The programme allows a computer to scour family snaps for facial features characteristic of conditions such as Down's syndrome.
In future, it could be used to identify people born with such disorders, allowing them to be given early treatment and support.
Between 30 per cent and 40 per cent of rare genetic disorders are thought to involve changes in the shape of the face and skull.
However, they may not be obvious. A majority of those with the conditions do not receive a genetic diagnosis.
The computer software employs a form of "artificial intelligence" to learn what aspects of a person's face are linked to particular conditions.
Analysing ordinary family photos, it maps the corners of the eyes, nose, mouth and other features and compares the results with a database of remembered features.
The programme automatically takes account of variations in lighting, image quality, background, pose, and facial expression.
Professor David FitzPatrick, of the Medical Research Council's Institute for Genetics and Molecular Medicine at the University of Edinburgh, said: "Getting to a firm diagnosis as quickly as possible is very important, but the extremely large number of possible disorders makes this very challenging.
"This technology will help doctors to pick up extremely rare disorders and may be able to identify new conditions. "
Lead researcher Dr Christopher Nellaker, from the Medical Research Council's Functional Genomics Unit at Oxford University, said: "A diagnosis can provide parents with some certainty and help with genetic counselling on risks for other children or how likely a condition is to be passed on.
"A diagnosis can also improve estimates of how the disease might progress, or show which symptoms are caused by the genetic disorder and which are caused by other clinical issues that can be treated."