Drones record how the environment shapes disease risk

Drones – or “unmanned aerial vehicles” – aren’t just for the military. They are already becoming important components of the ecologist’s toolkit, since they can collect lots of data at relatively low cost. Because of those two features, drones can be used to look at how landscapes change over time, by repeatedly mapping the same areas as frequently as a researcher would like.

In a recent issue of the journal Trends in Parasitology, a group of researchers from the London School of Hygiene and Tropical Medicine, the Menzies School of Health Research in Malaysia, the Queen Elizabeth Hospital in Malaysia, the Department of Health in Sabah, Malaysia, and the Department of Health in Muntinlupa City, Philippines argue that drones are particularly useful for understanding the complex interaction between environment and the spread of infectious zoonotic diseases, a prescient argument given the recent Ebola outbreak in West Africa.

Could drones be used for public health? For more practical applications, they already have been. After Typhoon Haiyan slammed into the Philippines, aid organizations used drones to plan their relief measures and to survey the extent of the storm’s damage. Some drones have even been used to deliver medical supplies to inaccessible areas, or to transfer biological samples from rural clinics to better-equipped laboratories in larger cities for analysis.

When it comes to epidemiological research, drones are also promising. One of the most exciting possibilities for using drones is for understanding and predicting the spread of infectious diseases. Changes in land use, such as deforestation or the expansion of agriculture, have already been documented as “major driver[s] of infectious disease emergence and spread.”

By disrupting existing ecosystems, these anthropogenic changes can alter the way that zoonotic and vector-borne diseases like malaria or ebola spread. In order to more accurately predict disease risk and to plan measures to control the spread of those diseases, researchers need to better understand how the interplay of human behavior, animal behavior, and the changing landscapes in which they coexist.

Image acquisition by traditional satellite technology is usually too slow to document rapid changes in land use, like wiping out a forest for agriculture. That’s where drones come in.

To demonstrate the utility of drones for understanding the relationship between epidemiology and ecology, the researchers, led by Kimberly M. Fornace, turned to a pair of study sites in Sabah, Malaysia, and a third site in Palawan, Philippines. They conducted a total of 158 drone flights in an effort to characterize the environmental risk factors for human infection by a parasite, Plasmodium knowlesi, which causes malaria. Theoretically, deforestation would result in increased contact between people, the mosquitos that carry that parasite, and their wildlife reservoirs, long-tailed macaques and pig-tailed macaques. To understand how that occurred, the researchers needed “detailed spatial information to integrate human and macaque movement…to understand the epidemiology of infection.”

They used a $25,000 commercially available drone called the Sensefly eBee, which can fly for up to fifty minutes at a time. (Drones range from several hundred dollars to tens of thousands; this particular model was chosen because the high cost was offset by the team’s ability to use it “without significant training or technical knowledge, allowing multiple members of the project team to be trained in operating [it].”) The eBee is equipped with a 16-megapixel camera to record aerial images, plus a GPS tracker. Previews of images are generated in real time, and full data processing from a single flight can be completed in a few hours. Eighty percent of the flights returned usable data. The main reasons for a failed flight were battery failure and weather conditions (high winds or rain). The researchers then combined their aerial imagery with GPS data on locations of households and malaria cases. That information allowed them to make better predictions as to disease risk and the spread of infection.

drone monkey

The drone proved more useful than satellite imagine for several reasons. Clouds often blanket tropical forests like the ones surveyed by the researchers, making satellites useless. And Google Earth imagery does not update fast enough for researchers to monitor changes in land use in real time. Compared to satellites, drones also allow users to have more control over the spatial resolution of resulting images, simply by adjusting the flight altitude. Most importantly, however, the drones allowed researchers to rapidly re-map areas as they were changing. One of the Malaysian sites was undergoing deforestation to make way for a rubber plantation, so researchers were able to document those changes as they were occurring.

On the other hand, drones weren’t perfect. They were quite finicky when it came to bad weather, and high temperatures caused them to overheat, something that’s obviously problematic in the tropics. Because weather is variable and unpredictable, it was difficult for the researchers to design their flight plans far ahead of time. The number of flights that could be conducted per day was limited by the availability of electricity, since the drone’s batteries need frequent recharging between flights.

Taken together, Fornace and her colleagues ultimately concluded that drones can’t replace more traditional forms of aerial mapping entirely. Instead, they fill an important niche. The use of drones, they write, “is most appropriate when detailed maps of relatively small geographical areas are needed in areas which high-resolution satellite data are not readily available,” or when “data are needed either in real time or at very frequent time points.” – Jason G. Goldman | 24 October 2014

Source: Fornace K.M., Drakeley C.J., William T., Espino F. & Cox J. (2014). Mapping infectious disease landscapes: unmanned aerial vehicles and epidemiology, Trends in Parasitology, DOI: 10.1016/j.pt.2014.09.001

Header image: Setting up the Sensefly eBee before a flight, via Fornace et al.