DSP/ARP/Service Applications & Valorisation
A Conceptual Approach of Tele‐epidemiology
Applied to the Rift Valley Fever in Senegal
C. Vignolles
ISPRS Commission VIII, Working Group 2 on Health
Symposium of Advances in Geospatial Technologies for Health
Santa Fe, NM, 12‐13 September 2011
Telehealth
Space technology for health
1 ‐ Improving access to healthcare
Treating patients at remote and mobile sites
2 ‐ Environment / Climate / Health
Monitor, predict and prevent epidemics
3 ‐ Crisis Management
Tele‐epidemiology consists in monitoring and studying the Better management of major humanitarian crises
propagation of human and animal diseases (water, air and vector borne diseases) which are closely linked to climate and environmental changes, based on space technology.
4 ‐ Education and Training
The French Space Agency (CNES) has thus developed, with its partners, a concept based on a deterministic approach of the Improving healthcare and learning thanks to Space
climate‐environment‐health relationships and on an original and really adapted space offer.
The tele‐epidemiology conceptual approach
Multidisciplinary approach based upon the study of the key mechanisms favoring emergence and propagation of infectious diseases linking disciplines like
environmental sciences, epidemiology, climatology, entomology, hydrology, microbiology…
The analysis of those processes is a key step in the development of new and original risk mapping using space technology
1‐ Experimental design mainly field studies • Observing strategy: monitoring and assembling multidisciplinary in‐situ datasets
• Diagnostic: extract and identify the main physical and biological mechanisms at stake
2‐ Obtaining well adapted products from Space
• Remote‐sensing monitoring of environment, linking epidemics with confounding factors
• Remote‐sensing from space: use of products, fully adapted to the various spatio‐temporal scales of variability
3‐ and dedicating modeling
•Modeling towards an operational Early Warning System (EWS);
•Validating/testing EWS
Projects in progress
•
Concept currently applied to different infectious diseases :
o Malaria in urban areas: Puerto Iguazu (Argentina), Dakar (Senegal), Bamako (Mali),
Ndjamena (Tchad)
o Malaria in rural areas in Burkina Faso
o Bilharzia in China
o Vibrio in the Mediterranean basin
o Dengue in Argentina and in Martinique Island
•
Elaboration of an operational system D RVF in Senegal
Marocco, Algeria, Tunisia
Mali
Tchad
Senegal
China
Argentina
Burkina Faso
Tele‐epidemiological conceptual approach
applied to
Rift Valley Fever (RVF) Monitoring in Senegal DIREL/DSV
Rationale: The Rift Valley Fever (RVF)
¾ The Rift Valley Fever (RVF) is an arthropod‐borne viral disease, ¾ Found essentially in Africa
¾ Primarily spread amongst domestic animals by the bites of infected mosquitoes, especially Aedes vexans
and Culex poicilipes Ö The abundance of such RVF vectors is directly linked to ponds’ dynamics and their vegetation cover and turbidity degree. The ponds’ dynamics is associated with the spatio‐temporal variability of rainfall events.
¾ RVF virus could also infect humans
¾ RVF is causing epizootics of spontaneous abortion and high mortality rate for domestic animals
BRVF can cause very serious economic losses in livestock.
Countries with endemic disease and substantial outbreaks of RVF
Gambia, Senegal, Mauritania, Namibia, South Africa, Mozambique,
Zimbabwe, Zambia, Kenya, Sudan, Egypt, Madagascar, Saudi
Arabia, Yemen
CDC source
Countries known to have some cases, periodic isolation of virus, or
serologic evidence of RVF :
Botswana, Angola, Democratic Republic of the Congo, Gabon, Congo,
Cameroon, Nigeria, Central African Republic, Chad, Niger, Burkina Faso,
Mali, Guinea, Tanzania, Malawi, Uganda, Ethiopia, Somalia
The studied area
Studied area centered around Barkedji
in the Ferlo region in Senegal
Sahelian Climate Rainfall 300 to 400 mm in 4 or 5 months (July to November)
Understanding mechanisms at stake From rainfall event to vectors’ aggressiveness
„ Productive rainfall for Aedes
Non productive rainfall events
2003 TRMM daily rainfall data
RVF mosquitoes
Aedes
Culex
Aedes vexans (%) versus pond distance (meter)
100%
Mosquitoes Flying range
Cumulative frequency (%)
90%
80%
70%
60%
Photo by Jacques-André Ndione, 2006
50%
40%
30%
20%
10%
0%
0
100
200
300
400
500
Pond distance (in meter)
Adapted from Bâ et al 2005
Photo by Jacques-André Ndione, 2006
Ndiaye et al., 2006
A Remote‐sensing tool
applied to Rift Valley Fever (RVF) Monitoring
Analyses and processing of high‐spatial resolution satellite images (SPOT 5, 10m) Computation of ponds’ area, their vegetation cover and turbidity
Evaluation of Zones Potentially Occupied by Mosquitoes (ZPOM)
Multi‐spectral SPOT 5 Image (high‐spatial resolution ‐10m)
43 km
Map of Senegal –
African Atlas (Jaguar Edit. )
Barkedji pond
Studied area : Ferlo region in Senegal
On the 26/08/2003
Studied area: 198 337 ha
