In most cases, these genetic changes are also expressed by characteristic facial features: for example, by clearly defined eyebrows, the base of the nose or the cheeks. However, this varies between diseases.
Professor Dr Peter Krawitz from the Institute for Genomic Statistics and Bioinformatics (IGSB) at the University Hospital Bonn (Germany) said: “The aim is to detect such diseases at an early stage and initiate therapy. appropriate as soon as possible.
In the new study, the AI system “GestaltMatcher” is described as a continuation of the “DeepGestalt”, which the IGSB team trained with other organizations a few years ago.
While DeepGestalt still requires about ten unrelated affected people as training references, its successor “GestaltMatcher” requires significantly fewer patients for feature matching. This is a great advantage in the group of very rare diseases, where only a handful of patients are reported worldwide.
More, The new AI system also considers similarities with patients who are also undiagnosed and thus incorporate undescribed characteristics..
The researchers used 17,560 patient photos, most of which came from digital health company FDNA, with which the team collaborated to develop a web service through which AI could be used. use.
Some 5,000 photographs and patient data were contributed by the team at the Institute of Human Genetics at the University of Bonn, along with nine other university sites in Germany and abroad. The researchers focused on disease patterns as diverse as possible.
They were able to look at a total of 1,115 different rare diseases.
Aviram Bar-Haim of FDNA Inc. “We are excited to finally have a phenotypic analysis solution for extremely rare cases that can help clinicians tackle difficult cases and researchers,” said in Boston, USA. research to improve understanding of rare diseases”.
Doctors were able to use their smartphones to take portraits of patients and use AI to make differential diagnoses. GestaltMatcher helps doctors evaluate and supplement expert opinion.
The GestaltMatcher Database (GMDB) will improve the comparability of algorithms and provide the basis for further development of artificial intelligence for rare diseases, including other medical imaging data such as X-rays. or retinal images from ophthalmology.