Georgia Tech researchers release open-source AI algorithm to predict effectiveness of cancer drugs

A team of researchers from Atlanta-based Georgia Institute of Technology introduced an open-source algorithm Oct. 26 that predicts a cancer drug's effectiveness based on a patient's genetic data.

The researchers developed the machine learning algorithm using gene expression and drug response data from the National Cancer Institute's panel of 60 human cancer cell lines. Their goal was to create an algorithm that predicts optimal drug therapies based on individual patient tumors.

In a study of 273 cancer patients, researchers found the algorithm to be about 85 percent accurate in assessing the effectiveness of nine drugs. By releasing the algorithm on an open-source platform, they hope other researchers will participate in refining their work.

"With our project, we're advertising that sharing should be what everybody does," said Dr. Fredrik Vannberg, a research collaborator on the open-source algorithm and assistant professor in Georgia Tech's School of Biological Sciences. "This can be a win for everybody, but really it's a win for the cancer patients."

To access the open-source algorithm or to read the full study, published in PLOS One, click here.

More articles on data analytics & precision medicine:
Harvard, IBM researchers develop prediction model for cirrhosis outcomes
Johns Hopkins, Northwestern among institutions leading new National Center for Data to Health
78% of physician leaders cite 'incomplete data' as a top threat to patient safety, survey finds

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months