{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T04:25:05Z","timestamp":1766377505249,"version":"3.41.2"},"reference-count":37,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2021,6,7]],"date-time":"2021-06-07T00:00:00Z","timestamp":1623024000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100014857","name":"National Natural Science Foundation of China-Guangdong Joint Fund","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100014857","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012151","name":"Sanming Project of Medicine in Shenzhen","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100012151","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Genet."],"abstract":"<jats:sec><jats:title>Background<\/jats:title><jats:p>Noonan syndrome (NS), a genetically heterogeneous disorder, presents with hypertelorism, ptosis, dysplastic pulmonary valve stenosis, hypertrophic cardiomyopathy, and small stature. Early detection and assessment of NS are crucial to formulating an individualized treatment protocol. However, the diagnostic rate of pediatricians and pediatric cardiologists is limited. To overcome this challenge, we propose an automated facial recognition model to identify NS using a novel deep convolutional neural network (DCNN) with a loss function called additive angular margin loss (ArcFace).<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>The proposed automated facial recognition models were trained on dataset that included 127 NS patients, 163 healthy children, and 130 children with several other dysmorphic syndromes. The photo dataset contained only one frontal face image from each participant. A novel DCNN framework with ArcFace loss function (DCNN-Arcface model) was constructed. Two traditional machine learning models and a DCNN model with cross-entropy loss function (DCNN-CE model) were also constructed. Transfer learning and data augmentation were applied in the training process. The identification performance of facial recognition models was assessed by five-fold cross-validation. Comparison of the DCNN-Arcface model to two traditional machine learning models, the DCNN-CE model, and six physicians were performed.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>At distinguishing NS patients from healthy children, the DCNN-Arcface model achieved an accuracy of 0.9201 \u00b1 0.0138 and an area under the receiver operator characteristic curve (AUC) of 0.9797 \u00b1 0.0055. At distinguishing NS patients from children with several other genetic syndromes, it achieved an accuracy of 0.8171 \u00b1 0.0074 and an AUC of 0.9274 \u00b1 0.0062. In both cases, the DCNN-Arcface model outperformed the two traditional machine learning models, the DCNN-CE model, and six physicians.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>This study shows that the proposed DCNN-Arcface model is a promising way to screen NS patients and can improve the NS diagnosis rate.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fgene.2021.669841","type":"journal-article","created":{"date-parts":[[2021,6,7]],"date-time":"2021-06-07T06:53:39Z","timestamp":1623048819000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":8,"title":["Automated Facial Recognition for Noonan Syndrome Using Novel Deep Convolutional Neural Network With Additive Angular Margin Loss"],"prefix":"10.3389","volume":"12","author":[{"given":"Hang","family":"Yang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xin-Rong","family":"Hu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ling","family":"Sun","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dian","family":"Hong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ying-Yi","family":"Zheng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ying","family":"Xin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hui","family":"Liu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Min-Yin","family":"Lin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Long","family":"Wen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dong-Po","family":"Liang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shu-Shui","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2021,6,7]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2570","DOI":"10.1002\/ajmg.a.37736","article-title":"Objective studies of the face of Noonan, Cardio-facio-cutaneous, and Costello syndromes: a comparison of three disorders of the Ras\/MAPK signaling pathway.","volume":"170","author":"Allanson","year":"2016","journal-title":"Am. J. Med. Genet. A"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1960","DOI":"10.1002\/ajmg.a.33518","article-title":"The face of Noonan syndrome: does phenotype predict genotype?","author":"Allanson","year":"2010","journal-title":"Am. J. Med. Genet. A"},{"key":"B3","doi-asserted-by":"publisher","first-page":"1415","DOI":"10.1109\/TCSII.2018.2865896","article-title":"A CNN accelerator on FPGA using depthwise separable convolution.","volume":"65","author":"Bai","year":"2018","journal-title":"IEEE. Trans. Circuits II."