HIMSS19: IBM Watson teams with Brigham, Vanderbilt on $50M AI research initiative

IBM Watson Health is making $50 million investment in healthcare artificial intelligence with the aim of exploring how the technology can be used to improve patient safety and health equity.

The company announced a 10-year investment in collaboration with Brigham and Women’s Hospital—which is a teaching hospital of Harvard Medical School—and Vanderbilt University Medical Center to research the use of AI to address major public health issues.

The collaborations will focus on critical health problems that are ideally suited for AI solutions such as improving the utility of electronic health records (EHRs) and claims data to address patient safety, precision medicine and health equity, IBM officials said in a release.

The research will also explore physician and patient user experience and interactions with AI technologies.

Kyu Rhee, M.D., vice president and chief health officer at IBM Watson Health, said physicians are spending an average of two hours with their EHRs and deskwork for every hour of patient care. It's an issue the American Medical Association said is leading to a steady increase in physician burnout.

"AI is the most powerful technology we have today to tackle issues like this one, but there is still a great deal of work to be done to demystify the real role of AI in healthcare with practical, proven results and clear-cut best practices," Rhee said in a statement. "By putting the full force of our clinical and research team together with two of the world’s leading academic medical centers, we will dramatically accelerate the development of real-world AI solutions that improve workflow efficiencies and outcomes."

RELATED: IBM and MIT launch $240M AI research lab with a focus on healthcare applications

The collaborations enable IBM Watson Health scientists the opportunity to work with leading health informatics researchers to advance the field in the areas of artificial intelligence, clinical decision support and implementation science, said Gretchen Purcell Jackson, M.D., IBM Watson Health's vice president and chief science officer.

Medical data is expected to double every 73 days by 2020, Jackson noted.

"As a practicing surgeon, I often had to make critical decisions about children’s lives without time to dig for information buried in electronic health records or sift through thousands of studies in the literature. Our collaborative research will unlock new insights that affect broad health stakeholders: from providers, payers, governments, and life science companies to ultimately the most important stakeholder, patients, and seek to improve health around the globe," she said.

David Bates, M.D., chief of general internal medicine at Brigham and Women's Hospital and professor of medicine at Harvard Medical School, said researchers are aiming to find new ways to leverage AI to improve the utility of the EHR and claims data to address major public health issues like patient safety.

Predicting cardiovascular disease

IBM Watson Health and the Broad Institute, which consists of MIT, Harvard and Harvard-affiliated hospitals, also announced a three-year project to use genomics, clinical data and AI to better predict the possibility of serious cardiovascular diseases.

Announced at the Healthcare Information and Management Sytems Society's (HIMSS) annual conference and exhibition on Wednesday, project leaders hope to help provide doctors with tools to tap into the potential of genomics data and better understand the intrinsic possibility an individual has for a certain disease.

With these insights, health professionals can potentially intervene and help to reduce this risk.  

The initiative will incorporate population-based and hospital-based biobank data, genomic information and EHRs to build upon and expand the predictive power of genetic risk scoring.

Researchers plan to develop AI tools to produce models that bring together and analyze a multitude of genetic risk factors within an individual’s genome along with existing health records and biomarkers to help clinicians more accurately predict the onset of complex and often fatal conditions in patients, such as heart attacks, sudden cardiac death and atrial fibrillation.

The developed AI technologies will require innovation on three fronts: the ability to integrate several disparate types of health data for modeling; the potential to transfer and apply models on patients from different health systems; and communicating generated insights and analysis results to patients and doctors in a way that is meaningful and actionable.