EMA Unveils ERATO its new AI Tool improving Pharmacovigilance Signal Detection

Among the initiatives discussed by the European Medicines Agency (EMA) in its recently published annual report on 2024 activities was its new AI tool Enhanced Review of Abstracts with Transformer models (ERATO).

The purpose of ERATO is to reduce by an estimated 80% to 90% the roughly 200,000 abstracts the EMA Pharmacovigilance Office reviews every year. This could save more than 0.5 FTE (full time equivalent) per week.

The tool screens scientific literature publications to detect safety signals, which are new or known adverse events that may be caused by a medicinal product and require further investigation.

In a related commentary, an EMA official commented that ERATO had also resulted in the “secondary benefit” of helping the EMA to establish a more robust tracking system for the review of safety signals identified during the screening process.