Automated syndrome diagnosis by three-dimensional facial imaging.

PURPOSE: Deep phenotyping is an emerging trend in precision medicine for genetic disease. The shape of the face is affected in 30-40% of known genetic syndromes. Here, we determine whether syndromes can be diagnosed from 3D images of human faces. METHODS: We analyzed variation in three-dimensional (3D) facial images of 7057 subjects: 3327 with 396 different syndromes, 727 of their relatives, and 3003 unrelated, unaffected subjects. We developed and tested machine learning and parametric approaches to automated syndrome diagnosis using 3D facial images. RESULTS: Unrelated, unaffected subjects were correctly classified with 96% accuracy. Considering both syndromic and unrelated, unaffected subjects together, balanced accuracy was 73% and mean sensitivity 49%. Excluding unrelated, unaffected subjects substantially improved both balanced accuracy (78.1%) and sensitivity (56.9%) of syndrome diagnosis. The best predictors of classification accuracy were phenotypic severity and facial distinctiveness of syndromes. Surprisingly, unaffected relatives of syndromic subjects were frequently classified as syndromic, often to the syndrome of their affected relative. CONCLUSION: Deep phenotyping by quantitative 3D facial imaging has considerable potential to facilitate syndrome diagnosis. Furthermore, 3D facial imaging of "unaffected" relatives may identify unrecognized cases or may reveal novel examples of semidominant inheritance.

Clinical and genetic heterogeneity in Dubowitz syndrome: Implications for diagnosis, management and further research.

Dubowitz syndrome was described in 1965 as a recognizable syndrome characterized by microcephaly, short stature, eczema, mild developmental delays, and an increased risk of malignancy. Since its original description, there have been over 200 reported cases though no single gene has been identified to explain a significant proportion of affected individuals. Since the last definitive review of Dubowitz syndrome in 1996, there have been 63 individuals with a clinical, or suspected, diagnosis of Dubowitz syndrome reported in 51 publications. These individuals show a markedly wide spectrum with respect to growth, facial gestalt, psychomotor development, and risk of malignancy; genetic causes were identified in 33% (21/63). Seven individuals had deleterious copy number variants, in particular deletions at 14q32 and 17q24 were reported and showed overlap with the Dubowitz phenotype. Several cases were shown to have single gene disorders that included de novo or biallelic pathogenic variants in several genes including NSUN2 and LIG4 frequently identified by next-generation sequencing methods. It appears that the inability to identify a single gene responsible for Dubowitz syndrome reflects its extreme clinical and genetic heterogeneity. However, detailed phenotyping combined with careful grouping of subsets of unsolved cases and in conjunction with data-sharing will identify novel disease genes responsible for additional cases. In the interim, for those clinically diagnosed with a Dubowitz phenotype, we recommend assessment by a Medical Geneticist, a microarray and, if available, clinical or research based genome-wide sequencing. Management suggestions, including decisions regarding malignancy screening in select patients will be discussed.