Efficient categorization leads to efficient analysis
Descriptions
of adverse drug events, medication errors and overall treatment have the
potential to tell us a lot about a medicine. If only we could easily make sense of all that
unstructured data!
If we want to efficiently analyze larger data masses, we need to efficiently categorize the variables.
Aggressive digitalization in Finland has resulted in a wealth of data about medication use and errors. What can we do with it all?
During this webinar, Post-doc. researcher Outi Laatikainen from the University of Oulu reflected on how artificial intelligence can be used to make sense of medication error reporting in Finnish specialized care. Using a data set of 15 000 patient safety incident reports, she has tested the capabilities of natural language processing for anonymization, drug name categorization, and event categorization. Challenges along the way included proper names and medical slang. The results are not bulletproof, but they are definitely a step in the right direction!
This webinar will be held in English.
Recording
Du kan laste ned podkasten til mobilen din på Apple Podcasts, Spotify eller Podbean. Søk etter "Norwegian Centre for E-health Research".