search for




 

Early detection of seasonal influenza through medical prescriptions of respiratory diseases
Journal of the Korean Data & Information Science Society 2018;29:391-402
Published online March 31, 2018
© 2018 Korean Data and Information Science Society.

Tae Heung Kim1 · Eun Jin Jang2 · Sun-Young Jung3 · Ji-Yeon Yang4

14Department of Applied Mathematics, Kumoh National Institute of Technology
2Department of Information Statistics, Andong National University
3College of Pharmacy, Chung-Ang University
Correspondence to: Associate professor, Department of Applied Mathematics, Kumoh National Institute of Technology, 61 Daehak-Ro, Gumi 39177, Korea. E-mail: jyang@kumoh.ac.kr
Received February 17, 2018; Revised March 13, 2018; Accepted March 14, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
This study proposes a new non-clinical surveillance system for seasonal influenza using NHIS-NSC. In particular, we take notice that the symptoms of influenza at early stage of infection are similar to those of respiratory disease, and we aim to detect an early outbreak of influenza by monitoring the activity of non-influenza respiratory disease. Our surveillance system yielded a quite reliable classification result with the average AUROC value above 0.75 and the precision rate above 80%. Furthermore, our system is able to effectively detect an outbreak that can be easily missed by disease control authorities, during the influenza “off-season”. The recall rate, however, is somewhat low, and thus it would be most beneficial to use our surveillance system as a complementary tool for the current surveillance program, not as a stand-alone replacement.
Keywords : AUROC, influenza-like illness, NHIS-NSC, precision, recall, respiratory disease, seasonal influenza.