Abstract
Purpose
Automated facial recognition technology based on deep learning has achieved high accuracy in diagnosing various endocrine diseases and genetic syndromes. This study attempts to establish a facial diagnostic system for Turner syndrome (TS) based on deep convolutional neural networks.
Methods
Photographs of 207 TS patients and 1074 female controls were collected from July 2016 to April 2019. Finally, 170 patients diagnosed with TS and 1053 female controls were included. Deep convolutional neural networks were used to develop the facial diagnostic system. A prospective study, which included two TS patients and 35 controls, was conducted to test the efficacy in the real clinical setting.
Results
The average areas under the curve (AUCs) in three different scenarios were 0.9540 ± 0.0223, 0.9662 ± 0.0108 and 0.9557 ± 0.0119, separately. The average sensitivity and specificity of the prospective study were 96.7% and 97.0%, respectively.
Conclusions
The facial diagnostic system achieved high accuracy. Prospective study results demonstrated the application value of this system, which is promising in the screening of Turner syndrome.
Similar content being viewed by others
Data availability
The statistical data are available on request. The photographs are not available due to the privacy of participants.
References
J.E. Pinsker, Turner syndrome: updating the paradigm of clinical care. J. Clin. Endocrinol. Metab. 97(6), E994–E1003 (2012). https://doi.org/10.1210/jc.2012-1245
L.L. Levitsky, A.H. Luria, F.J. Hayes, A.E. Lin, Turner syndrome: update on biology and management across the life span. Curr. Opin. Endocrinol. Diabetes. Obes. 22(1), 65–72 (2015). https://doi.org/10.1097/MED.0000000000000128
M. Ye, J.Y. Yeh, I. Kosteria, L. Li, Progress in Fertility Preservation Strategies in Turner Syndrome. Front. Med. 7, 3 (2020). https://doi.org/10.3389/fmed.2020.00003
A. Linglart, S. Cabrol, P. Berlier, C. Stuckens, K. Wagner, M. de Kerdanet, C. Limoni, J.C. Carel, J.L. Chaussain, Growth hormone treatment before the age of 4 years prevents short stature in young girls with Turner syndrome. Eur. J. Endocrinol. 164(6), 891–897 (2011). https://doi.org/10.1530/eje-10-1048
G. Massa, F. Verlinde, J. De Schepper, M. Thomas, J.P. Bourguignon, M. Craen, F. de Zegher, I. Francois, M. Du Caju, M. Maes, C. Heinrichs, Trends in age at diagnosis of Turner syndrome. Arch. Dis. Child. 90(3), 267–268 (2005). https://doi.org/10.1136/adc.2004.049817. Belgian Study Group for Paediatric, E.
J. Frelich, T. Irzyniec, K. Lepska, W. Jeż. New insight into clinical features, karyotypes and age at diagnosis in women with Turner Syndrome. ENDOKRYNOLOGIA POLSKA (2019). https://doi.org/10.5603/EP.a2019.0016
K. Stochholm, S. Juul, K. Juel, R.W. Naeraa, C.H. Gravholt, Prevalence, Incidence, Diagnostic Delay, and Mortality in Turner Syndrome. J. Clin. Endocrinol. Metab. 91(10), 3897–3902 (2006). https://doi.org/10.1210/jc.2006-0558
C.G. Palmer, A. Reichmann, Chromosomal and Clinical Findings in 110 Females with Turner Syndrome. Human Genet. 35, 35–49 (1976)
L. Sävendahl, M.L. Davenport, Delayed diagnoses of Turner’s syndrome: proposed guidelines for change. J. Pediatr. 137(4), 455–459 (2000). https://doi.org/10.1067/mpd.2000.107390
J. Miguel-Neto, A.B. Carvalho, A.P. Marques-de-Faria, G. Guerra-Júnior, A.T. Maciel-Guerra, New approach to phenotypic variability and karyotype-phenotype correlation in Turner syndrome. J. Pediatr. Endocrinol. Metab. 29(4), 475–479 (2016). https://doi.org/10.1515/jpem-2015-0346
D. Liu, W.M. Hsu, Oriental eyelids. Anatomic difference and surgical consideration. Ophthal. Plastic Reconstruct. Surg. 2(2), 59–64 (1986)
