|Evaluation of facial attractiveness for patients with malocclusion A machine-learning technique employing Procrustes|
|Yu, Xiaonan1; Liu, Bin2; Pei, Yuru2; Xu, Tianmin1|
|关键词||Facial attractiveness Machine learning|
|收录类别||SCI ; SSCI|
|WOS标题词||Science & Technology|
|类目[WOS]||Dentistry, Oral Surgery & Medicine|
|研究领域[WOS]||Dentistry, Oral Surgery & Medicine|
|关键词[WOS]||COSMETIC DENTISTRY ; US ORTHODONTISTS ; GOLDEN RATIOS ; PHOTOGRAPHS ; PRINCIPLES ; CHINESE ; PROFILE ; SHAPE|
Objective: To establish an objective method for evaluating facial attractiveness from a set of orthodontic photographs.
Materials and Methods: One hundred eight malocclusion patients randomly selected from six universities in China were randomly divided into nine groups, with each group containing an equal number of patients with Class I, II, and Ill malocclusions. Sixty-nine expert Chinese orthodontists ranked photographs of the patients (frontal, lateral, and frontal smiling photos) before and after orthodontic treatment from "most attractive" to "least attractive" in each group. A weighted mean ranking was then calculated for each patient, based on which a three-point scale was created. Procrustes superimposition was conducted on 101 landmarks identified on the photographs. A support vector regression (SVR) function was set up according to the coordinate values of identified landmarks of each photographic set and its corresponding grading. Its predictive ability was tested for each group in turn.
Results: The average coincidence rate obtained for comparisons of the subjective ratings with the SVR evaluation was 71.8% according to 18 verification tests.
Conclusions: Geometric morphometrics combined with SVR may be a prospective method for objective comprehensive evaluation of facial attractiveness in the near future.
|资助机构||Specific Research Project of the Health Pro Bono Sector, Ministry of Health, China ; Nature Science Foundation of China|
|作者单位||1.Peking Univ, Sch & Hosp Stomatol, Dept Orthodont, Beijing 100081, Peoples R China|
2.Peking Univ, Key Lab Machine Percept, Beijing 100081, Peoples R China
|Yu, Xiaonan,Liu, Bin,Pei, Yuru,et al. Evaluation of facial attractiveness for patients with malocclusion A machine-learning technique employing Procrustes[J]. ANGLE ORTHODONTIST,2014,84(3):410-416.|
|APA||Yu, Xiaonan,Liu, Bin,Pei, Yuru,&Xu, Tianmin.(2014).Evaluation of facial attractiveness for patients with malocclusion A machine-learning technique employing Procrustes.ANGLE ORTHODONTIST,84(3),410-416.|
|MLA||Yu, Xiaonan,et al."Evaluation of facial attractiveness for patients with malocclusion A machine-learning technique employing Procrustes".ANGLE ORTHODONTIST 84.3(2014):410-416.|