Abstract: Currently there is no way to keep track of the cohort establishment methods (including cohort definition, confirmation and attribute extraction) used in cohort studies. For many diseases, there are no standardized way to establish the cohorts. Our work proposed a framework that combined reference mining and semantic filtering to identify previous work cited in the reference that studied the same cohort. In this way, we could identify potential cohort establishment discrepancies exist among studies.

Learning Objective 1: Understand the need to standardize cohort establishment method (including cohort definition, confirmation and attribute extraction).

Learning Objective 2 (Optional): Understand how reference mining and semantic filter can be combined to enhance the precision of reference retrieval.


Yiqing Zhao (Presenter)
University of Wisconsin Milwaukee

Yanshan Wang, Mayo Clinic
Feichen Shen, Mayo Clinic
Benjamin Yan, Mayo Clinic
Shulan Tian, Mayo Clinic
Jerry Yang, University of Minnesota-Twin Cities
Nan Wang, University of Minnesota-Twin Cities
Hongfang Liu, Mayo Clinic

Presentation Materials: