It’s been roughly a year since Hurricane Maria — the tenth-most intense Atlantic hurricane on record — devastated the island of Puerto Rico. The storm killed more than 2,900 people, left millions of homes and businesses without power, and caused an estimated $ 8 billion in damage. It’s the worst natural disaster on record in Puerto Rico’s history.
San Juan-based Wovenware, a 15-year-old software engineering company, was on the front lines. CEO and founder Christian G. González described the recovery effort as “ongoing.”
“We are still living with its effects,” he said, “from an unstable power grid that needs to be redesigned from the ground up to a collective sense of PTSD that compels us to buy excessive amounts of water and fuel whenever a new storm forms in the Atlantic Ocean.”
Another, less-reported consequence of Hurricane Maria is an explosion of disease-carrying mosquitoes brought on by stagnant water. (The mosquitoes themselves don’t cause disease; rather, they pick up diseases from infected blood and spread them through bites.) Two years ago, Puerto Rico registered 38,058 confirmed cases of Zika, dengue, and chikungunya.
The Puerto Rico Science, Technology and Research Trust — a San Juan-based nonprofit organization that aims to foster growth in the island’s tech sector — in 2016 won a $ 50 million grant from the Centers for Disease Control and Prevention (CDC) to combat the spread of mosquitos on the island, with the goal of learning why a number of species have developed immunities to FDA-approved adulticides and larvicides.
But monitoring, testing, and labeling the more than 40 different species in Puerto Rico can be laborious. Currently, research scientists spend weeks capturing and classifying thousands of mosquitoes across difficult terrain.
That’s why Wovenware tapped artificial intelligence (AI) to help.
In early August, the company partnered with the Research Trust to develop a machine learning system that can automate the classification of Aedes aegypti, a specious known to carry infectious diseases. Its small team of data scientists are putting together a dataset of mosquitoes images and labels that will be used to train a computer vision algorithm.
The goal is to deploy a system within six months, González said.
“This work has serious implications for controlling the spread of mosquito-borne illness across the nation,” he said. “[It] also underscores the impact AI-based technology can have when bright minds are augmented by smart technology to help solve some of the world’s most pressing challenges.”
Wovenware’s work builds on earlier efforts.
In 2016, Verily, the health-focused subsidiary of Google parent company Alphabet, pulled back the covers on the Debug Project, which taps AI algorithms to quickly distinguish between male and female mosquitoes of Aedes aegypti. (Male mosquitoes subsist on plant nectar rather than blood.) And in 2017, entomologists at Stanford University built an AI-powered, sound-sensing smartphone app that can tell mosquito species apart by the noise produced by their wings.
“While we often hear about the use of AI in the business world – within insurance firms, banks, ecommerce, and manufacturing,” González said, “[the] possibilities of AI beyond the office walls or factory floor [are endless].”