Schneller zur genetischen Diagnose bei Kindern mit auffälliger Fazies durch künstliche Intelligenz
摘要
A diagnosis at first glance is one of the oldest and most fascinating skills in medicine. It enables a diagnosis to be made or at least a preliminary suspicion to be formulated, solely by an experienced eye. In the case of rare genetic disorders, however, this type of diagnosis is often reserved for dysmorphologists, experts specifically trained to recognize subtle characteristic facial features that can indicate a syndrome. Many pediatricians also have a pronounced intuition for recognizing an unusual facial appearance in a child. In clinical records this is sometimes noted with the abbreviation funny looking child (FLC), an expression that describes the difficulty of clearly naming and classifying a visually observed anomaly. With the help of artificial intelligence (AI), particularly through the GestaltMatcher system (so-called next generation phenotyping, NGP) and subsequent rapid initiation of genetic diagnostics (next generation sequencing, NGS) for selected children, pediatricians are now empowered to make a diagnosis quicker than before of even the rarest genetic diseases in FLCs. This not only shortens the time of uncertainty until making the diagnosis but can also be of considerable therapeutic and psychological benefit to the affected children and their families.