One of the most significant challenges in aquaculture is to mitigate or avoid production losses due to disease agents caused by parasites (such as sea lice), algae, bacteria, viruses, and fungi. This creates a critical need for the early detection of dangerous microorganisms and disease-causing agents in the water. Currently, the samples are collected manually and analyzed by trained technicians in a laboratory, often centralized, using conventional lens-based microscopes. This approach is expensive, laborious, time consuming and limiting scalability of operations. Furthermore, delays with disease agent identification often eliminate many mitigation strategies forcing approaches leading to production losses or pre-mature harvesting.
To address this critical need, Lucendi has developed Aqusens – an AI-based holographic microscopy platform capable of rapid automated monitoring of water samples for the presence and accurate quantification of dangerous organisms, such as harmful algae, sea lice and others. Unlike conventional lens-based microscopy systems, Aqusens relies on capturing interference patterns as objects are passing through a pulsating light field. These patterns are then processed and characterized by deep learning. With this novel approach, Aqusens can identify objects anywhere from 2 micron to 4 mm range and above. The system screens the water with unprecedented 100 mL/hr throughput, which can be increased to liters per hour depending on the application needs. For every object Aqusens generates an intensity and phase images which are then analyzed by custom AI to determine the object type and to compute concentration. Collected data can then be immediately shared with key decision makers to provide advanced warning, optimize operations, and save costs.
Aqusens is currently deployed at several in-field facilities. It was validated in the laboratory studies demonstrating accurate identification of various harmful algae and sea lice. Aqusens can also be customized for specific operational conditions and applications, and is developed to support scalability and safety of the operations.