– Machine learning and clinical natural language processing have led to a new record time for genetic sequencing while offering hope to seriously ill pediatric patients with rare conditions.
At Rady Children’s Institute for Genomic Medicine (RCIGM), the research arm of Rady Children’s Hospital in San Diego, researchers have employed a novel technique for whole genome sequencing that can reduce the time-to-diagnosis by more than 20 hours.
For children with genetic conditions presenting to intensive care, this speedy process can significantly increase the chances of matching patients with precision therapies that produce positive results.
“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.”
Around four percent of babies are born with genetic conditions, and these conditions are collectively the leading cause of infant deaths in North America.
Rare diseases with genetic components also contribute to around 15 percent of admissions to children’s hospitals, offering researchers an important opportunity to apply machine learning to a complex and widespread issue.
The same team holds a Guinness World Record for fasting diagnosis through genetic sequencing. Artificial intelligence helped complete the process in about 19 hours, supporting the fastest automated passes through the data.
Reducing the time and expense of genetic sequencing could democratize access to cutting-edge treatments for pediatric patients and potentially improve outcomes.
With fewer than 1600 certified clinical medical geneticists practicing across the nation in 2017, scaling up access to these critical insights could allow more children to receive better, faster treatment for their conditions.
Between July of 2016 and March of 2019, RCIGM has completed testing on 750 children, one-third of whom were then diagnosed with a genetic disease. Of those patients, 25 percent have seen immediate clinical benefits from the diagnosis and a change in treatment patterns due to the new knowledge.
“This is truly pioneering work by the RCIGM team—saving the lives of very sick newborn babies by using AI to rapidly and accurately analyze their whole genome sequence ” said Eric Topol, MD, Professor of Molecular Medicine at Scripps Research.
In order to continue accelerating the availability of precision diagnostics for pediatric patients, Kingsmore and his team used technologies from a variety of vendors, including Illumina’s gene sequencing offerings, machine learning components from Fabric Genomics and Diploid, and natural language processing tools from Clinithink.
The platform returned results in a median of 20 hours and 10 minutes, offering 97 percent recall and 99 percent precision in 95 children with 97 genetic diseases – comparable to human experts.
Clinical natural language processing automatically extracted data on deep phenomes from patient EHR data with 80 percent precision and 93 percent recall, adding to the speed and accuracy of the results.
“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.”