Facial scans picked for genetic prediction
A new project could make ‘selfies’ the diagnostic tool of the future.
A computer program has been designed to identify rare genetic conditions by analysing a face from an ordinary digital photograph.
It may even be able to find unknown genetic disorders, by scanning groups of photos in its database for shared facial features.
While genetic tests are beginning to emerge or a range of inherited conditions, rare disorders are still though to affect about 6 per cent of the population, and are notoriously difficult to spot.
This means that the rate of missed diagnoses could be high.
Researchers on the facial recognition tool say doctors too often have to rely on the pronounced facial features that occur in between 30 and 40 per cent of rare disorders to make a diagnosis, though few are trained to recognise them.
The software, developed at the University of Oxford, should help family doctors or general paediatricians make a preliminary diagnosis quickly and accurately.
To train the system, the team fed its computer vision algorithm over one thousand pictures of people covering eight different genetic disorders, including Down's syndrome, fragile X syndrome and progeria.
The computer used a pattern of 36 facial features in each shot to learn each condition from next, incorporating details such as the shapes of eyes, brows, lips and noses.
The Oxford team has recently expanded the software to now recognise 90 different disorders.
It does not give an exact diagnosis every time, but as the number of faces in database grows into the thousands, researchers estimate that the system is already about 30 times more likely to help make a correct diagnosis than by chance alone.
Researchers Christoffer Nellåker and Andrew Zisserman say they want the program to be an efficient and effective first point-of-call for maternity wards, genetic counsellors, GPs and other frontline staff.
“Worldwide, its main use will be in countries where you don't have any access to clinical genetics at all,” said clinical geneticist Dr David FitzPatrick.
“It's potentially a tremendous step forward in shortening the diagnostic journey that families embark on following birth of a child with dysmorphic features,” said Alastrair Kent, director of the charity Genetic Alliance UK.
“If validated, it will provide access to expert advice and guidance for families more quickly and efficiently than is currently possible.”
More details are available in the recent research paper, published in eLife.