Document Citation

Title:
Domain Adaptation of Coreference Resolution For Radiology Reports.

Author(s):
Apostolova E, Tomuro N, Mongkolwat P, Demner-Fushman D.

Institution(s):
1) College of Computing and Digital Media, DePaul University, Chicago, IL Department of Radiology, Northwestern University Medical School, Chicago, IL Communications Engineering Branch, National Library of Medicine, Bethesda, MD

Source:
Proceedings of the 2012 Workshop on Biomedical Natural Language Processing (BioNLP '12), Association for Computational Linguistics. Stroudburg, PA. 2012:118-121.

Abstract:
In this paper we explore the applicability of existing coreference resolution systems to a biomedical genre: radiology reports. Analysis revealed that, due to the idiosyncrasies of the domain, both the formulation of the problem of coreference resolution and its solution need significant domain adaptation work. We reformulated the task and developed an unsupervised algorithm based on heuristics for coreference resolution in radiology reports. The algorithm is shown to perform well on a test dataset of 150 manually annotated radiology reports.

Publication Type: CONFERENCE

More about this article:

Abstract | Full Text (PDF)