Montefiore tests semantic data lake's predictive capabilities

Montefiore Medical Center in New York plans to go live in July with a program that incorporates its semantic data lake to predict when patients are likely to have a major event within 48 hours, according to HealthITAnalytics.

Montefiore built the data lake and partnered with Mayo Clinic on a pilot of a predictive algorithm based on data from 68,000 patients from the two institutions.

Data lakes are based on graph database technology, which allow users to explore relationships between various data elements. The data elements are standardized in the semantic data lake, allowing them to be queried and matched in unique ways to answer questions on the fly.

For the pilot program, Montefiore creates patient risk scores for a major event within 48 hours, then another engine evaluates what can be done and sends personalized recommendations for proposed interventions to the doctor, Parsa Mirhaji, associate professor of systems and computational biology and the director of clinical research informatics at the Montefiore Medical Center-Institute for Clinical Translational Research, said in the article.

So far, it's being tested in comparison with usual care to determine its effectiveness. Once the system is live, researchers will collect data on whether practitioners use the recommendations and why or why not, according to the article. In addition, because it is a learning system, all the feedback will go back into the system to improve it.

Montefiore also is looking to use the system to determine how behavioral or mental health issues affect outcomes and for medication reconciliation, discharge education and continuous monitoring.

To learn more:
- here's the article