Feb 7 2019
An innovative artificial intelligence (AI) program recently developed by researchers can automatically enable species-level identification of extremely small marine organisms.
Foraminifera, or forams, as seen through the viewfinder of a microscope. (Image credit: Tom Marchitto)
The subsequent step is to integrate this AI program into a robotic system that will potentially improve people’s knowledge of oceans in the world, both in the prehistoric past and in the present day. In particular, the AI program has been shown to have the ability to identify six species of organisms—foraminifera or forams—that have been existing in Earth’s oceans for over 100 million years.
Forams are actually protists—that is, they are neither animal nor plant. When these organisms die, they tend to leave behind their microscopic shells, with most of them being less than one millimeter wide. Such shells provide researchers a deeper understanding of the characteristics of the world’s oceans since they existed when the forams were thriving. For instance, different kinds of ocean environments are a haven for different kinds of foram species, and through chemical measurements, researchers can learn about everything, right from the ocean’s temperature to its chemistry when the shells were being formed.
Conversely, assessing those foram fossils and shells is not only time-intensive but also laborious. That is the reason why an interdisciplinary research team, with know-how ranging from paleoceanography to robotics, has collaborated to automate the process.
At this point, the AI correctly identifies the forams about 80 percent of the time, which is better than most trained humans. But this is only the proof of concept. We expect the system to improve over time, because machine learning means the program will get more accurate and more consistent with every iteration. We also plan to expand the AI’s purview, so that it can identify at least 35 species of forams, rather than the current six.
Edgar Lobaton, Study Co-Author and Associate Professor, Department of Electrical and Computer Engineering, North Carolina State University
In the present system, a foram is placed under a microscope that has the ability to take photos. An LED ring then illuminates light onto the foram from 16 directions in a sequential order, while simultaneously taking the foram’s image with each variation in light. All the 16 images are integrated to offer as much geometric data as possible about the shape of the foram. The AI program subsequently applies this data to spot the foram’s species. The combined task of scanning and identification takes only a few seconds, and is already as fast, or even faster, than the fastest human experts.
Plus, the AI doesn’t get tired or bored. This work demonstrates the successful first step toward building a robotic platform that will be able to identify, pick and sort forams automatically.
Edgar Lobaton, Study Co-Author and Associate Professor, Department of Electrical and Computer Engineering, North Carolina State University
The National Science Foundation (NSF) has awarded a grant to Lobaton and his collaborators to construct the fully-functional robotic system, beginning in January 2019.
This work is important because oceans cover about 70 percent of Earth’s surface and play an enormous role in its climate. Forams are ubiquitous in our oceans, and the chemistry of their shells records the physical and chemical characteristics of the waters that they grew in. These tiny organisms bear witness to past properties like temperature, salinity, acidity and nutrient concentrations. In turn we can use those properties to reconstruct ocean circulation and heat transport during past climate events. This matters because humanity is in the midst of an unintentional, global-scale climate ‘experiment’ due to our emission of greenhouse gases,” Marchitto says. “To predict the outcomes of that experiment we need a better understanding of how Earth’s climate behaves when its energy balance is altered. The new AI, and the robotic system it will enable, could significantly expedite our ability to learn more about the relationship between the climate and the oceans across vast time scales.
Tom Marchitto, Study Corresponding Author and Associate Professor, Geological Sciences, University of Colorado, Boulder
The paper titled, “Automated species-level identification of planktic foraminifera using convolutional neural networks, with comparison to human performance,” has been published in the journal Marine Micropaleontology.
Ritayan Mitra, the paper’s lead author and a former postdoctoral researcher at NC State and the University of Colorado Boulder, is currently at IIT Bombay. Other co-authors of the study include Q. Ge and B. Zhong, PhD students at NC State; B. Kanakiya, a former master’s student at NC State; M.S. Cook of Williams College; J.S. Fehrenbacher of Oregon State University; J.D. Ortiz of Kent State University; and A. Tripati of UCLA.
The NSF under grant number 1637039 supported the AI work.