According to the National Oceanic and Atmospheric Administration, aquaculture in the United States represents a $1.5 billion-a-year industry. Like land-based agriculture, shellfish aquaculture requires robust seed production to maintain a sustainable industry. Aquaculture hatchery production of crustacean larvae – seeds – requires close monitoring to track mortality rates and assess health status from the earliest stages of life.
Careful observation is necessary to plan production, determine the impact of naturally occurring harmful bacteria, and ensure sustainable seed production. This is an necessary step for shellfish hatcheries, but is currently a time-consuming manual process, prone to human error.
With funding from the Abdul Latif Jameel Water and Food Systems Lab (J-WAFS), MIT Sea Grant is working with Associate Professor Otto Cordero of MIT’s Department of Civil and Environmental Engineering, Professor Taskin Padir and Research Scientist Mark Zolotas of the Northeastern University Institute for Experiential Robotics, and others at the Aquaculture Research Corporation (ARC) and the Cape Cod Commercial Fishermen’s Alliance to develop the technology for the aquaculture industry. Based on Cape Cod, ARC is a leading shellfish hatchery, grower, and wholesaler that plays a key role in providing high-quality shellfish seed to local and regional farmers.
Two MIT students joined in on the effort this semester, working with Robert Vincent, associate director of advising services at MIT Sea Grant, through the Undergraduate Research Opportunities Program (UROP).
First-year student Unyime Usua and sophomore Santiago Borrego are using microscopic images of crustacean seeds from ARC to train machine learning algorithms that will lend a hand automate the identification and counting process. The resulting user-friendly image recognition tool is intended to lend a hand aquaculture industries differentiate and count robust, unhealthy and dead crustacean larvae, improving accuracy and reducing time and effort.
Vincent explains that artificial intelligence is a powerful tool in environmental science that enables researchers, industry and resource managers to address challenges that have long been key points in true data collection, analysis, prediction and process improvement. “Financial support from programs like J-WAFS allows us to tackle these issues head-on,” he says.
ARC faces challenges in manually quantifying larval classes, which is an crucial step in the seed production process. “While the larvae are growing, their size is constantly measured and counted,” explains Cheryl James, ARC larval/juvenile production manager. “This process is critical to encouraging optimal growth and strengthening the population.”
Developing an automated identification and counting system will lend a hand streamline this stage of the production process, saving time and costs. “This is not an easy task,” says Vincent, “but thanks to the guidance of Dr. Zolotas from Northeastern University’s Institute for Experimental Robotics and the work of UROP students, we have made significant progress.”
Both scientists and students benefit from the UROP program. Engaging MIT UROP students in developing these types of systems provides insight into applications of artificial intelligence they may not have considered, providing opportunities to discover, learn, and apply themselves while contributing to solving real challenges.
Borrego saw this project as an opportunity to apply what he learned in Class 6.390 (Introduction to Machine Learning) to real-world problems. “I started imagining how computers could look at images and extract information from them,” he says. “I wanted to explore it further.”
Usua decided to continue with the project because of the direct impact it could have on the industry. “I’m very interested in how we can use machine learning to make people’s lives easier. We are using artificial intelligence to help biologists facilitate the counting and identification process.” Although Usua was unfamiliar with aquaculture before embarking on this project, she explains, “Hearing about the hatcheries that Dr. Vincent told us about, it was unfortunate that not many people knew what was going on and what problems they were facing.” to face again.”
On Cape Cod alone, aquaculture is an $18 million-a-year industry. However, the Massachusetts Division of Marine Fisheries estimates that hatcheries can only meet 70 to 80 percent of seed demand annually, impacting local growers and the economy. With this project, partners aim to develop technology that will boost seed production, expand industry capabilities, and lend a hand understand and improve the hatchery microbiome.
Borrego explains the initial challenge of having constrained data to work with. “At first we had to go through and label all the data, but this process helped me learn a lot.” In true MIT fashion, he shares his impressions of the project: “Try to make the best of what you have at your disposal with the data you have to work with. You will need to adapt and change your strategies depending on what you have.
Usua describes her experiences during the research process, communicating within the team, and deciding what approach to take. “Research is a difficult and long process, but you can gain a lot from it because it teaches you to look for things on your own and find your own solutions to problems.”
In addition to increasing seed production and reducing human labor required in the hatching process, collaborators expect this project will result in cost savings and technology integration to support one of the most underserved industries in the United States.
Both Borrego and Usua plan to continue working for a second semester at MIT Sea Grant. Borrego wants to learn more about how technology can be used to protect the environment and wildlife. Usua says he hopes to explore more aquaculture projects. “It seems like there are an infinite number of ways to solve these problems.”