This week we learned about a facial recognition system for cows, a bacterium that consumes toxic metals and poops out gold without poisoning itself, and a live worm that lives inside a computer and can balance a pole on the tip of its tail. Together with the car now orbiting Earth, this week got a lot of mileage out of science.
What is it? The Irish machine vision company Cainthus is putting facial recognition and remote sensing out on the pasture. The company, which just received an investment from the U.S. agricultural giant Cargill, has developed “proprietary software” that can identify cows by their hide patterns and faces, and track “key data such as food and water intake, heat detection and behavior patterns,” according to a Cargill news release. “The software then delivers analytics that drive on-farm decisions that can impact milk production, reproduction management and overall animal health.”
Why does it matter? Predictive analytics are already monitoring planes and power plants, but they are clearly spreading into farming. “Our shared vision is to disrupt and transform how we bring insights and analytics to dairy producers worldwide,” said SriRaj Kantamneni, managing director for Cargill’s digital insights business, said the release. “Our customers’ ability to make proactive and predictive decisions to improve their farm’s efficiency, enhance animal health and wellbeing, reduce animal loss, and ultimately increase farm profitability are significantly enhanced with this technology.”
How does it work? Cainthus’ imaging technology “can identify individual cows by their features in several seconds to memorize a cow’s unique identity, recording individual pattern and movements,” Cargill said. “That information is used as part of an artificial intelligence-driven mathematical algorithm that conveys imagery into feed and water intake analysis, behavioral tracking and health alerts that can be sent directly to the farmer. Data gleaned from those images is used to anticipate issues and adjust feeding regimens. What used to be a manual process that took days or weeks now takes place in near real-time.”
What is it? University researchers working in Germany and Australia have shed light on a trick that allows a special kind of bacteria to swallow miniscule amounts of toxic metals and poop out “tiny gold nuggets” — without poisoning itself.
Why does it matter? The process could help miners produce gold “from ores containing only a small percentage of gold without requiring toxic mercury bonds as was previously the case,” Martin Luther University Halle-Wittenberg said in a news release.
How does it work? The team, working at Martin Luther University Halle-Wittenberg, the Technical University of Munich and the University of Adelaide, studied C. metallidurans, a bacterium known to extract gold from soils full of poisonous heavy metals. “Apart from the toxic heavy metals, living conditions in these soils are not bad,” said Dietrich H. Nies, a microbiologist at MLU. “There is enough hydrogen to conserve energy and nearly no competition. If an organism chooses to survive here, it has to find a way to protect itself from these toxic substances.” The team observed that the bacteria use a pair of enzymes to transform copper and gold compounds found in the soil into forms that are difficult to absorb. “This assures that fewer copper and gold compounds enter the cellular interior,” Nies said. “The bacterium is poisoned less and the enzyme that pumps out the copper can dispose of the excess copper unimpeded. Another consequence: the gold compounds that are difficult to absorb transform in the outer area of the cell into harmless gold nuggets only a few nanometres in size.”
What is it? It’s going to be a while before we can upload our brains into a computer. But programmers at the Vienna University of Technology in Austria have taken what might be the first step on that journey, blurring the boundaries between a living being and a being living inside a computer. They uploaded into a computer a copy of the neural system of a simple worm, C. elegans. Once running inside, they trained the worm’s digital doppelganger to balance a pole at the tip of its tail.
Why does it matter? “The project raises the question whether there is a fundamental difference between living nerve systems and computer code,” the team wrote in a news release. “Is machine learning and the activity of our brain the same on a fundamental level? At least we can be pretty sure that the simple nematode C. elegans does not care whether it lives as a worm in the ground or as a virtual worm on a computer hard drive.”
How does it work? With only 300 neurons, the worm is “the only living being whose neural system has been analyzed completely,” the team reported. The team uploaded the worm’s neural system inside the computer and started manipulating it by emulating the ways it normally responds to touch and other basic stimuli in nature. “With the help of reinforcement learning, a method also known as ‘learning based on experiment and reward,’ the artificial reflex network was trained and optimized on the computer,” explained the university’s Mathias Lechner. The approach allowed the team to teach the virtual worm to balance a pole. “The result is a controller, which can solve a standard technology problem – stabilizing a pole, balanced on its tip,” said Lechner’s colleague Radu Grosu. “But no human being has written even one line of code for this controller, it just emerged by training a biological nerve system.”
What is it? Researchers at the University of Pennsylvania and the University of California, Irvine, have found a way to help wounds heal without scars. “Essentially, we can manipulate wound healing so that it leads to skin generation rather than scarring,” said George Costarelis, the chair of the Department of Dermatology at Penn and the project’s principal investigator.
Why does it matter? The university said in a news release that “the first and most obvious use” would be to develop a therapy that helps wounds heal without scarring. Skin regeneration could also lead to new anti-aging treatments.
How does it work? The team found a way to transform myofibroblasts, the most common cells found in healing wounds, into fat cells called adipocytes — a feat long considered impossible. Adipocytes are common in healthy skin but not in scars. “The secret is to generate hair follicles first,” Costarelis said. “After that, the fat will generate in response to the signals from those hair follicles.” His colleague Maksim Plikus working at the University of California, Irvine, said that “the findings show we have a window of opportunity after wounding to influence the tissue to regenerate rather than scar.
What is it? Scientists at Columbia University have developed a lithium battery shaped like the human spine that they say retains “high energy density and stable voltage no matter how it is flexed or twisted.”
Why does it matter? The battery could power a wide range of flexible wearable devices, patches, sensors and smart fabrics. “The energy density of our prototype is one of the highest reported so far,” said Yuan Yang, assistant professor of materials science and engineering at Columbia. “We’ve developed a simple and scalable approach to fabricate a flexible spine-like lithium ion battery that has excellent electrochemical and mechanical properties. Our design is a very promising candidate as the first-generation, flexible, commercial lithium-ion battery. We are now optimizing the design and improving its performance.”
How does it work? Like the human spine, the battery alternates “thick, rigid segments” that store energy with flexible parts that connect them together. “Our spine-like design is much more mechanically robust than are conventional designs,” Yang says. “We anticipate that our bio-inspired, scalable method to fabricate flexible Li-ion batteries could greatly advance the commercialization of flexible devices.”