Using machine learning and big data in academic medicine

Two new research from the Department of Computational Biomedicine in Cedars-Sinai is advancing what they know about improving healthcare and medical research using machine learning and big data. Both studies were published in peer-reviewed journals pattern.

In the first study, Cedars-Sinai investigators applied advanced statistical techniques to analyze electronic health records from nearly 100,000 hospital stays. This approach identified drugs that were unexpectedly associated with an increase or decrease in blood glucose levels in hospitalized patients.

“Our findings provide practical insights that will help clinicians predict and manage drug-related glucose changes and ultimately improve glucose safety for hospital patients,” said Dr. Jesse G. Meyer, assistant professor of computational biomedicine at Cedars-Sinai and author of the study.

In the second study co-led by Cedars-Sinai, investigators developed a safe method to pool patient data from multiple hospitals for their research. This method allows hospitals to send statistical summary of patient characteristics rather than individual healthcare data to a central location for investigator analysis, reducing the risk of careless disclosure of sensitive patient information.

“Our innovative approach opens the door to larger and diverse research that better protects patient privacy, improves research quality, and supports the development of more effective treatments,” says Dr. Ruowang Li, assistant professor of computer biomedicine at Cedars-Sinai and co-author of the study.

“Both studies highlight our unique approach to using machine learning and big data in academic medicine,” says Dr. Jason Moore, professor at Cedars-Sinai's School of Computational Biomedical Sciences and co-author of the study. “These studies will encourage collaboration and ultimately lead to patient care and research driven by data, overcoming the outcome gaps and creating healthier lives.”