Los Angeles-based Cedars-Sinai is becoming more like Netflix and Google Maps in its COVID-19 response by using predictive analytics to meet the needs of patients and staff as the disease progresses, according to a Newswise report.
The health system's data science team built a platform that uses thousands of data points to predict effective treatments, likelihood for readmission and patient satisfaction. In much the same way Netflix uses algorithms to suggest new movies or TV shows based on previous choices, the health system uses data to estimate factors such as patient adherence to medication prescription after leaving the hospital. Cedars-Sinai is also using an algorithm to estimate patient care plans and disease progression, similar to the algorithm Google Maps uses to recommend the best driving routes.
The algorithm adapts as the situation changes. During the pandemic, the health system used the machine learning platform to meet daily bed, staff and PPE needs.
"If it predicts that tomorrow we'll have 100 COVID-19 patients, but that actual number turns out to be 90, then the platform automatically goes back and tries to relearn what changed to cause the outcome to be different," said Michael Thompson, Cedars-Sinai executive director of enterprise data intelligence.
The platform can predict COVID-19 patient volume with an 85 percent to a 95 percent degree of accuracy.
More articles on digital transformation:
Health Catalyst grows with Northwell partnership, revenue cycle acquisition: 4 details
3 health systems relying more on big tech during the pandemic
Why UCLA Health's IT team was uniquely prepared for the pandemic and how it will change moving forward