A Hybrid Approach to Precision Medicine-related Biomedical Article Retrieval and Clinical Trial Matching

We describe our systems implemented for the Text Retrieval Conference (TREC 2017) Precision Medicine track. We submitted five runs for biomedical article retrieval and five runs for clinical trial matching. Our approaches combine strict rule matching with an ontology-based solution. Evaluation results demonstrate that our best run obtained the 2nd highest precision (P@5) score for the clinical trial matching task and was consistently ranked within top 5 teams in all evaluation metrics for the biomedical literature retrieval task.

[1]  Ellen M. Voorhees,et al.  The effect of sampling strategy on inferred measures , 2014, SIGIR.

[2]  Martha Palmer,et al.  Verb Semantics and Lexical Selection , 1994, ACL.

[3]  Dan Wu,et al.  Toward a Robust data fusion for document retrieval , 2008, 2008 International Conference on Natural Language Processing and Knowledge Engineering.

[4]  Ellen M. Voorhees,et al.  Overview of the TREC 2020 Precision Medicine Track , 2017, TREC.

[5]  Oladimeji Farri,et al.  Automated clinical diagnosis: The role of content in various sections of a clinical document , 2017, 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[6]  Oladimeji Farri,et al.  Using Neural Embeddings for Diagnostic Inferencing in Clinical Question Answering , 2015, TREC.

[7]  Javed A. Aslam,et al.  Models for metasearch , 2001, SIGIR '01.