Number of ponds: 1354
Ponds’ area: 1703 ha
Ponds’ percentage: 0,9%
© CNES 2003, Distribution Spot Image
SA
46 km
A Remote‐sensing tool
applied to Rift Valley Fever (RVF) Monitoring
Developing brand‐new indices by combining various wavelengths
Ponds detection
NDPI f(MIR, Green)
Ponds characteristics
% Vegetation
NDVI f(NIR, Red)
% Turbidity
NDTI f(Red, Green)
A Remote‐sensing tool
applied to Rift Valley Fever (RVF) Monitoring
Spot 5, multi‐spectral high‐spatial resolution (10‐m) August 26th, 2003 (during the rainy season)
Pond south‐west Barkedji
15 ha (peak of rainy season)
55% covers by vegetation
45% free water
1.4 km
1
Photo by Jacques-André Ndione, 2006
1‐ pond vegetation
2
© CNES 2003, Distribution Spot Image SA
1.4 km
False color composite
The new Normalized Difference Pond Index or:
NDPI = ( MIR‐Green)/ (MIR+Green)
Photo by Jacques-André Ndione, 2006
2‐sahelian savanna
A Remote‐sensing tool
applied to Rift Valley Fever (RVF) Monitoring
Identify environmental factors of Aedes & Culex presence by remote sensing to obtain risk map
SPOT5 multispectral Image
high spatial resolution -10m
(Program ISIS/CNES)
Vegetation
activity
gradient
Niaka
43 km
1354 ponds
Ponds detection
NDPI
Characterization
NDVI & NDTI
Ponds’ area
Ponds’ characterization
Turbidity
gradient
26/08/2003
© CNES 2003, Distribution
46 km Spot Image SA
Barkedji
Zones Potentially
Occupied by Mosquitoes (ZPOM)
Vegetation Cover
Turbidity
Mosquitoes flying range
~500m (Bâ et al., 2005)
Mapping the ZPOM
© CNES/OMP product, CNES 2003, Distribution Spot Image
26/08/2003
Ponds ~ 1%
ZPOM = 25%
Analyses and processing of high‐spatial resolution images (SPOT 5, 10m) allows to detect
ponds’ area like their vegetation cover and turbidity and finally evaluate Zone Potentially
Occupied by Mosquitoes (ZPOM)
The integrated conceptual approach for RVF In situ data ‐> main mechanisms
Remote sensing data
¾ Entomology
¾ Ponds’ detection (Spot 5)
¾ Rainfall (TRMM)
¾ Parks localization (Quickbird & Spot 5)
¾ Free water area (TerraSarX)
• aggressiveness
• embryogenesis
• flying range
¾ Breeding sites and ponds’ dynamics
¾ Contact host (cattle) vectors (mosquitoes)
¾ ZPOM
¾ Productive rainfall
¾ Dynamics of ponds
Hazard map
Dynamic ZPOM
(daily, 10m)
Vulnerability map
Cattle parks localization
Risk map
Environmental Risk
From remote sensing to risk
Some parks …
© CNES 2003, Distribution Spot Image SA
Ponds
ZPOM
Photo by Jacques-André Ndione, 2006
Parks
ZPOM
Photo by Jacques-André Ndione, 2008
Hazard
Vulnerability
Risk
Photo by Jacques-André Ndione, 2008
Photo by Jacques-André Ndione, 2008
© CNES/OMP product, CNES 2003, Distribution Spot Image
@Emercase, 2003
Results 2010
31st July 2010: productive rainfall
Agressiveness of Aedes
61,3mm from 31st July to 15 August 2010
Hazard : Potential Presence of Aedes
Aucun
Faible
Moyen
Fort
Très fort
13/08/10
12/08/10
10/08/10
16/08/10
9/08/10
8/08/10
7/08/10
6/08/10
5/08/10
4/08/10
3/08/10
11/08/10
15/08/10
14/08/10
Aggressiveness' of Aedes vexans
Studied area (15km*15km) centered on Barkedji
Conclusion
¾ Development of an integrated methodological approach to obtain environmental risk
mapping based on space technology
¾ Concept currently applied to different infectious diseases at different places
¾ Towards the elaboration of operational Early Warning Systems
¾ This conceptual approach is meant to be applied to other diseases in other places with
different environments
Contacts & informations
Telehealth CNES : www.cnes.fr/telesante & http://RedGems.org
• Lacaux J‐P, Tourre Y. M., Vignolles C., Ndione J‐A, and M. Lafaye, 2007: Classification of Ponds from High‐Spatial Resolution Remote Sensing: Application to Rift Valley Fever Epidemics in Senegal, Remote Sensing of Environment, Elsevier Publishers 106 66–74
• Tourre Y.M., Lacaux J‐P, Vignolles C., Ndione J‐A, and M. Lafaye, 2008: Mapping of zones potentially occupied by Aedes vexans and Culex poicilipes mosquitoes, the main vectors of Rift Valley Fever in Senegal. Geospatial Health, 3, 69‐79
• Vignolles C., J‐P. Lacaux, Y. M. Tourre, G. Bigeard, J‐A. Ndione, and M. Lafaye, 2009: Rift Valley Fever (RVF) in a Zone Potentially Occupied by Aedes vexans in Senegal: Dynamics, Mapping and Risks, Geospatial Health 3(2), 211‐220
• Tourre, Y.M., J‐P. Lacaux, C. Vignolles, and M. Lafaye, (2009): Climate impacts on environmental risks evaluated from space: The case of the Rift Valley Fever and its conceptual approach, Global Health Action, DOI: 10.3402/gha.v2i0.2053 • Dambach, P., A. Sie, J‐P. Lacaux, C. Vignolles, V. Machault, and R.Sauerborn, (2009): Using high spatial resolution remote sensing for risk mapping of malaria occurrence in the Nouna district, Burkina Faso, Global Health Action, DOI:10.3402/gha.v2i0.2094
• Vignolles, C., Y. M. Tourre, O. Mora, L. Imanache and M. Lafaye, (2010): TerraSAR‐X high‐resolution radar remote sensing: an operational warning system for Rift Valley fever risk, Geospatial Health, 5(1), 23‐31