This project aims at an open source implementation of a software that automatically detects potential adverse drug events from any kind of text (both scientific, social media, news, electronic health records) and in several languages (english, german, italian, chinese, japanese, ...).
- Web user interface for mobile users
- Standalone user interface for desktop users
- Web service interface (e.g. for KNIME users)
- UIMA interface (e.g. for GATE and LUXID users)
- piped commandline interface (e.g. for emedding in Un*x piped workflows)
- XML output for all non GUI use cases
- check against "known" adverse drug reactions (using latest SIDER version)
- links to individual official case reports ( US FAERS; UK Yellow Cards)
- links to individuual scientific case reports ( MEDLINE )
- multilingual (english first, then other EMEA languages and CJK )
- multistyle (scientific, journalists, social media, hospital records )
Adverse drug events are a burning issue for society: thousands of casualties (some of them fatal) are the result of unexpected ( from clinical trials or from post marketing surveillance ) and certainly unwanted reactions. This tool address the need of timely up-to-date detection of such occurrences in any kind of written media, ideally in any language.