| Issue |
MATEC Web Conf.
Volume 189, 2018
2018 2nd International Conference on Material Engineering and Advanced Manufacturing Technology (MEAMT 2018)
|
|
|---|---|---|
| Article Number | 07002 | |
| Number of page(s) | 6 | |
| Section | Bio Issues & Environment | |
| DOI | https://doi.org/10.1051/matecconf/201818907002 | |
| Published online | 10 August 2018 | |
Automatic summarization of medical interviews
Department of computer science, Yangzhou University, Yangzhou, Jiangsu, China
*
Corresponding author: jpqiang@yzu.edu.cn
Abstract. The genomic-based targeted therapy (Crizotinib) has been emerged as an alternative option for the treatment of patients with locally advanced or metastatic non-small cell lung cancer, comprising the 85\% of lung cancer. However, Crizotinib is not listed in VA drug formulary- and is not available for VA oncologists to treat lung cancer currently. Therefore, for understanding physicians’ views on using genomic services, semi-structured interviews were collected. In this paper, we will present an innovative method to extract summarization from medical interviews automatically. Different from keyword-based method, automatic summarization can help to understand the intention of physicians. Compared with the existing summarization methods, our work is based on latent Dirichlet allocation and recent results m word embeddings that learn seinantically meaningful representations for words from local cooccurrences in sentences. Experiments on medical interviews demonstrate that the proposed algorithm achieves good results compared with a gold standard file using manual extraction technique.
© The Authors, published by EDP Sciences, 2018
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.