eHealth 2011, Málaga, 21-23 November 2011
1
Event-based surveillance with MedISys
Tracking Media Reports on the Shiga toxin-producing
Escherichia coli O104:H4 outbreak in Germany
4th ICST International Conference on eHealth
21-23 November 2011
Málaga, Spain
Erik van der Goot & the OPTIMA team (OPensource Text Information Mining and Analysis )
European Commission – Joint Research Centre (JRC)
Institute for the Protection and Security of the Citizen (IPSC)
erik.van-der-goot@jrc.ec.europa.eu
JRC - who
eHealth 2011, Málaga, 21-23 November 2011
Joint Research Centre (JRC)
The European Commission’s
Research-Based Policy Support Organisation
IPSC - Institute for the Protection and Security of the Citizen
Ispra - Italy
http://ipsc.jrc.ec.europa.eu/
http://www.jrc.ec.europa.eu/
2
JRC - where
eHealth 2011, Málaga, 21-23 November 2011
3
MedISys - Overview
eHealth 2011, Málaga, 21-23 November 2011
Objective:
Provide open source data collection and analysis for surveillance and epidemiology
Replace manual scanning of multiple newspapers and web portals
Support national and international Public Health organisations to monitor issues of Public
Health concern (infectious diseases, CBRN threats, food & feed contaminations)
Functionality:
Gather, filter, classify, extract and aggregate health-related information
Monitor trends, detect breaking news
Visualise analysis results
Alert users
Allows customised views
In combination with NewsDesk tool, allows manual moderation.
4
Background - History
eHealth 2011, Málaga, 21-23 November 2011
Based on JRC’s Europe Media Monitor (EMM) technology
(EMM live since 2002; http://emm.newsbrief.eu).
On request / initiative of the EC’s Directorate General
for Health and Consumer Protection (DG SANCO).
Password-protected service for Public Health bodies since 2005.
Public service since early 2007
(http://medisys.newsbrief.eu, restricted functionality).
5
EMM/MediSys System Overview
eHealth 2011, Málaga, 21-23 November 2011
6
EMM Newsbrief
MediSys Newsbrief
EMM Open Source Monitoring Engine
NewsDesk Service (a.k.a. RNS)
Editorial Interface
MediSys Today
eHealth 2011, Málaga, 21-23 November 2011
7
Some statistics
Gathers approximately 100,000 new news articles per day
In >42 languages
From ~ 3100 news portals world-wide (roughly 6500 pages/RSS feeds), plus 20 news wires and
370 specialist sites
Classifies all news according to hundreds of subjects and countries.
(~1400 category definitions, ~50.000 key word patterns and pattern combinations)
24 / 7: Runs 24 hours per day, 7 days a week.
Started in 2002
90.000 hits per day for public MediSys
But more importantly, 4500 visitors/day on the public site
Developed, built, maintained and run by the JRC.
Importance of multilingual information gathering
Locations mentioned in MedISys medical articles across languages
eHealth 2011, Málaga, 21-23 November 2011
8
Italian - German
English - French
Spanish - Portuguese
Event-based surveillance & Epidemic intelligence
eHealth 2011, Málaga, 21-23 November 2011
Taken from:
Paquet, C., Coulombier, D., Kaiser, R., Ciotti, M.
Epidemic intelligence: a new framework for strengthening disease surveillance in Europe. Euro Surveill. 11, 665 (2006).
http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=665
9
eHealth 2011, Málaga, 21-23 November 2011
10
E. coli outbreak in May 2011
eHealth 2011, Málaga, 21-23 November 2011
Number of articles in the E. coli category per day (data from 1 May to 4 July 2011; all languages).
11
Timeline of early events
eHealth 2011, Málaga, 21-23 November 2011
•19 May: Robert-Koch-Institute was informed about a cluster of
three cases of HUS in Hamburg.
• 21 May: MedISys detected the first media report in the newspaper
Die Welt on Saturday 21 May at 12:14 CEST
• Several other articles followed the developing story in the afternoon.
• Altogether, 23 German news items triggered the MedISys E. coli
category, of which 22 were about the outbreak
(one irrelevant article was about the water quality in German lakes
in the region Oberbergischer Kreis).
• Furthermore, there were two news reports in Farsi and one report
in Mandarin; these reports were from the science pages
of Iranian and Chinese newspapers and were not related to the outbreak.
• Due to the sudden increase in media reports on E.coli on Saturday,
MedISys issued an automatic alert.
• 22 May: The Early Warning and Response System (EWRS) of the EU
received a first communication by the German authorities
on Sunday 22 May at 11:40.
12
eHealth 2011, Málaga, 21-23 November 2011
13
eHealth 2011, Málaga, 21-23 November 2011
14
Country of publication
eHealth 2011, Málaga, 21-23 November 2011
Number of articles in the E. coli category published in:
Luxembourg (LU), Sweden (SE), France (FR), Russia (RU), Spain (ES) and Germany (DE)
15
Main events
eHealth 2011, Málaga, 21-23 November 2011
•
•
•
•
•
•
•
•
sudden rise on articles in Germany (21-25 May), when the first cases
became public,
a peak with articles on a Swedish tourist group who got infected in Lower
Saxony (26 May).
the reporting on alleged E. coli contaminations in Spanish cucumber,
tomatoes and salad (peak on 27 May with 107 articles),
the reporting on the financial impact on Spanish farmers and the
announcement that Spanish cucumbers had
tested negative for E. coli
(peak on 31 May with 300 articles),
the discussion on trade restrictions in Russia for EU vegetable products
(peak on 2 June),
the announcements by German authorities that bean sprouts were the
source of infection (rise in volume on 5-10 June),
the E.coli cluster of cases in Bordeaux, France (peak on 16 June with 203
articles), and
the reporting on fenugreek seeds imported from Egypt in 2009 and 2010
(from end of June on).
16
Entity extraction
eHealth 2011, Málaga, 21-23 November 2011
Number of articles in the E.coli category that also mention one of the entities Robert Koch Institute, Commissioner John Dalli, European
Commission, ECDC and EFSA.
17
Conclusions: E.coli outbreak in May 2011
eHealth 2011, Málaga, 21-23 November 2011
18
Early media reports on the E.coli outbreak in May 2011 were
identified quickly and accurately; an alert was published on
Saturday 21st of May via MedISys.
Media monitoring was used to identify key aspects of the developing
story which were further disseminated, e.g. in reports for the
European Commission.
Since MedISys covers both general media and official sites, users
were able to analyze their reports in parallel.
NewsDesk Screenshot
eHealth 2011, Málaga, 21-23 November 2011
19
Feeds
(in a directory)
Articles
(in a feed)
Newsletter
MedISys – Current subscribers and users include …
eHealth 2011, Málaga, 21-23 November 2011
Supranational organisations
Directorate General Health and Consumer Protection (SANCO)
European Centre for Disease Control, Stockholm (ECDC)
European Food Safety Authority (EFSA)
World Health Organisation (WHO)
National Public Health organisations
EU27 public health authorities & ministries of health
Global Public Health Intelligence Network (Canada)
National Institute of Infectious diseases (Japan)
The (general?) public
Currently ~ 5000 visitors, ~ 100000 hits per day on public system
You?
http://medisys.newsbrief.eu
20