We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Attention-based Clinical Note Summarization

Abstract: The trend of deploying digital systems in numerous industries has induced a hike in recording digital information. The health sector has observed a large adoption of digital devices and systems generating large volumes of personal medical health records. Electronic health records contain valuable information for retrospective and prospective analysis that is often not entirely exploited because of the dense information storage. The crude purpose of condensing health records is to select the information that holds most characteristics of the original documents based on reported disease. These summaries may boost diagnosis and extend a doctor's interaction time with the patient during a high workload situation like the COVID-19 pandemic. In this paper, we propose a multi-head attention-based mechanism to perform extractive summarization of meaningful phrases in clinical notes. This method finds major sentences for a summary by correlating tokens, segments and positional embeddings. The model outputs attention scores that are statistically transformed to extract key phrases and can be used for a projection on the heat-mapping tool for visual and human use.
Comments: 9 Pages, 6 Figure, 2 Tables, Conference
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
Cite as: arXiv:2104.08942 [cs.CL]
  (or arXiv:2104.08942v1 [cs.CL] for this version)

Submission history

From: Neel Kanwal [view email]
[v1] Sun, 18 Apr 2021 19:40:26 GMT (3250kb,D)
[v2] Fri, 1 Oct 2021 10:51:26 GMT (3294kb,D)
[v3] Mon, 28 Feb 2022 11:15:16 GMT (4042kb,D)

Link back to: arXiv, form interface, contact.