I recently had the opportunity to speak at the O’Reilly Artificial Intelligence Conference in Silicon Valley, sharing my experiences with environmental AI, and more specifically, how we’re applying it to help wipe out mosquito-borne diseases. I was honored to join the ranks of speakers from world-class organizations, such as Google, UC Berkley, Bloomberg and others, to discuss how AI is really changing the world.
During my presentation, I discussed our work for the Puerto Rico Science, Technology & Research Trust to identify and classify mosquitoes that may be carrying diseases such as Zika, Dengue and Chikungunya.
It probably came as a big surprise to many of the session attendees that the world’s most deadly animal is the mosquito – and, they’ve killed 32 times as many people as every war in human history combined. They also may not have known that the Zika virus is behind an unprecedented rise in the number of children being born with microcephaly, or that there are 390 million Dengue infections each year.
As we know all too well here in Puerto Rico thanks to Hurricanes Maria and Irma, stagnant water and weather events can make the mosquito population explode. During the conference I shared how we at Wovenware are working to apply AI to automate much of the work involved in classifying mosquitoes, so entomologists can concentrate on research such as why mosquitoes have become immune to insecticides.
Wovenware has developed an RPA solution composed of deep learning models and other processes that’s designed to automate the identification and classification of mosquito species, reducing the amount of manual processes previously required of entomologists. This automation is making it faster to count, classify and identify the species and gender of Aedes Aegypti captured mosquitoes.
I shared with the group the architectures and technologies we’ve employed to build the neural networks, including RetinaNet, ResNet and TensorFlow.
As I mentioned to attendees, we’re still in the early stages of the project, but future plans call for us to integrate to other related projects, analyze time series and remotely monitor and classify species.
As we continue to refine and evolve our solution, we hope to take our work to other countries and geographies where mosquitoes are a major issue.
In addition to the incredible response I received from my presentation, the O’Reilly Artificial Intelligence Conference truly was inspirational. Listening to other presentations and speaking with data scientists, as well as academics and business leaders was a confirmation that amazing discoveries, technologies and use cases continue to evolve around AI, and we’re really just at the tip of the iceberg in terms of what it is capable of achieving. It’s clear that AI is moving beyond the business world and beginning to solve environmental problems, while making the world a safer place.