A Software can conclude on the basis of portrait photos quite reliable on rare genetic diseases of the people. Researchers from the USA, Israel and Germany in the journal "Nature Medicine" the program "deep design" that is more than 200, can usually detect very rare syndromes. In the case of an initial suspicion, the Software could narrow down the number of possible genetic causes and diagnosis accelerate, says Co-author Peter Krawitz from the University hospital Bonn. The medical possibilities of such networks are rich, therefore, but much more.
a Difficult, often lengthy diagnosis
About two to eight percent of the population genetically have syndrome, says Krawitz. A third to half of the diseases associated with mental retardation, the show often already in infancy. "Because of the large number of possible syndromes, and its rarity, the correct diagnosis is a lengthy and expensive process," writes the Team of Yaron Gurovich from the Boston-based company FDNA. So far, only a few experts unusual appearance could recognize pictures, or extremely rare the symptoms. Automated systems have the potential to improve the Situation significantly.
The Software deep the shape of investigated frontal shots of faces on characteristic abnormalities, and analysed around the shape of the eyes, the mouth, the chin or the distance between the eyebrows. The artificial neural network technology that mimics the pattern recognition in the brain - mind, among other things, to 130 points in the face, and compares these with 216 syndromes and lists on the basis of the Similarities and differences of the most likely genetic causes. "In the database, the patient is matched to a photo with lots of pictures, and a total similarity is calculated," says the bioinformatics scientist Krawitz.
Software to 17000 images
The most elaborate trained according to Krawitz, the Training of the Software to a data set of more than 17,000 images. So, for example, the detection of the Cornelia de Lange syndrome (CdLS) was practiced on the basis of 614 images of Affected and nearly 1100 pictures of other people. In a subsequent Test of whether a person has this syndrome or not, has reached the program has a reliability of 97 percent. In Angelman syndrome, for which Training images of nearly 770 people Affected and about 2700 other people were used, the reliability of 92 percent. In these two Tests, however, it was only a matter of whether someone has this syndrome or not. It served, ultimately, the sensitivity of the Software.
In two other Tests, the researchers tested then, as well-deep shape to be able to assign a face photo, one of 216 different genetic defects. According to the analysis, the System gave a Top Ten of possible diagnoses: The probability of this was finding the actual gene defect in this Top Ten, was about 90 percent. About 65 percent of the cases, the most likely current diagnosis even met.
support for children's doctors to more targeted diagnostics
to initiate the Software could be of children's doctors, would those parents with eye-catching children, says Krawitz. However, the program was trying to deliver a suspected diagnosis, which would then need to be laboratory verified. "The value lies in the fact, that some children, doctors can make in the ideal case, in consultation with a human geneticist, targeted diagnostics, Vera," says the bioinformatics scientist. This will speed up the diagnosis. Currently, even the experts are overwhelmed with the diagnosis, often because of the number of newly discovered syndromes increases continuously.More about
Faster for the diagnosis of Googling in the genomeSascha Karberg
The diagnosis of genetic diseases is only one of many applications of Artificial intelligence in medicine. Similar software to be developed according to Krawitz currently for the analysis of other images, such as MRI scans or photos of the retina. And the development standing at the very beginning, says Krawitz. "In a few years time, people will begin to associate the networks with each other." Walter Willems (dpa)