A team at Stanford University have computationally extracted information from free-text clinical notes in medical records, in order to identify
drug-induced adverse events and adverse reactions from
drug-drug interactions. Information from the notes was deidentified and transformed into a patient-feature matrix using a medical ontology hierarchy system. They demonstrate that the method could be used to flag potential adverse reactions earlier than FDA alerts are reported by current methods.
Pharmacovigilance Using Clinical Notes
Lependu P, Iyer SV, Bauer-Mehren A, Harpaz R, Mortensen JM, Podchiyska T, Ferris TA, Shah NH.
Clin Pharmacol Ther. 2013 Mar 4. doi: 10.1038/clpt.2013.47.
