IBM aims machine learning at type 1 diabetes with JDRF partnership

Big Blue scientists will apply algorithms to spot patterns that could signal ways to delay or prevent juvenile diabetes.
By Mike Miliard
01:44 PM
IBM machine learning

A new initiative between IBM and JDRF, one of the leading groups funding the fight against type 1 diabetes, will apply machine learning to troves of worldwide research data accumulated over the years with the aim of uncovering commonalities that could point to diabetes risks in children.

The partnership is meant to give type 1 diabetes a foothold in emerging precision medicine efforts, officials say, combining JDRF's global research with the computing power of IBM.

Scientists will start by examining three separate data sets, applying algorithms to help spotlight certain patterns or common factors that might identify ways to delay or prevent type 1 diabetes in kids. The analytics tools will mine genetic, familial, autoantibody and other data to arrive at a set of features that is common to all three data sets.

[Also: IBM Watson thinks it can use AI to fight opioid addiction]

The models that emerge should quantify the risk for juvenile diabetes from the combined dataset using this foundational set of features, officials say. That will enable JDRF to better identify top predictive risk factors, cluster patients based on them and explore a number of data-driven models for predicting onset.

A bit further on, the partners have eyes toward putting big data to work helping understand root causes of type 1 diabetes and hope to apply analytics to more complex datasets, such as microbiome and genomics or transcriptomics data. The hope, of course, is to eventually put the insights gleaned from these projects toward a cure.

"JDRF supports researchers all over the world, but never before have we been able to analyze their data comprehensively, in a way that can tell us why some children who are at risk get T1D and others do not," said JDRF CEO Derek Rapp in a statement. "IBM's analysis of the existing data could open the door to understanding the risk factors of T1D in a whole new way, and to one day finding a way to prevent T1D altogether."

"Nearly 40,000 new cases of type 1 diabetes will be diagnosed in the U.S. this year. And each new patient creates new records and new data points that, if leveraged, could provide additional understanding of the disease," said Jianying Hu, senior manager and program director at IBM Research's Center for Computational Health.

"The deep expertise our team has in artificial intelligence applied to healthcare data makes us uniquely positioned to help JDRF unlock the insights hidden in this massive data set and advance the field of precision medicine towards the prevention and management of diabetes," he said

Twitter: @MikeMiliardHITN
Email the writer: