AQUA 2024

August 26 - 30, 2024

Copenhagen, Denmark

AQUSENS: AI-POWERED PLATFORM FOR AUTOMATED QUANTIFICATION OF LIVE FEED

 Maxim Batalin*, Zoltan Gorocs , Dhruvin Naik, Jessica Keszey
Address: Lucendi, Inc., 11040 Santa Monica Blvd., Suite 220, Los Angeles, CA 90025
Email: mbatalin@lucendi.ai

 



Many fish hatcheries rely on live feeds (i.e., microorganisms, such as copepods, rotifers, and artemia). One of the critical capabilities missing today is to enable automated live feed counting and characterization. Introducing automation, will result in elimination of manual processes, will improve accuracy, will promote standardization, and will reduce errors and culture crashes.

 Existing solutions to live feed quantification and characterization are based predominantly on manual counts performed by trained experts on conventional microscopes. Furthermore , re-purposing existing  lens-based imaging flow cytometers for automated live feed characterization is not ideal due to the high cost of the equipment and performance limitations of these solutions, including low-throughput processing, significant maintenance requirements, issues with adapting for large size range of organisms that is the case with live feed species.

Lucendi has developed the Aqusens platform, shown on Figure 1, which is capable of rapid and label free evaluation of the live feed samples with throughputs of anywhere from 100 mL/hr – 5 L/hr , depending on use case needs with sufficient resolution to characterize organisms in a diverse size range from  single microns to millimeter scale. Furthermore, since the Aqusens is built primarily from the off the shelf components, the cost of the hardware solution is  significantly  lower than  any of the existing alternatives.

Aqusens AI infrastructure  is  capable of automated live feed  characterization  by type (i.e. copepods, rotifers, artemia, etc.),  and  by phase of development (i.e. egg, nauplii, copepod). Aqusens can also identify cross-contamination to avoid production inefficiencies and losses.

 Aqusens has been deployed in-field and was demonstrated to provide continuous live feed sample monitoring and quantification  that is at least  on par or better  and more repeatable than that of the human operator.