Boston Children's Hospital plans to use IBM's Watson supercomputer to speed up diagnoses for children with a rare and genetically linked kidney disease.
As a first step, Watson will be “trained” in nephrology by ingesting medical literature and other information about the causative mutations for steroid-resistant neophrotic syndrome, a form of kidney disease. Then, specialists from Boston Children's will feed the computer genetic-sequencing data from its own patients.
According to IBM, the Boston Children's project represents the first time the technology developed in its cancer initiative will be applied to disease diagnosis and treatment.
Staying clinically up to date, even within a medical specialty, has long exceeded the capacity of a single human brain. Even coupling available knowledge with genetic information from a single patient is time-consuming, according to Steve Harvey, vice president of the Watson Health unit at IBM.
For example, researchers at Boston Children's publish 150 articles every month, Harvey said. Around the world, 150,000 articles are published every year on cancer alone.
“It's really challenging for humans to keep up with all of the knowledge that's being generated,” Harvey said. “The second issue is that it takes somewhere on average of 25 hours, once the sequencing has been done, for an expert in the space to analyze the data and come up what they think” might be the correct diagnosis, he said.
At Boston Children's, Watson should considerably speed up that analysis. A process that would normally take weeks for a physician to complete is expected to take Watson only a few minutes.
“Coping with an undiagnosed illness is a tremendous challenge for many of the children and families we see,” said Dr. Christopher Walsh, chief of Boston Children's division of genetics, in a news release. “Watson can help us ensure we've left no stone unturned in our search to diagnose and cure these rare diseases, so we can uncover all relevant insights from the patient's clinical history, DNA data, supporting evidence and population health data.”
This spring, IBM launched a cancer research initiative with Watson to analyze genomic data with a goal to find personalized treatment options. The effort includes research nodes with 14 hospitals and cancer-treatment centers in the U.S. and Canada.
IBM has also pledged to work with electronic health-record developer Epic Systems Corp. to integrate Watson's cognitive abilities into its EHR system. Watson will use patient-specific data to provide medical literature and case studies in real time in conjunction with Epic's advanced decision-support technology.