Health record relationships. KESER Knowledge Extraction.
VA, ORNL, Harvard Develop Novel Method to Identify Complex Medical Relationships
Oak Ridge National Laboratory, April 28, 2022
A team of researchers from the U.S. Department of Veterans Affairs (VA), the U.S. Department of Energy's Oak Ridge National Laboratory (ORNL), Harvard Medical School, and Brigham and Women’s Hospital has developed a novel technique to identify complex medical relationships from electronic health records. The KESER (Knowledge Extraction via Sparse Embedding Regression) method combines data from the VA and Boston-based non-profit hospital and physicians network Partners Healthcare, and facilitates automated feature selection to support phenotype identification algorithms and knowledge discovery. KESER converts data into a structured format, builds a low-dimensional-vector model of each medical code, selects features to attribute importance, and charts attributed relationships into a network. The researchers processed vast medical datasets, built a co-occurrence matrix of 100,000-plus healthcare codes, and automated data pre-processing .... '
No comments:
Post a Comment