Hospital diagnoses critically ill children using artificial intelligence

The Rady Children’s Hospital used Moon, software developed by the Leuven-based company Diploid. Moon is the first software worldwide to use AI for the diagnosis of rare diseases.

This new method accelerates the diagnosis of seriously ill children in intensive care and opens the door to the use of genome sequencing as a first-line diagnostic test for babies with unknown disorders. An estimated 4 percent of newborns are affected by genetic diseases, which are the leading cause of death in infants. Rare genetic diseases also account for around 15 percent of admissions to children's hospitals.

“Some people call this artificial intelligence, we call it augmented intelligence,” said Stephen Kingsmore, MD, DSc, President and CEO of RCIGM. “Patient care will always begin and end with the doctor. By harnessing the power of technology, we can quickly and accurately determine the root cause of genetic diseases. We rapidly provide this critical information to intensive care physicians so they can focus on personalizing care for babies who are struggling to survive.”

By reducing the need for labor-intensive manual analysis of genomic data, the automated approach saved a considerable amount of time. In February 2018, the same team set the world record for the fastest diagnosis by sequencing the entire genome.

"Saving the lives of very sick newborn babies by using AI in genome analysis is truly groundbreaking work for the Rady team," said Eric Topol, MD, professor of molecular medicine at Scripps Research.

The Rady Children's Hospital has optimized various time-saving technologies and integrated them into a rapid process to screen the entire genetic composition of a child for thousands of genetic abnormalities, based on a blood sample.

A crucial element in this process is the Moon software developed by Diploid in Leuven. The platform automates the interpretation of the genetic data via AI to automatically filter and classify pathogenic mutations. Integration of the patient's symptoms, based on analysis of the medical literature, is one of the main pillars of this automated interpretation. With Moon, it takes only five minutes to find the causal mutation amidst the 4.5 million mutations in a fully human genome.

The study emphasizes that these automated diagnoses corresponded to manual interpretation by experts. The Leuven Moon software achieved 99 percent precision in 95 children with 97 genetic disorders.

"This publication of the Rady Children's Hospital is an absolute first," said Peter Schols, CEO of Diploid. “For the first time in history, an algorithm matches the precision of a geneticist in diagnosing rare diseases. Of course there is always a manual check by a geneticist and the treating physician, but as the research results indicate, our Moon software achieved 99% precision in finding the genetic cause, which is unseen in our sector. We are delighted with these results. The Rady Children's Hospital is one of the very best children's hospitals in the US and even worldwide, and it is an honor that they have chosen our software. ”

Researchers concluded that this highly automated approach allows the use of genome analysis in intensive care units, giving seriously ill children the rapid and accurate medical care they require.

Increased process automation removes a barrier to scaling up clinical use of genome sequencing by reducing the need for time-consuming manual analysis and interpretation of data by scarce clinical geneticists.

“Using machine-learning platforms doesn’t replace human experts. Instead it augments their capabilities,” said Michelle Clark, PhD, statistical scientist at RCIGM and the first author of the study. “By informing timely targeted treatments, rapid genome sequencing can improve the outcomes of seriously ill children with genetic diseases.”