Astley S J, Clarren S K
Department of Epidemiology, School of Public Health and Community Medicine, University of Washington, Seattle, USA.
J Pediatr. 1996 Jul;129(1):33-41. doi: 10.1016/s0022-3476(96)70187-7.
The purpose of this study was to demonstrate that a quantitative, multivariate case definition of the fetal alcohol syndrome (FAS) facial phenotype could be derived from photographs of individuals with FAS and to demonstrate how this case definition and photographic approach could be used to develop efficient, accurate, and precise screening tools, diagnostic aids, and possibly surveillance tools.
Frontal facial photographs of 42 subjects (from birth to 27 years of age) with FAS were matched to 84 subjects without FAS. The study population was randomly divided in half. Group 1 was used to identify the facial features that best differentiated individuals with and without FAS. Group 2 was used for cross validation.
In group 1, stepwise discriminant analysis identified three facial features (reduced palpebral fissure length/inner canthal distance ratio, smooth philtrum, and thin upper lip) as the cluster of features that differentiated individuals with and without FAS in groups 1 and 2 with 100% accuracy. Sensitivity and specificity were unaffected by race, gender, and age.
The phenotypic case definition derived from photographs accurately distinguished between individuals with and without FAS, demonstrating the potential of this approach for developing screening, diagnostic, and surveillance tools. Further evaluation of the validity and generalizability of this method will be needed.
本研究的目的是证明可以从胎儿酒精综合征（FAS）患者的照片中得出FAS面部表型的定量多变量病例定义，并证明如何使用该病例定义和摄影方法来开发高效、准确和精确的筛查工具、诊断辅助工具以及可能的监测工具。
将42名患有FAS的受试者（年龄从出生到27岁）的正面面部照片与84名无FAS的受试者的照片进行匹配。研究人群被随机分成两半。第1组用于确定最能区分有FAS和无FAS个体的面部特征。第2组用于交叉验证。
在第1组中，逐步判别分析确定了三个面部特征（睑裂长度/内眦距离比值降低、人中平滑、上唇薄）作为在第1组和第2组中区分有FAS和无FAS个体的特征簇，准确率达100%。敏感性和特异性不受种族、性别和年龄的影响。
从照片得出的表型病例定义能够准确区分有FAS和无FAS的个体，证明了这种方法在开发筛查、诊断和监测工具方面的潜力。需要对该方法的有效性和可推广性进行进一步评估。
