Knowledge extraction

The matrix is full of anonymized information on this immense cohort of patients that can be probed with different methods, such as KESER, to gain new insights into human health. Using a series of modern statistical methods, the team transformed summary data into vectors, tuned a model that encodes the relatedness of each vector and extracted the most important features and feature weights for each phenotype.


“These statistical methods, which include Gaussian graphical models for sparse modeling of covariance structures, are particularly capable in attribution of importance that exposes potential causal relationships, a concept with which classical AI technology, such as deep learning, tends to struggle," said George Ostrouchov, ORNL senior research scientist and lead statistician on the MVP-CHAMPION project.

After running the KESER method, the team selected eight phenotypes — including depression, rheumatoid arthritis and ulcerative colitis — to explore. Using the features selected by KESER, they trained models to identify the phenotypes of interest.

Future research

The possibilities enabled by KESER’s novel ability to anonymize, integrate and analyze data from multiple healthcare institutions seem limitless.

Tianxi Cai, professor of Biomedical Informatics at Harvard Medical School and a principal investigator of KESER, said, “We are excited to have a highly scalable approach that can handle matrices an order of magnitude bigger than what we are working with now.”

The team is already incorporating more clinical descriptors into the knowledge graphs. Additionally, the team has started exploring the knowledge graphs to better understand emerging diseases.

“In a situation like COVID, for example, where everybody needs to share data and we need to start investigating all the different things that are related to this specific disease, you would potentially be able to do that with this system,” said Chuan Hong, assistant professor at Duke University, who led research on the KESER project as an instructor at Harvard last year. “It’s basically plug-and-play; you go to the data warehouse, follow the four-step process and directly integrate your results.”

The potential for future collaboration and discovery may be the project’s greatest success. “This innovation will facilitate multi-center collaborations,” the team wrote in Nature, “and bring the field closer to the promise of creating distributed networks for learning across institutions while maintaining patient privacy.”

Back to HCB News

Health IT Homepage