Progressive meets Society – Interview with Annamaria Conte

“In the coming decades, the greatest challenge will be moving from big data toward creating value. What we have from Earth Observation offers tremendous possibilities for creative Earth Observation applications, researches and commercial exploitation. But it is fundamental cooperation, dialogue, open data, multidisciplinarity and innovative minds. If each of us brings a piece of the puzzle in the right place, the final picture will be knowledge and benefit for everyone”

ANNAMARIA CONTE

The Earth Observation sector is evolving rapidly thanks to the ever-increasing availability of high-quality Earth Observation data both in terms of resolution and revisit time. This is creating opportunities for private and public entities that have recognized the potential of satellite observations to support their decision-making on a wide range of societal and environmental challenges.

We have interviewed Annamaria Conte, Head of Statistics and GIS Unit at the Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise “G. Caporale” (IZSAM), the public health institute which operates as a technical and scientific arm of the Italian Ministry of Health. Annamaria shared with us the experience of IZSAM with Earth Observation data and how their exploitation is adding value to the work at the Institute.

IZSAM has been working to safeguard the health of humans and animals for almost 80 years. Can you tell us more about the work you do?

The ongoing epidemic clearly shows us that human health is strictly linked to animal health and the ecosystem in which we live. Protecting the environment and animals, taking care of their health means preserving human health. And this is something that IZSAM has been always working on.

Let me start mentioning the present, what we have been doing since March 16 2020 during the COVID-19 epidemic. IZSAM has processed until today more than 60.000 human samples in support of the diagnostic workflow for SARS-CoV-2 of the Abruzzo and Lombardy regions. Moreover, IZSAM is engaged in the data analysis with the production of an online dashboard and a daily report. IZSAM is also promoting researches and studies to better elucidate the factors influencing the infection spread and carrying out epidemiological investigations. In addition, IZSAM is involved in the post lockdown serological surveillance of workers employed in various enterprises in the Abruzzo region.
In a very short time, IZSAM has made available its experts, its technologies, its resources to respond to an unknown virus epidemic. And this has been done thanks to our experience in working in emergency and following a multidisciplinary approach.

For over 30 years the IZSAM has understood that humans, animals and the ecosystem must be studied together, and this cannot be limited to human and animal doctors. It is necessary to base the knowledge on multidisciplinary teams. In IZSAM, in addition to the health figures of veterinarians and biologists, we have statisticians, mathematicians, GIS experts, computer scientists, engineers, bioinformaticians, trainers, social sciences experts, psychologists, and this means that health issues are addressed from every possible angle, approached transversely, satisfying the societal demand. All in full respect and harmony with what Europe asks us about research: analysis of problems, solutions, generate economy and improve citizens’ health.

This approach that has always characterized IZSAM, in recent years has led us to partnerships unthinkable until a few years ago (e.g. the European Space Agency, the Engineering Departments of the most prestigious Universities), driving IZSAM more and more towards innovation, towards the Big Data era, towards the Earth Observation and Artificial Intelligence fields.

The opportunity to collaborate with the most important Intergovernmental Agencies (the World Organisation for Animal Health – OIE, the Food and Agriculture Organization of the United Nations – FAO) brings our experts in many countries, in Africa, the Middle East, Eastern Europe, where they are asked to study problems and find solutions, while collecting fundamental field experiences on diseases that we do not have in Europe.

The Institute has been exploiting Earth Observation data in diverse initiatives. Can you share with us a couple of examples?

For several years, our Institute has been using EO data to study the influence of climatic and environmental factors on vector-borne diseases (VBD) and vector population dynamics. Most of the environmental variables (geographical, climatological, and hydrological) that influence the transmission cycle of VBDs between pathogen, vectors and intermediate hosts can be monitored efficiently from satellites capable of capturing these parameters frequently and on a global scale.
Knowing where conditions are favorable for the spread of the virus and for the abundance of vectors is of primary importance for directing surveillance activities. The Ministry of Health funds researches and surveillance plans for millions of euros each year and it is important to get the most in terms of knowledge from these programs. Combining the data collected in field with Earth Observation data and the most innovative and powerful big data analysis techniques, is the direction in which IZSAM is going.

