Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: Journal of Medical Internet Research

Date Submitted: Dec 15, 2022
Date Accepted: Aug 4, 2023

The final, peer-reviewed published version of this preprint can be found here:

Influenza Epidemic Trend Surveillance and Prediction Based on Search Engine Data: Deep Learning Model Study

Yang L, Zhang T, Han X, Yang J, Sun Y, Ma L, Chen J, Li Y, Lai S, Li W, Feng L, Yang W

Influenza Epidemic Trend Surveillance and Prediction Based on Search Engine Data: Deep Learning Model Study

J Med Internet Res 2023;25:e45085

DOI: 10.2196/45085

PMID: 37847532

PMCID: 10618884

Influenza epidemic trend surveillance and prediction based on search engine data: a deep learning model study

  • LiuYang Yang; 
  • Ting Zhang; 
  • Xuan Han; 
  • Jiao Yang; 
  • Yanxia Sun; 
  • Libing Ma; 
  • Jialong Chen; 
  • Yanming Li; 
  • Shengjie Lai; 
  • Wei Li; 
  • Luzhao Feng; 
  • Weizhong Yang

ABSTRACT

Background:

When emerging of re-emerging infectious outbreak, reported case will not reflect the real epidemic curve, like COVID-19 outbreak in China. Influenza-like Illness (ILI) could take the role of surveillance. Once the correlation between internet search engine and symptomatic surveillance been identified, respiratory infectious like influenza, COVID-19 could develop a supplemented modern intelligence epidemic surveillance method.

Objective:

To find the correlation between the internet search engine and influenza epidemic, providing evidence for supplementing modern intelligence epidemic surveillance methods.

Methods:

We collected the daily proportion of ILI sentinel virological testing results and Baidu query data across the 31 provinces of mainland China from 1 January 2011 to 31 June 2018 with a total of 3,691,865 and 1,563,361 respective samples. We applied Pearson analysis and distributed non-linear lag models to clarify the correlation between selected Baidu search items and the daily influenza-positive rate. We applied a four-in-one deep learning model to identify the contribution of Baidu search information for seasonal influenza trend prediction.

Results:

The traditional disease surveillance systems should be complemented with information from online data sources with advanced technical support. Search engine information can help detect warning signs of influenza outbreaks earlier. However, supplemented search engine information may also reduce the effectiveness of the prediction of modern surveillance. Modern surveillance should be complemented with caution.

Conclusions:

The traditional disease surveillance systems should be complemented with information from online data sources with advanced technical support. Search engine information can detect warning signals of influenza outbreaks earlier, saving time for preparedness. However, it is worth highly noting that supplemented search engine information may also reduce the effectiveness of the prediction that modern surveillance. Modern surveillance should be complemented with caution rather than exaggerating the effect.


 Citation

Please cite as:

Yang L, Zhang T, Han X, Yang J, Sun Y, Ma L, Chen J, Li Y, Lai S, Li W, Feng L, Yang W

Influenza Epidemic Trend Surveillance and Prediction Based on Search Engine Data: Deep Learning Model Study

J Med Internet Res 2023;25:e45085

DOI: 10.2196/45085

PMID: 37847532

PMCID: 10618884

The author of this paper has made a PDF available, but requires the user to login, or create an account.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.