AI Planet- While artificial intelligence (AI) has been known for providing value to a variety of industries, such as health care, manufacturing and finance, it also can play a major role in helping to protect our planet. Environmental projects may not be as well-known as business applications, but there are a number of them underway across the globe that address such issues as preserving wildlife, preventing forest fires protecting marine life and providing a better planet for humans.
In fact, we’ve been fortunate to participate in an important environmental and health project right here in our own backyard: helping to prevent mosquito-borne illnesses in Puerto Rico. After Hurricane Maria last year, there was growing concern about the potential rise of diseases, including Zika, Dengue and Chikungunya, which are spread by a particular species of mosquito that thrives in water-soaked environments. The CDC provided a $50 million grant to the Puerto Rico Science, Technology and Research Trust to help stop the spread of these diseases and learn why many mosquitoes have become immune to FDA-approved insecticides.
To gain this insight, the Trust’s Puerto Rico Vector Control Unit (PRVCU) was tasked with capturing and classifying thousands of mosquitoes, across different areas in Puerto Rico — an extremely time-consuming process that requires specialized skills. We are working closely with the organization to develop an AI-based solution that automates the counting of mosquito eggs present in a given trap and the identification and classification of the mosquitoes.
The first step of identifying and labeling thousands of images of mosquitoes and datasets took place over a period of three to six months. Following that, these images and data were used to develop an algorithm and trained it to classify the species. The AI solution is expected to save the organization months of time so that it can more quickly discover the cause for insecticide resistance and find ways to stop and prevent the spread of diseases.
It’s very rewarding to apply our AI expertise to help address issues affecting the general population and the environment. But this is just one example of how AI is being leveraged to solve these types of problems. Others are doing fascinating things around the world:
• Monitoring wildlife: Microsoft’s AI for Earth awards grants to help organizations tackle pressing environmental issues. The Snow Leopard Trust, which received a grant through the program, is using AI to automate the monitoring and counting of snow leopards in the wild.
• Using AI to predict where forest fires might erupt: The University of Alberta is working with a doctoral student at the University of Oklahoma to develop an AI program that can help identify forests that are potentially vulnerable to fires. By using historical meteorological data and an artificial neural network (based on the way the brain and nervous system work), fire officials in Canada will be able to predict where extreme weather conditions could result in forest fires. The AI algorithms that are developed will include extensive weather information, among other data, to identify areas that are most at risk.
• Addressing climate change: Using AI and data from NASA, scientists are able to identify patterns and monitor changes occurring on land surfaces to determine potential risk. AI is also helping researchers uncover new approaches to tackle pollution and other contributors to climate change.
• Fighting wildlife poachers: According to scientists from the Digital Geography Lab at the University of Helsinki, AI could be used to help environmental groups and law enforcement monitor illegal wildlife trafficking to prevent these practices and better enforce government bans. Machine learning could be used to search social media for images of people participating in illegal wildlife trade, such as showing off rhino horns on their social platforms, and it could then identify the context through images, videos and text.