주메뉴 바로가기 본문 바로가기
  • KDCA
  • Contact us
  • E-Submission

PHRP : Osong Public Health and Research Perspectives

OPEN ACCESS. pISSN: 2210-9099. eISSN: 2233-6052
About
Browse articles
Editorial policies
For authors
Original Article

Risk Assessment Program of Highly Pathogenic Avian Influenza with Deep Learning Algorithm

Osong Public Health and Research Perspectives 2020;11(4):239-244.
Published online: July 31, 2020

aVeterinary Epidemiology Division, Animal and Plant Quarantine Agency, Gimcheon, Korea

bKorea Telecom, Seoul, Korea

*Corresponding author: Hachung Yoon, Veterinary Epidemiology Division, Animal and Plant Quarantine Agency, 177 Hyeoksin 8-ro, Gimcheon, 39660, Korea, E-mail: heleney@korea.kr
• Received: March 22, 2020   • Revised: May 2, 2020   • Accepted: June 22, 2020

Copyright ©2020, Korea Centers for Disease Control and Prevention

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

  • 8,394 Views
  • 118 Download
  • 3 Web of Science
  • 3 Crossref
  • 3 Scopus
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • Avian Influenza: Lessons from Past Outbreaks and an Inventory of Data Sources, Mathematical and AI Models, and Early Warning Systems for Forecasting and Hotspot Detection to Tackle Ongoing Outbreaks
    Emmanuel Musa, Zahra Movahhedi Nia, Nicola Luigi Bragazzi, Doris Leung, Nelson Lee, Jude Dzevela Kong
    Healthcare.2024; 12(19): 1959.     CrossRef
  • Big data-based risk assessment of poultry farms during the 2020/2021 highly pathogenic avian influenza epidemic in Korea
    Hachung Yoon, Ilseob Lee, Hyeonjeong Kang, Kyung-Sook Kim, Eunesub Lee, Mathilde Richard
    PLOS ONE.2022; 17(6): e0269311.     CrossRef
  • Artificial Intelligence Models for Zoonotic Pathogens: A Survey
    Nisha Pillai, Mahalingam Ramkumar, Bindu Nanduri
    Microorganisms.2022; 10(10): 1911.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Risk Assessment Program of Highly Pathogenic Avian Influenza with Deep Learning Algorithm
Osong Public Health Res Perspect. 2020;11(4):239-244.   Published online August 31, 2020
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
Risk Assessment Program of Highly Pathogenic Avian Influenza with Deep Learning Algorithm
Osong Public Health Res Perspect. 2020;11(4):239-244.   Published online August 31, 2020
Close

Figure

  • 0
  • 1
  • 2
Risk Assessment Program of Highly Pathogenic Avian Influenza with Deep Learning Algorithm
Image Image Image
Figure 1 Distribution of risks of HPAI for 23,174 evaluations at the farm level in association with 544 “outbreak” farms from January 2014 to April 2017, predicted by the selected multilayer perceptron model: Comparison by outbreak status of receiver farms. HPAI = highly pathogenic avian influenza.
Figure 2 Recall and precision of the risk predicted by the assessment model, according to a series of cut-off values.
Figure 3 Positive likelihood ratio and posterior probability estimated using risk predicted by the risk assessment model.
Risk Assessment Program of Highly Pathogenic Avian Influenza with Deep Learning Algorithm

Details of the explanatory variables included in the risk assessment model.

Group Label Variable type Dummy
Farm Source/receiver Binary No
Animal species Categorical Yes
Farming type Categorical Yes
Number of heads Continuous No
Geographical coordinates Continuous No
History of HPAI outbreaks Binary No

Environment Farm density Continuous No
Surface area Continuous No
Affiliation Binary Yes

Vehicles Purpose of operation Categorical Yes
Owner (Corporation/Individual) Binary Yes
Driver (Owner/Employee) Binary No

Visit records Frequency of visits on farms Integer No
Frequency of visits on livestock facilities Integer No
Time interval between the visits to the source and receiver farms Continuous No
Visits to any livestock facilities between the visits to the 2 farms Binary No
Interval between visit and outbreak dates Continuous No

Visit records and dangerous contacts generated by livestock-related vehicles during the epidemics of HPAI in Korea from 2014 to 2017.

Category of use for the vehicle Total visit records Dangerous contacts


No. of vehicles No. of visits No. of vehicles No. of visits Attack rate (%) by vehicles

25th 50th 75th
Feed 11,096 22,230 173 1,571 8.7 14.3 27.3

Live animal 9,110 12,464 140 590 6.7 10.5 22.4

Egg 916 2,549 61 446 33.3 50.0 66.7

Husk 976 1,727 29 209 16.7 25.0 40.0

Consultant 914 1,382 14 79 16.9 28.9 38.3

Veterinary pharmaceuticals 740 1,186 6 35 6.6 10.8 12.3

Livestock compost 343 984 10 159 18.6 50.0 72.9

Manure 280 731 4 105 21.3 29.2 47.9

Veterinarian 434 636 4 13 12.7 17.5 20.6

Others 9,534 14,137 1* 1 - 9.1 -

Total 34,343 58,026 442 3,208 9.1 17.2 36.4

*Livestock machine repairman.

Supplementary Table 1 Details of the explanatory variables included in the risk assessment model.
Table 1 Visit records and dangerous contacts generated by livestock-related vehicles during the epidemics of HPAI in Korea from 2014 to 2017.

Livestock machine repairman.