S. Zhang, H. Xue, Upper arch flap combined with extended incision of lower eyelid: a modified epicanthoplasty in correcting epicanthus. Aesthetic Plast. Surg. 42(4), 1033–1038 (2018). https://doi.org/10.1007/s00266-018-1175-1
Y. Gurovich, Y. Hanani, O. Bar, G. Nadav, N. Fleischer, D. Gelbman, L. Basel-Salmon, P.M. Krawitz, S.B. Kamphausen, M. Zenker, L.M. Bird, K.W. Gripp, Identifying facial phenotypes of genetic disorders using deep learning. Nature Med. 25(1), 60–64 (2019). https://doi.org/10.1038/s41591-018-0279-0
P. Shukla, T. Gupta, A. Saini, P. Singh, R. Balasubramanian. A Deep Learning Frame-Work for Recognizing Developmental Disorders. 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), 705-714 (2017). 10.1109/wacv.2017.84
S. Hadj-Rabia, H. Schneider, E. Navarro, O. Klein, N. Kirby, K. Huttner, L. Wolf, M. Orin, S. Wohlfart, C. Bodemer, D.K. Grange, Automatic recognition of the XLHED phenotype from facial images. Am. J. Med. Genet. Part A 173(9), 2408–2414 (2017). https://doi.org/10.1002/ajmg.a.38343
S. Chen, Z.-x Pan, H.-j Zhu, Q. Wang, J.-J. Yang, Y. Lei, J.-q Li, H. Pan, Development of a computer-aided tool for the pattern recognition of facial features in diagnosing Turner syndrome: comparison of diagnostic accuracy with clinical workers. Scientific Rep. 8(1), 9317 (2018). https://doi.org/10.1038/s41598-018-27586-9
K. He, X. Zhang, S. Ren, J. Sun. Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016). 10.1109/CVPR.2016.90
W. Song, Y. Lei, S. Chen, Z. Pan, J.-J. Yang, H. Pan, X. Du, W. Cai, Q. Wang, Multiple facial image features-based recognition for the automatic diagnosis of turner syndrome. Comput. Industry 100, 85–95 (2018). https://doi.org/10.1016/j.compind.2018.03.021
Funding
This work was supported by the Beijing Municipal Natural Science Foundation (Grant No. 7192153), the National Natural Science Foundation of China (Grant No. 81673184), the CAMS Initiative for Innovative Medicine (2016-I2M-1–008), the National Natural Science Foundation of China (Grants Nos. 61773382, U190920015 & 61773381), and the CAS Key Technology Talent Program (Z.S.).
Author information
Authors and Affiliations
Contributions
The first draft of the paper was written by Z.P., and all authors commented on previous versions of the paper. Material preparation and data collection were performed by X.L., Z.P., S.L., and S.W. Y.B., Z.S., L.N., X.D., and X.S. were responsible for the establishment of the automatic classification system. Y.B. and L.N. wrote and run the process of face classification using a computer. Statistical analyses were performed by Z.P. S.C., H.P., G.X. and H.Z. designed the study and provided guidance. All authors contributed to the study conception and design. All authors read and approved the final paper.
Corresponding authors
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The ethical approval for the present study was obtained from the Institutional Review Board of Chinese Academy of Medical Sciences, PUMCH (Project No: ZS-1242). The study procedures were performed in accordance with the approved guidelines.
Informed consent
All patients provided written informed consent. In the case of minors, consent is provided by parents or legal guardians.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
About this article
Cite this article
Pan, Z., Shen, Z., Zhu, H. et al. Clinical application of an automatic facial recognition system based on deep learning for diagnosis of Turner syndrome. Endocrine 72, 865–873 (2021). https://doi.org/10.1007/s12020-020-02539-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12020-020-02539-3