Facial scans could aid diagnosis of rare genetic disorders

11 September 2007

A UK scientist has developed a computer-based programme that compares 3D facial scans of a patient with those of specific genetic disorders, with a view to speeding up diagnosis. Many genetic syndromes are associated with dysmorphology, including characteristic facial features, which can help clinical geneticists to identify the actual genetic disorder. However, many genetic abnormalities are extremely rare, so that it is difficult for clinicians to associate a particular set of facial anomalies with a likely genetic abnormality.

The software is based on surface modelling techniques developed at UCL, and was developed using multiple facial scans of patients with specific genetic conditions, as well as controls, to produce composite images characteristic of each disorder. Analysis of a subject’s face with the programme can identify significant similarities to the catalogued genetic conditions. Describing his work at the BA Festival of Science in York, UCL Institute of Child Health researcher Professor Peter Hammond reported a 90% success rate thus far when using the programme to diagnose “over 30 conditions with an underlying genetic abnormality” including Cornelia de Lange, Fragile X, Noonan, Smith-Magenis and Velocardiofacial syndromes (see press release). In the future, Professor Hammond is reportedly hoping to compile facial data from individuals of different sexes and ethnic backgrounds with specific genetic disorders.

The software can potentially speed diagnosis by identifying the type of genetic abnormality that may be present in a patient, so that clinical geneticists have a lead to investigate via standard genetic analysis; it is hoped that this sort of tool could avoid inappropriate (and costly) genetic investigations. It could also add to the increasing movement towards international databases such as DECIPHER, which record rare and novel genetic abnormalities linked with phenotypic information from the patients so that clinical geneticists can share information that may help in diagnosis and/or clinical management of individuals with chromosomal abnormalities.

More from us

Genomics and policy news