},{"key":"B4","doi-asserted-by":"publisher","first-page":"1082","DOI":"10.1038\/sj.ejhg.5201673","article-title":"Syndrome identification based on 2D analysis software.","volume":"14","author":"Boehringer","year":"2006","journal-title":"Eur. J. Hum. Genet."},{"key":"B5","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1007\/s10479-005-5724-z","article-title":"A tutorial on the cross-entropy method.","volume":"134","author":"Boer","year":"2005","journal-title":"Ann. Oper. Res."},{"key":"B6","doi-asserted-by":"publisher","first-page":"407","DOI":"10.18178\/ijmlc.2019.9.4.818","article-title":"Personal verification system using ID card and face photo.","volume":"9","author":"Chinapas","year":"2019","journal-title":"Int. J. Mach. Learn. Comput."},{"key":"B7","doi-asserted-by":"publisher","first-page":"4690","DOI":"10.1109\/CVPR.2019.00482","article-title":"ArcFace: Additive Angular Margin Loss for Deep Face Recognition","author":"Deng","year":"2019","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1895","DOI":"10.1162\/089976698300017197","article-title":"Approximate statistical tests for comparing supervised classification learning algorithms.","volume":"10","author":"Dietterich","year":"1998","journal-title":"Neural Comput."},{"key":"B9","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1038\/s41591-018-0279-0","article-title":"Identifying facial phenotypes of genetic disorders using deep learning.","volume":"25","author":"Gurovich","year":"2019","journal-title":"Nat. Med."},{"key":"B10","doi-asserted-by":"publisher","first-page":"212","DOI":"10.1111\/j.1601-6343.2009.01455.x","article-title":"Genetic studies of craniofacial anomalies: clinical implications and applications.","volume":"12","author":"Hart","year":"2009","journal-title":"Orthod. Craniofac. Res."},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep Residual Learning for Image Recognition","author":"He","year":"2016","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"B12","doi-asserted-by":"crossref","DOI":"10.1161\/CIRCGEN.119.002690","article-title":"LZTR1-related hypertrophic cardiomyopathy without typical noonan syndrome features.","volume":"13","author":"Jenkins","year":"2020","journal-title":"Circ. Genom. Precis. Med."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1867","DOI":"10.1109\/cvpr.2014.241","article-title":"One Millisecond Face Alignment with an Ensemble of Regression Trees","author":"Kazemi","year":"2014","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"B14","doi-asserted-by":"publisher","first-page":"2323","DOI":"10.1002\/ajmg.a.38362","article-title":"Noonan syndrome in diverse populations.","volume":"173","author":"Kruszka","year":"2017","journal-title":"Am. J. Med. Genet. A"},{"key":"B15","doi-asserted-by":"crossref","first-page":"105","DOI":"10.1016\/j.artmed.2014.08.003","article-title":"Biomedical visual data analysis to build an intelligent diagnostic decision support system in medical genetics.","volume":"62","author":"Kuru","year":"2014","journal-title":"Artif. Intell. Med."},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-20894-0","article-title":"Integrating functional analysis in the next-generation sequencing diagnostic pipeline of RASopathies.","volume":"8","author":"Leung","year":"2018","journal-title":"Sci. Rep."},{"key":"B17","doi-asserted-by":"publisher","first-page":"290","DOI":"10.1111\/cge.13588","article-title":"Molecular and phenotypic spectrum of Noonan syndrome in Chinese patients.","volume":"96","author":"Li","year":"2019","journal-title":"Clin. Genet."},{"key":"B18","doi-asserted-by":"publisher","first-page":"555","DOI":"10.1038\/sj.ejhg.5200997","article-title":"Computer-based recognition of dysmorphic faces.","volume":"11","author":"Loos","year":"2003","journal-title":"Eur. J. Hum. Genet."},{"key":"B19","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1002\/ajmg.1320210312","article-title":"Noonan syndrome: a review.","volume":"21","author":"Mendez","year":"1985","journal-title":"Am. J. Med. Genet."},{"key":"B20","doi-asserted-by":"publisher","first-page":"77","DOI":"10.1007\/s10462-018-09679-z","article-title":"Machine learning and deep learning frameworks and libraries for large-scale data mining: a survey.","volume":"52","author":"Nguyen","year":"2019","journal-title":"Artif. Intell. Rev."},{"key":"B21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICTKE.2017.8259629","article-title":"Artificial Intelligence, Machine Learning and Deep Learning","author":"Ongsulee","year":"2017","journal-title":"Proceedings of the 15th International Conference on ICT and Knowledge Engineering (ICT&amp;KE)"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1002\/mgg3.1636","article-title":"Objective differential diagnosis of Noonan and Williams-Beuren syndromes in diverse populations using quantitative facial phenotyping.","author":"Porras","year":"2021","journal-title":"Mol. Genet. Genomic. Med."