Some examples? A research completed in 2018 (1) produced a classification of the Italian territory in similar ecoregions at 250m resolution. A multivariate statistical clustering algorithm was applied to seven climatic and environmental remote-sensed variables and the final ecoregion map (Figure 1) was adopted by some regions to choose targeted areas for West Nile Virus (WNV) surveillance. In case of entry of a new virus transmitted by vectors, knowing which area is similar to the ones in which the virus starts spreading, can better direct us in field research.

Figure 1

Another example is the project started on September 2019, funded by the European Space Agency (ESA) in the framework of the ESA EO Science for Society Permanently Open Call for Proposals – ‘Artificial Intelligence and Earth Observation Data: innovative methods for monitoring West Nile Disease (WND) spread in Italy’ – AIDEO. Its principal aim is to verify the feasibility of an automated process integrating past WND outbreaks, EO data and Artificial Intelligence algorithms to predict where and when WND cases could re-emerge and spread in Italy.

And Progressive is partner of this project, with its support in acquiring and processing satellite images – Sentinel 1, 2 and 3, Landsat 8 and SRTM (Figure 2). Such images are then analysed using Deep learning techniques by AImageLab, a research laboratory of the Engineering Department “Enzo Ferrari” at the University of Modena and Reggio Emilia, Italy, specialized in Computer Vision, Pattern Recognition and Artificial Intelligence.

Figure 2

This project is a perfect example of the multidisciplinary approach necessary to acquire knowledge and implement new solutions especially in the context of VBD.

Another application of EO data we are exploring relates to the Copernicus Emergency Management Service (Copernicus EMS) which provides information for emergency response in relation to different types of natural and man-made disasters. IZSAM manages the Italian National Animal Identification and Registration System and the Livestock Data Bank. The opportunity to have satellite data with timely and accurate geo-spatial information before, during and after disasters, is fundamental to support Veterinary Services in their prevention, preparedness, response and recovery activities.

Can you tell us what makes the use of satellite data beneficial for your projects?

The changing of climatic and ecological conditions, human behaviour, mobility, as well as the rapid and uncontrolled urbanization, are key factors that influence the seasonal and geographic distribution of vectors’ population and therefore the transmission of pathogens. Having the possibility to acquire data explaining and monitoring such factors with excellent spatial resolution (just think of the Sentinel 2’s 10 – 20 m) and such a frequent revisit time on a global level is a huge resource.
It gives the opportunity to find information on the territory that can guide field activities. Surveillance activities in field are extremely expensive, both from an economic and human resources point of view. Knowing where and when to go to collect samples would save resources and better allocate efforts.

What kind of role partnerships with Earth Observation experts do you expect in your field?

First of all, a clear and constant dialogue is needed to understand which datasets are produced, what characteristics they have and understand the potential use together. Then it is fundamental the support for manipulation and processing of the data to be associated with the health data to answer research questions. There is a real need to assemble datasets from multiple, disparate sources. Analysis of EO data is complicated: data is often at different spatial and temporal resolution, and datasets must be aggregated or disaggregated to harmonize the spatio-temporal scale. Specific technical skills, computational power and processing power are not always present in our sector. For this reason, I see the development of small and medium enterprises to support the Health System (human and animal) to assist epidemiologists in the collection and processing of EO data. Don’t forget that the ultimate goal in research is to generate knowledge, to translate abundant, diverse, and rapidly growing big data into meaningful insights for animal and human health. And if everyone does its part, the end goal is achieved faster.
In IZSAM we have in part the expertise to process and manipulate EO data, we have epidemiologists and disease experts, laboratory facilities, so we cover quite well what the scientific approach needs to meet the requirements of our Ministry of Health, but much more can be done with partnerships with EO experts.

