Technology advancements aim to boost aquaculture sector.

Reportedly, aquaculture in the United States is a thriving $1.5 billion industry annually as per the National Oceanic and Atmospheric Administration. Similar to traditional farming, successful shellfish aquaculture hinges on robust seed production to ensure industry sustainability. The production of shellfish larvae, also known as seeds, at aquaculture hatcheries necessitates rigorous monitoring to track mortality rates and evaluate health right from the larvae’s early life stages.

A detailed observation process is essential to guide production scheduling, analyze the impacts of naturally occurring harmful bacteria, and secure the continuity of sustainable seed production. While crucial for shellfish hatcheries, this monitoring is presently a laborious manual task that leaves room for human errors.

Thanks to funding from MIT’s Abdul Latif Jameel Water and Food Systems Lab (J-WAFS), MIT Sea Grant is collaborating with Associate Professor Otto Cordero of the MIT Department of Civil and Environmental Engineering, Professor Taskin Padir, Research Scientist Mark Zolotas at Northeastern University Institute for Experiential Robotics, and several experts at Aquaculture Research Corporation (ARC) and the Cape Cod Commercial Fishermen’s Alliance to drive technological advancements in the aquaculture sector. Positioned in Cape Cod, ARC is a key player in the shellfish industry, serving as a prominent hatchery, farm, and wholesaler that provides top-notch shellfish seed to regional and local growers.

This semester, two MIT students have joined forces in this initiative, collaborating with Robert Vincent, the assistant director of advisory services at MIT Sea Grant, through the Undergraduate Research Opportunities Program (UROP).

Unyime Usua, a first-year student, and Santiago Borrego, a sophomore, are currently leveraging microscopy images of shellfish seed from ARC to train machine learning algorithms aimed at automating the identification and counting processes. The envisioned user-friendly image recognition tool is designed to assist aquaculturists in distinguishing and tallying healthy, unhealthy, and deceased shellfish larvae for enhanced precision, reduced time investment, and improved efficiency.

Vincent underlines the significant role of AI in environmental science, empowering researchers, industry players, and resource managers to tackle long-standing obstacles in accurate data collection, analysis, predictions, and process enhancement. He emphasizes the critical support from programs like J-WAFS in confronting these challenges head-on.

ARC is currently grappling with the manual quantification of larvae classes, a crucial phase in its seed production process. According to Cheryl James, ARC’s larval/juvenile production manager, “Sizing and counting larvae in their growth phases is a continuous process crucial for optimal growth and population reinforcement.”

By developing an automated identification and counting framework, this process can be streamlined, offering time and cost efficiencies. Vincent acknowledges the complexity of this endeavor but highlights substantial progress made under the guidance of Dr. Zolotas at the Northeastern University Institute for Experiential Robotics and the diligent efforts of UROP students.