Researchers from the University Hospital Bonn in Berlin, Germany have successfully used artificial intelligence image scanning to diagnose patients with Glycolipid (GPI) anchor deficiencies, reports Healthcare Analytics News. They recently published their results in the Genome Medicine journal. While this is great for the GPI community, they are hoping this technology will be applied to other diseases as well.
GPI anchor deficiency is a condition that damages cell-to-cell communication. It’s caused by a genetic mutation that prevents the body from producing sufficient GPI.
Alexej Knaus, a PhD from the the University of Hospital Bonn of Genome of Statistics and Bioinformatics shares her excitement for the potential of this AI technology. She believes it will greatly improve the number of diagnosed individuals with GPI anchor deficiencies and eventually other conditions.
The researchers were also successful in simulating characteristics of another genetic condition called Mabry syndrome, which causes distinct facial features and intellectual disability, among other symptoms. They relied on data looking at surface texture, facial features, as well as genetic make-up of cells.
Peter Krawitz, a co-author with Knaus on the study, shared that patients with Mabry syndrome often spend years looking for an accurate diagnosis. He believes this technology will be a huge game changer. Considering the lack of widespread knowledge on the disease, it’s hard for these patients to be correctly diagnosed. With this technology, they will be be able find answers quicker, even if a doctor is unaware of these characteristics.
In the study, they photographed 91 patients with Mabry syndrome to look for similarities, and were successful. They believe that if they continue to gather photographs from many different patients of the same rare disease, they can find facial similarities that might not be so noticeable by the eye.
Krawaitz was one of the first skeptics to this technology, but he couldn’t even deny the amazing performance it has shown and grew to be a huge advocate for it. While more research needs to be done, and more data needs be gathered, researchers are well on their way to further implement AI technology into rare disease diagnosis.