Do you think that the Health sector is nowadays fully taking advantage of the opportunities coming from the Earth Observation sector?

I believe we are only at the beginning of a new era in which new products are available and new frontiers are opening up. Although for years remote sensed data has been used in modelling the spread of animal diseases there is still a lot to study and explore. The increasing availability and complexity of EO data has led to new opportunities and challenges in human and veterinary epidemiology. Products that have a spatial resolution of 10 m and a revisit time of 3-5 days inevitably need experts in the sector who support experts in the Health domain.

If you could turn a wish into a reality, what will an ideal health application based on EO data look like that would really give a boost to your work?

We expect a lot from the one we are already working on: setting up an early warning model based on EO data and AI algorithms which highlights when climatic and environmental conditions are favorable for the spread of a VBD. If the warning arrives early in advance, Veterinary and Public Health Authorities are able to intervene in field with deeper monitoring and control measures. Of course, being disease/pathogen/vector-specific applications, it is certainly not easy to have a single approach, a single model. But the availability of ‘ready to use’ data would already be a wonderful starting point to develop such systems.

Where do you see the society taking the most benefit from Earth Observation in 5/10 years from now?

There is a deluge of data now being returned by Earth observation (EO) satellites. The archives of the big space agencies and companies are filling up with petabytes of information. The future will be to use all this information in ever smarter ways, using machine learning, deep learning to extract even more relevant insights into the changes taking place on our Earth. The fields of application will be ever larger and ever more disparate: climate, energy, protection of the environment and natural resources. But definitely Health is in the list.
Last March 10, ESA organized an online and live streamed workshop “EO and AI for Health and Urban Resilience Workshop” with specific objectives of identifying the main challenges related to Global Health and Earth Observation at large, and to discuss with scientists, policy makers and industry the most pressing needs and information requirements. A clear sign of the openness that is being created between EO and Health.

For a close, is there anything else you would like to add?

In the coming decades, the greatest challenge will be moving from big data toward creating value. What we have from EO offers tremendous possibilities for creative EO applications, researches and commercial exploitation. But it is fundamental cooperation, dialogue, open data, multidisciplinarity and innovative minds. If each of us brings a piece of the puzzle in the right place, the final picture will be knowledge and benefit for everyone.

————————————————————————–

To learn more about Annamaria Conte

Annamaria Conte graduated in Statistics at ‘La Sapienza’ University in Rome in 1994. In 1995 she joined the Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise “G. Caporale” (IZSAM) – Teramo, Italy (www.izs.it), a Veterinary Public Health Institute serving the Italian Ministry of Health. Since 2009, she leads the Statistics and GIS Unit, coordinating and managing the activities required by the National and International Reference Centers operating at IZSAM (the National Reference Centre for Epidemiology, the OIE Collaborating Centre for Veterinary Training, Epidemiology, Food Safety and Animal Welfare and the FAO Reference Centre for Veterinary Epidemiology).

Her research focus is on epidemiological analytical methods and spatial epidemiology of major animal infectious diseases, including zoonoses, and on the identification of factors influencing the spread and persistence of vector borne diseases. She leads the development of several web-based GIS which provide near-real time surveillance and modelling tools to share and analyse animal disease data and information.

Recent interests are on the integration of remoted sensed data and deep learning methods to produce early warning systems for preventing and controlling, emerging and re-emerging infectious diseases.

She has an extensive training activity on basic and advanced statistical methods, Spatial Analysis and GIS for the European Union, Italian Ministry of Health, Universities, Veterinary Services and Veterinary Public Health Institution.
Over 100 publications on peer reviewed journal.

—————————————————————————–

1 – Ippoliti C, Candeloro L, Gilbert M, Goffredo M, Mancini G, Curci G, Falasca S, Tora S, Quaglia M, Conte A. (2019). Defining ecological regions in Italy based on a multivariate clustering approach: A first step towards a targeted vector borne disease surveillance. PLoS ONE 14(7): e0219072. https://doi.org/10.1371/journal.pone.0219072