Machine Learning for Data Governance in a Hospital Setting

Learn more about how Clinovera worked with a leading U.S. healthcare system to develop a searchable catalog of more than 100,000 analytical insights that are organized and classified automatically with innovative machine learning tools – with minimal human involvement.

About the Client

• Leading US System of Academic Medical Centers and Care Facilities
• Client Business: Patient Care, Clinical Research
• Summary: Developing a searchable catalog of actionable, valuable patient insights using machine learning
• Services: Artificial Intelligence, Machine Learning, Application Development

Description

The widespread adoption of EHR systems brings along a large amount of data available for reporting and other analytic insights. In fact, health organizations routinely generate thousands of such analytical artifacts daily or weekly. This continuous stream of analytical insights is often taxing for the infrastructure and is a burden for supporting IT staff.

To increase the value of available analytics and potentially reduce the number of such artifacts, it’s important to answer the following questions:

• Which clinical, operational or financial metrics are associated with a report or dashboard?
• Is a given report a complete and/or partial duplicate of another report?
• Do these reports refer to the same and correct data elements in the source EHR?
• Who are the right consumers for these insights in the organizations, and who should have the right to access them?

The Clinovera team was engaged by a leading U.S. healthcare system to create a searchable catalog that contains more than 250,000 of reports that can be automatically organized and classified.

Results

Using machine learning algorithms to detect duplicates and similarities in the reports and provide assignments of the right metrics and portfolio membership, we created a usable and dynamic system for analytics. As new analytical insights are developed, these processes continue to evolve and improve with minimal human involvement.