation.Multimed Tools Appl 2020:79:24929-24947 aging and impact on facial rejuvenation.Aesther Surg/2012:32:937- 25.Long F.Microscopy cell nuclei segmentation with enhanced U-Net. 942 BMC Bioinform 2020:21 7.Wong C.Mendelson B.Newer understanding of specific anatomic 26.Pessa JE.The potential role of stereolithography in the study of facial targets in the aging face as applied to injectables:aging changes in the aging.Am J Orthod Dentofacial Orthop 2001;119:117-120 craniofacial skeleton and facial ligaments.Plast Reconstr Surg 27.Krizhevsky A.Sutskever I.Hinton GE.ImageNet classification with 2015:136:44S48S deep convolutional neural networks.Ady Neural Inf Process Syst 8.Pessa JE.An algorithm of facial aging:verification of Lambros's theory 2012:25:1097-1105 by three-dimensional stereolithography,with reference to the 28.Srivastava N.Hinton G,Krizhevsky A,et al.Dropout:a simple way to pathogenesis of midfacial aging,scleral show,and the lateral suborbital prevent neural networks from overfitting.J Mach Learn Res trough deformity.Plast Reconstr Surg 2000:106:479-488 2014:15:1929-1958 9.Mendelson BC.Hartley W.Scott M,et al.Age-related changes of the 29.Belongie S.Malik J,Puzicha J.Shape matching and object recognition orbit and midcheek and the implications for facial rejuvenation.Aesther using shape contexts.IEEE Trans Pattern Anal Mach Intell Plast Surg2007:31:419423 2002:24:509-522 10.Shaw RB.Kahn DM.Aging of the midface bony elements:a three- 30.Greg M,Serge B,Jitendra M.Efficient shape matching dimensional computed tomographic study.Plast Reconstr Surg using shape contexts.IEEE Trans Pattern Anal Mach Intell 2007:119:675-681 2005:27:1832-1837 11.Jeon A,Sung KH,Kim SD,et al.Anatomical changes in the East Asian 31.Kingma DP,Ba J.Adam:A method for stochastic optimization.3rd midface skeleton with aging.Folia Morphol (Warsz)2017:76:730-735 International Conference on Leaming Representations 2015:1-13 12.Escaravage GK,Dutton JJ.Age-related changes in the pediatric human 32.Cover T,Hart P.Nearest neighbor patter classification.IEEE Trans orbit on CT.Ophthalmic Plast Reconstr Surg 2013:29:150-156 Inform Theory 1967:13:21-27 13.Toledo AL.Cardoso MA.Santos BL,et al.Aging and sexual differences 33.Fushiki T.Estimation of prediction error by using K-fold cross- of the human skull.Plast Reconstr Surg Glob Open 2017:5:e1297 validation.Stat Comput 2011:21:137-146 318 Copyright o 2021 The Author(s).Published by Wolters Kluwer Health,Inc.on behalf of Mutaz B.Habal,MDrealize high-precision segmentation of the orbit contour, and with the CNN-based orbit contour sorting algorithm, the aging degree of the bony orbit can be identified precisely. It is preliminarily validated that the aging mode of Mongolian bony orbit contour is that the bone resorption of the superior orbital rim is more obvious than that of the inferior orbital rim, and the change of the orbit area, perimeter, height and circularity is not obvious in the aging process. In our future work, we will collect more data, validate such data with the method put forward in this article, and quantize changes of aging features of orbit contours with automatic methods. REFERENCES 1. Lambros V. Observations on periorbital and midface aging. Plast Reconstr Surg 2007;120:1367–1376 2. Varani J, Spearman D, Perone P, et al. Inhibition of type I procollagen synthesis by damaged collagen in photoaged skin and by collagenase￾degraded collagen in vitro. Am J Pathol 2001;158:931–942 3. Gosain AK, Klein MH, Sudhakar PV, et al. A volumetric analysis of soft-tissue changes in the aging midface using high-resolution MRI: implications for facial rejuvenation. Plast Reconstr Surg 2005;115:1143–1152 4. Farkas JP, Pessa JE, Hubbard B, et al. The science and theory behind facial aging. Plast Reconstr Surg Glob Open 2013;1:8–15 5. Matros E, Momoh A, Yaremchuk MJ. The aging midfacial skeleton: implications for rejuvenation and reconstruction using implants. Facial Plast Surg 2009;25:252–259 6. Shaw JRB, Katzel EB, Koltz PF, et al. Facial bone density: effects of aging and impact