Format

Send to

Choose Destination
JMIR Public Health Surveill. 2018 Dec 21;4(4):e11361. doi: 10.2196/11361.

Real Time Influenza Monitoring Using Hospital Big Data in Combination with Machine Learning Methods: Comparison Study.

Author information

1
Laboratoire Traitement du Signal et de l'Image, Université de Rennes 1, Rennes, France.
2
INSERM, U1099, Rennes, France.
3
Centre d'Investigation Clinique de Rennes, Université de Rennes 1, Rennes, France.
4
Centre Hospitalier Universitaire de Rennes, Centre de Données Cliniques, Rennes, France.
5
Comprehensive Cancer Regional Center, Eugene Marquis, Rennes, France.
6
Centre d'Etudes et de Recherche en Informatique Médicale EA2694, Université de Lille, Lille, France.
7
Public Health Department, Centre Hospitalier Régional Universitaire de Lille, Lille, France.

Abstract

BACKGROUND:

Traditional surveillance systems produce estimates of influenza-like illness (ILI) incidence rates, but with 1- to 3-week delay. Accurate real-time monitoring systems for influenza outbreaks could be useful for making public health decisions. Several studies have investigated the possibility of using internet users' activity data and different statistical models to predict influenza epidemics in near real time. However, very few studies have investigated hospital big data.

OBJECTIVE:

Here, we compared internet and electronic health records (EHRs) data and different statistical models to identify the best approach (data type and statistical model) for ILI estimates in real time.

METHODS:

We used Google data for internet data and the clinical data warehouse eHOP, which included all EHRs from Rennes University Hospital (France), for hospital data. We compared 3 statistical models-random forest, elastic net, and support vector machine (SVM).

RESULTS:

For national ILI incidence rate, the best correlation was 0.98 and the mean squared error (MSE) was 866 obtained with hospital data and the SVM model. For the Brittany region, the best correlation was 0.923 and MSE was 2364 obtained with hospital data and the SVM model.

CONCLUSIONS:

We found that EHR data together with historical epidemiological information (French Sentinelles network) allowed for accurately predicting ILI incidence rates for the entire France as well as for the Brittany region and outperformed the internet data whatever was the statistical model used. Moreover, the performance of the two statistical models, elastic net and SVM, was comparable.

KEYWORDS:

Sentinelles network; big data; electronic health records; influenza; infodemiology; infoveillance; machine learning

Supplemental Content

Full text links

Icon for JMIR Publications Icon for PubMed Central
Loading ...
Support Center