The device is capable of detecting and preventing diseases in seaweed farms before they spread.
A graduate engineering student from MIT has developed technology that is capable of detecting and preventing diseases on seaweed farms. According to lead researcher Charlene Xia, currently a PhD student in the mechanical engineering department at MIT, the objective was to keep the device affordable while equipping it with the ability to forecast and control disease transmission in aquaculture. The device utilizes a holographic microscope and a machine learning system to monitor the microbiota present in seaweed in order to identify the presence of biological threats. These tools allow it to convert 2D images into a representative 3D environment of the existing microbiome, providing marine researchers with relevant data.
To achieve this, the team combined both old and new technology. The 2D image is captured using a submersible holographic digital microscope. The machine learning neural network then takes the 2D image and transforms it into a representation of the microbiome present in the 3D environment. The neural network software can be run using a small Raspberry Pi, which can be fastened on to the holographic microscope.
“Using a machine learning network, you can take a 2D image and reconstruct it almost in real time to get an idea of what the microbiome looks like in a 3D space,” shared Xia.
To deliver the data to users, Xia took inspiration from her master’s degree research. Instead of using expensive underwater communication devices, the team constructed miniature low-cost devices that would lower the cost barrier to less than $100. This is a large difference compared to these devices’ typical $4,000 price tag.
Seaweed is responsible for combatting different environmental threats by absorbing excess carbon dioxide in the atmosphere as well as fertilizer run-off, subsequently helping mitigate climate change. However, just like other marine life, it is not immune to the effects of climate change. Warm temperatures coupled with minimal sunshine are more likely to increase the development of harmful bacteria that can develop into the likes of ice-ice disease.
Ice-ice disease is typically caused by changes in salinity, ocean temperature and light intensity, which enables stress in seaweeds. This makes them produce a moist organic substance that attracts bacteria in the water, which eventually turns seaweed tissue white and hardens them.
This unregulated proliferation of bacteria can result in the rapid destruction of seaweed farms within days.
The team is now currently working on designing a low-cost underwater camera system to work alongside the wireless communication device. This will serve as a real-time monitoring system that can be scaled to cover large seaweed farms. Through this, researchers will be able to access data about the microbiome. Users can then detect whether diseases are about to strike and jeopardize seaweed or livestock before they worsen.
“It’s almost like having the ‘internet of things’ underwater,” explained Xia.
As climate change continues to be at the forefront of global issues, there is a need for more innovative climate change mitigation strategies. Student-led projects are vital in mapping and addressing local environmental challenges, subsequently tackling underlying drivers of climate change. By focusing on experiential learning, students are able to use their expertise and creativity towards developing more innovative solutions. Similarly, it shows the capacity to design and implement sustainability plans that target root causes.
The project was initially proposed under a new seed funding program by computing software company MathWorks in MIT’s Department of Mechanical Engineering. Besides Xia, the team included Professor David Wallace and Assistant Professor Stefanie Mueller.