},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics10070487","article-title":"Automatic identification of down syndrome using facial images with deep convolutional neural network.","volume":"10","author":"Qin","year":"2020","journal-title":"Diagnostics"},{"key":"B24","doi-asserted-by":"publisher","first-page":"333","DOI":"10.1016\/S0140-6736(12)61023-X","article-title":"Noonan syndrome.","volume":"381","author":"Roberts","year":"2013","journal-title":"Lancet"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1155\/2016\/3054578","article-title":"Exploring the underlying genetics of craniofacial morphology through various sources of knowledge.","volume":"2016","author":"Roosenboom","year":"2016","journal-title":"Biomed. Res. Int."},{"key":"B26","doi-asserted-by":"publisher","first-page":"127","DOI":"10.1023\/A:1010091220143","article-title":"The cross-entropy method for combinatorial and continuous optimization.","volume":"1","author":"Rubinstein","year":"1999","journal-title":"Methodol. Comput. Appl."},{"key":"B27","doi-asserted-by":"publisher","first-page":"676","DOI":"10.1210\/er.2017-00232","article-title":"The RASopathy family: consequences of germline activation of the RAS\/MAPK pathway.","volume":"39","author":"Tajan","year":"2018","journal-title":"Endocr. Rev."},{"key":"B28","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1016\/j.beem.2010.09.002","article-title":"Noonan syndrome and clinically related disorders.","volume":"25","author":"Tartaglia","year":"2011","journal-title":"Best Pract. Res. Clin. Endocrinol. Metab."},{"key":"B29","doi-asserted-by":"publisher","first-page":"718","DOI":"10.1002\/bdr2.1675","article-title":"Noonan syndrome on the African continent.","volume":"112","author":"Tekendo-Ngongang","year":"2020","journal-title":"Birth Defects Res."},{"key":"B30","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1002\/ajmg.1320530213","article-title":"Clinical and molecular studies in a large Dutch family with Noonan syndrome.","volume":"53","author":"Van Der Burgt","year":"1994","journal-title":"Am. J. Med. Genet."},{"key":"B31","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1016\/j.jmsy.2018.01.003","article-title":"Deep learning for smart manufacturing:methods and applications.","volume":"48","author":"Wang","year":"2018","journal-title":"J. Manuf. Syst."},{"key":"B32","doi-asserted-by":"publisher","first-page":"499","DOI":"10.1007\/978-3-319-46478-7_31","article-title":"A Discriminative Feature Learning Approach for Deep Face Recognition","author":"Wen","year":"2016","journal-title":"Proceedings of the Computer vision-European Conference on Computer Vision 2016"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1186\/s12920-017-0298-6","article-title":"Targeted\/exome sequencing identified mutations in ten Chinese patients diagnosed with Noonan syndrome and related disorders.","volume":"10","author":"Xu","year":"2017","journal-title":"BMC Med. Genomics"},{"key":"B34","doi-asserted-by":"publisher","first-page":"701","DOI":"10.1002\/sim1011","article-title":"A non-parametric method for the comparison of partial areas under ROC curves and its application to large health care data sets.","volume":"21","author":"Zhang","year":"2002","journal-title":"Statist. Med."},{"key":"B35","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.media","article-title":"Medical image classification using synergic deep learning.","volume":"54","author":"Zhang","year":"2019","journal-title":"Med. Image Anal."},{"key":"B36","doi-asserted-by":"publisher","first-page":"1499","DOI":"10.1109\/LSP.2016.2603342","article-title":"Joint face detection and alignment using multitask cascaded convolutional networks.","volume":"23","author":"Zhang","year":"2016","journal-title":"IEEE Signal Proc. Lett."},{"key":"B37","doi-asserted-by":"publisher","first-page":"29857","DOI":"10.1109\/ACCESS.2020.2972859","article-title":"Deep learning algorithms for bearing fault diagnostics\u2013a comprehensive review.","volume":"8","author":"Zhang","year":"2020","journal-title":"IEEE Access"}],"container-title":["Frontiers in Genetics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fgene.2021.669841\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,6,29]],"date-time":"2021-06-29T15:23:12Z","timestamp":1624980192000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fgene.2021.669841\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,7]]},"references-count":37,"alternative-id":["10.3389\/fgene.2021.669841"],"URL":"https:\/\/doi.org\/10.3389\/fgene.2021.669841","relation":{},"ISSN":["1664-8021"],"issn-type":[{"type":"electronic","value":"1664-8021"}],"subject":[],"published":{"date-parts":[[2021,6,7]]},"article-number":"669841"}}