on facial rejuvenation. Aesthet Surg J 2012;32:937– 942 7. Wong C, Mendelson B. Newer understanding of specific anatomic targets in the aging face as applied to injectables: aging changes in the craniofacial skeleton and facial ligaments. Plast Reconstr Surg 2015;136:44S–48S 8. Pessa JE. An algorithm of facial aging: verification of Lambros’s theory by three-dimensional stereolithography, with reference to the pathogenesis of midfacial aging, scleral show, and the lateral suborbital trough deformity. Plast Reconstr Surg 2000;106:479–488 9. Mendelson BC, Hartley W, Scott M, et al. Age-related changes of the orbit and midcheek and the implications for facial rejuvenation. Aesthet Plast Surg 2007;31:419–423 10. Shaw RB, Kahn DM. Aging of the midface bony elements: a three￾dimensional computed tomographic study. Plast Reconstr Surg 2007;119:675–681 11. Jeon A, Sung KH, Kim SD, et al. Anatomical changes in the East Asian midface skeleton with aging. Folia Morphol (Warsz) 2017;76:730–735 12. Escaravage GK, Dutton JJ. Age-related changes in the pediatric human orbit on CT. Ophthalmic Plast Reconstr Surg 2013;29:150–156 13. Toledo AL, Cardoso MA, Santos BL, et al. Aging and sexual differences of the human skull. Plast Reconstr Surg Glob Open 2017;5:e1297 14. Kim J, Park SW, Choi J, et al. Ageing of the bony orbit is a major cause of age-related intraorbital fat herniation. J Plast Reconstr Aesthet Surg 2018;71:658–664 15. Kahn D, Shawjr R. Aging of the bony orbit: a three-dimensional computed tomographic study. Aesthet Surg J 2008;28:258–264 16. Pessa JE, Chen Y. Curve analysis of the aging orbital aperture. Plast Reconstr Surg 2002;109:751–755 17. Ching JA, Ford JM, Decker SJ. Aging of the adult bony orbit. J Craniofac Surg 2020;31:1082–1085 18. Jeon A, Lee U, Kwak D, et al. Aging of the bony orbit in East Asians: a three-dimensional computed tomographic study. Surg Radiol Anat 2020;42:617–626 19. Wang S, Kang B, Ma J, et al. A deep learning algorithm using CT images to screen for Corona virus disease (COVID-19). Eur Radiol 2021;31:6096–6104 20. Tan X, Li K, Zhang J, et al. Automatic model for cervical cancer screening based on convolutional neural network: a retrospective, multicohort, multicenter study. Cancer Cell Int 2021;21:35 21. Li L, Song X, Guo Y, et al. Deep convolutional neural networks for automatic detection of orbital blowout fractures. J Craniofac Surg 2020;31:400–403 22. Ronneberger O, Fischer P, Brox TU. U-Net: Convolutional Networks for Biomedical Image Segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention 2015:234–241 23. Byra M, Jarosik P, Szubert A, et al. Breast mass segmentation in ultrasound with selective kernel U-Net convolutional neural network. Biomed Signal Process Control 2020;61:102027 24. Long J, Ma G, Liu H, et al. Cascaded hybrid residual U-Net for glioma segmentation. Multimed Tools Appl 2020;79:24929–24947 25. Long F. Microscopy cell nuclei segmentation with enhanced U-Net. BMC Bioinform 2020;21 26. Pessa JE. The potential role of stereolithography in the study of facial aging. Am J Orthod Dentofacial Orthop 2001;119:117–120 27. Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 2012;25:1097–1105 28. Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 2014;15:1929–1958 29. Belongie S, Malik J, Puzicha J. Shape matching and object recognition using shape contexts. IEEE Trans Pattern Anal Mach Intell 2002;24:509–522 30. Greg M, Serge B, Jitendra M. Efficient shape matching using shape contexts. IEEE Trans Pattern Anal Mach Intell 2005;27:1832–1837 31. Kingma DP, Ba J. Adam: A method for stochastic optimization. 3rd International Conference on Learning Representations 2015;1–13 32. Cover T, Hart P. Nearest neighbor pattern classification. IEEE Trans Inform Theory 1967;13:21–27 33. Fushiki T. Estimation of prediction error by using K-fold cross￾validation. Stat Comput 2011;21:137–146 Li et al The Journal of Craniofacial Surgery Volume 33, Number 1, January/February 2022 318 Copyright # 2021 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of Mutaz B. Habal, MD