Weeding out duplicates to better detect side-effects – New podcast episode

Research / 26 August 2024

Duplicate reports are a big problem when it comes to signal detection, but with the help of machine learning and new ways of comparing reports, we may more effectively detect them.


In this episode, Jim Barrett, Senior Data Scientist at UMC and Fredrik Brounéus discuss algorithms: specifically, the duplicate detection algorithm and its use in VigiBase, and generally on different approaches to creating algorithms and the challenges in evaluating their performance.

This episode is part of the Uppsala Reports Long Reads series – the most topical stories from UMC’s pharmacovigilance news site, brought to you in audio format. Find the original article here. Subscribe by visiting the Drug Safety Matters website.

Related Articles

AI meets psychedelic storytelling: What online forums can teach pharmacovigilance

Revisiting the role of medication utilisation evaluations to assess adherence to risk minimisation measures

Beyond numbers, quality in ADR reporting – New podcast episode

You may also like

AI meets psychedelic storytelling: What online forums can teach pharmacovigilance

Online psychedelic forums hold untapped safety data. AI analysis of user narratives could help pharmacovigilance systems detect risks missed by traditional reporting.

13 November 2025

Beyond numbers, quality in ADR reporting – New podcast episode

Spontaneous adverse event reports pose significant challenges to pharmacovigilance scientists. How may we turn them into opportunities that benefit our pharmacovigilance systems?

04 June 2025

Revisiting the role of medication utilisation evaluations to assess adherence to risk minimisation measures

MUEs are typically used to improve the safety of drugs, however, may they also have a role in measuring the adherence to risk minimisation measures?

06 August 2025