FDA Explains New LabelComp Tool

FDA researchers are using an open-source article in Drug Safety to introduce LabelComp, an artificial intelligence (AI) tool that automates the identification of changes in adverse event terms made during drug labeling updates. The article says that traditional methods of updating drug labeling with new adverse event information have been manual, time-consuming, and error-prone.

The researchers say LabelComp uses a combination of text analytics and a trained Bidirectional Encoder Representations from Transformers model to automatically identify changes in adverse event terms from updated drug labeling documents.

“The LabelComp AI tool was validated and demonstrated high accuracy and thus is a fit-for-purpose tool for the identification of adverse event changes in drug labeling that can enhance regulatory science research and characterization of adverse event profiles over time,” the article says.