This proposal is a preliminary engagement between Abbvie and the Division of Health and Biomedical Informatics (HBMI) at the Northwestern University Feinberg School of Medicine. This proposal contains two sub-projects. The goal of these projects is to provide background and preliminary data to assess the value and feasibility of a pharmacovigilance project combining text mining with the clinical data contained in the Northwestern Medicine Enterprise Data Warehouse (NMEDW). In particular, this project will focus on Neupogen (filgrastim), as well as a newly-approved biosimilar drug called Zarxio (filgrastim-sndz). Previous pharmacovigilance projects have focused on conventional pharmaceutical agents, for which the side effect profile of the generic agent is close to that for the innovator compound. Biologic agents are now reaching their patent life and the FDA has begun approving bio-similar compounds. An open question is how pharmacovigilance should be approached for these new classes of “generic” agents. Detection of clinical outcomes and side effects in pharmacovigilance can be viewed as a subset of the more general problem of detecting clinical phenotypes. Even with the advent of electronic health records (EHRs), it is still reported that roughly 90% of the total clinical information is in textual notes and reports, rather than in coded data. Work by the Electronic Medical Records and Genomics (eMERGE) consortium has demonstrated that accurate clinical phenotypes require the use of text mining techniques to extract information from these textual documents. “Text mining”, as the name implies, describes a collection of computer techniques and algorithms for computing unstructured text documents into structured information that can be used for further analysis. There are many types of text mining. Some leverage an understanding of language, grammar, syntax and semantics to extract the information. These are typically grouped under computational linguistics. Other techniques increasingly utilize statistical pattern discovery methods treating documents as bags of words and are often grouped under “machine learning”. Both types of methods have advanced considerably in recently years. This project has two sub-projects: • Sub-project 1: Review of the current state of text mining for pharmacovigilance. • Sub-project 2: Identification and location of mentions of Neupogen (filgrastim) and Zarxio (filgrastim-sndz).
|Effective start/end date
|12/4/15 → 12/3/16
- AbbVie Inc. (Agmt 12/04/15)
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.