Analytics in Action News

Using Artificial Intelligence to Diagnose Rare Pediatric Disorders

Researchers found that artificial intelligence technology can quickly diagnose rare disorders in children, allowing patients to acquire treatment sooner.

artificial intelligence

Source: Getty Images

By Erin McNemar, MPA

- According to a collaborative research study from the University of Utah Health, Fabric Genomics, and Rady Children’s Hospital, artificial intelligence-based technology can rapidly diagnose rare disorders in critically ill children with high accuracy.

These findings indicate the next phase of medicine where technology can help clinicians quickly determine the root cause of disease, providing patients with the proper treatment sooner.

“This study is an exciting milestone demonstrating how rapid insights from AI-powered decision support technologies have the potential to significantly improve patient care,” co-corresponding author on the paper Mark Yandell, PhD, said in a press release.

About seven million children are born with serious genetic disorders each year. For those children, life typically begins in intensive care. Various NICUs in the United States are now looking for genetic causes of disease by reading, or sequencing, the three billion DNA letters that made up the human genome.

While it takes hours to sequence the whole genome, manual analysis can take days or weeks to diagnose the illness. For some infants, that’s not fast enough, Yandell said. Understanding the cause of newborns’ illnesses is vital to treating them. Arriving at a diagnosis within the first 24 to 48 hours after birth gives infants the best chance to improve their condition.

Understanding that speed and accuracy are essential, Yandell’s group collaborated with Fabric to create the new Fabric GEM algorithm, which uses artificial intelligence to find DNA errors that lead to disease.

In the study, scientists tested GEM by analyzing whole genomes from 179 previously diagnosed pediatric cases from Rady’s Children’s Hospital and five other medical centers worldwide. GEM found the causative gene as one of its top two candidates 92 percent of the time, outperforming existing tools.

“Dr. Yandell and the Utah team are at the forefront of applying AI research in genomics,” said Martin Reese, PhD, CEO of Fabric Genomics and a co-author on the paper. “Our collaboration has helped Fabric achieve an unprecedented level of accuracy, opening the door for broad use of AI-powered whole-genome sequencing in the NICU.”

GEM cross-references large databases of genomic sequencing from diverse populations, clinical disease information, and other repositories of medical and scientific data and then combines this information with the patient’s genome sequencing and medical records. To sort through medical records, GEM can be used with natural language processing tools.

While existing technologies can identify small genomic variants, GEM can find structural variants as causes of disease. These changes are larger and are often more complex. According to researchers, it’s estimated that structural variants are responsible for 10 to 20 percent of genetic diseases.

“To be able to diagnose with more certainty opens a new frontier,” said Luca Brunelli, MD, neonatologist and professor of pediatrics at U of U Health, who leads a team using GEM and other genome analysis technologies to diagnose patients in the NICU.

According to Brunelli, these AI tools can now explain why a child is sick, enable doctors to improve disease management, and lead to recovery.

“This is a major innovation, one made possible through AI,” Yandell said. “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.”