The powerful algorithm can correctly identify disease-causing genes 92 per cent of the time and includes structural variants as causes of disease, improving certainty in diagnoses for critically ill children.
Scientists from the University of Utah Health and Fabric Genomics, collaborators on a study led by Rady Children’s Hospital in San Diego, have reported that an artificial intelligence (AI)-based technology can rapidly and accurately diagnose rare diseases in critically ill children. The benchmark technology foreshadows next-generation healthcare, where AI helps clinicians quickly determine the root cause of disease to provide the correct treatments 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,” commented Mark Yandell, Ph.D., co-corresponding author on the paper, professor of human genetics, Edna Benning Presidential Endowed Chair at the University of Utah Health, and a founding scientific advisor to Fabric.
Globally, an estimated seven million infants are born with serious genetic disorders each year, causing them to begin their first years of life in intensive care. Although many neonatal intensive care units in the U.S. can now search for genetic causes of disease through DNA analysis, it takes hours to sequence the whole genome and it can take days or weeks of computational and manual analysis to diagnose illnesses.
For some infants, Yandell believes our current speed is not fast enough since understanding the cause of a newborn’s illness is critical for effective treatment. Reaching a diagnosis within the first 24 to 48 hours after birth provides these patients with the best chance to improve their clinical condition, or better yet recovery.
Being at the forefront of applying AI research in genomics, Yandell’s team collaborated with Fabric to develop the new Fabric GEM algorithm that incorporates AI to detect disease-causing DNA errors.
The researchers assessed GEM’s performance by analysing whole genomes from 179 previously diagnosed paediatric cases from Rady’s Children’s Hospital and five other medical centres from across the globe. The new algorithm successfully identified the causative gene as one of its top two candidates 92 per cent of the time. Doing so outperformed existing tools that accomplished the same task less than 60 per cent of the time, demonstrating GEM’s superior accuracy.
GEM leverages AI to learn from a vast and ever-growing body of knowledge that has become challenging for clinicians and scientists to keep up with. The technology cross-references large databases of genomic sequences from diverse populations, clinical disease information, and other repositories of medical and scientific data, and combines all the data with the patient’s genome sequence and medical records. To assist medical record search, GEM can be coupled with a natural language processing tool called Clinithink’s CLiX focus, which scans reams of doctors’ notes for clinical presentations of patients’ diseases.
“Critically ill children rapidly accumulate many pages of clinical notes,” Yandell said. “The need for physicians to manually review and summarise note contents as part of the diagnostic process is a massive time sink. The ability of Clinithink’s tool to automatically convert the contents of these notes in seconds for consumption by GEM is critical for speed and scalability.”
Most existing technologies are primarily built to identify small genomic variants such as single DNA letter changes, insertions, or deletions of a small string of DNA letters. However, GEM is also equipped to find “structural variants” as causes of disease. These changes are not only larger and more complex but are also estimated to be the root of 10 to 20 per cent of genetic disease.
“To be able to diagnose with more certainty opens a new frontier,” said Luca Brunelli, M.D., neonatologist and professor of paediatrics at the University of Utah Health, who leads a team using GEM and other genome analysis technologies to diagnose patients in the NICU. Dr. Brunelli aims to provide answers to families who would have had to live with uncertainty before the development of these tools. He believes these advances now provide an explanation for why a child is sick, thereby enabling doctors to improve disease management, and, at times, 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.” [APBN]
Source: De La Vega et al. (2021) Artificial intelligence enables comprehensive genome interpretation and nomination of candidate diagnoses for rare genetic diseases. Genomic Medicine, 13, 153.