New AI technology helps diagnose genetic disorders in newborns
SALT LAKE CITY (ABC4)- Scientists have developed an artificial intelligence-based technology that helps diagnose rare disorders in critically ill children.
The new findings coming from a study led by Rady Children’s Hospital in San Diego, in collaboration with University of Utah Health scientists, and Fabric Genomics,
The findings of the study were published in Genome Medicine, a publication specializing in genetic and genomic research. According to the University of Utah, the findings “foreshadow the next phase of medicine, where technology helps clinicians quickly determine the root cause of disease so they can give patients the right treatment sooner.”
Roughly 7.9 million infants around the world are born with a “birth defect of genetic or partially genetic origin” yearly, according to the National Center for Biotechnology Information. Of those infants, an estimated 3.3 million children under the age of five die from those defects.
Neonatal intensive care units (NICU) in the United States, including the one at U of U Health, are working toward finding genetic causes of disease. They do this by reading, or sequencing, the 3 billion DNA letters that make up the human genome, which takes a long time.
However, an infant may not have that much time to wait, says co-author of the study and U of U genetics professor Mark Yandell. The findings of the study offer hope in treating infants with genetic disorders by diagnosing illnesses before they can develop further.
“This study is an exciting milestone demonstrating how rapid insights from AI-powered decision support technologies have the potential to significantly improve patient care,” Yandell says. Once a child is born, time is of the essence in diagnosing any possible genetic disorders.
“Arriving at a diagnosis within the first 24 to 48 hours after birth gives these patients the best chance to improve their condition,” A press release from U of U Health says. Yandell and Fabric Genomics helped develop a technology called “GEM” whose algorithm incorporates AI to find errors in the DNA that lead to disease.
The scientists in the study tested GEM by analyzing whole genomes from 179 previously diagnosed pediatric cases from the children’s hospital in San Diego and five other medical centers worldwide. GEM was able to identify the gene that causes disorders 92% of the time. Existing tools and methods were only able to identify the causative gene 60% of the time.
GEM uses AI to learn from a body of information that continues to grow. It cross-references databases of genomic sequences from various and diverse populations as well as clinical disease information and other repositories of medical and scientific data.
Once it has analyzed that information, it combines it with the patient’s genome sequence and medical records.
Current technologies identify only small genomic differences. GEM is different in that it can also find “structural variants” that cause disease, which are often more complex.
“This is a major innovation, one made possible through AI,” Yandell says. “GEM makes genome sequencing more cost-effective and scalable for NICU applications. It took an international team of clinicians, scientists, and software engineers to make this happen. Seeing GEM at work for such a critical application is gratifying.”
More information on the study and its findings can be found here.
Gloss