Tech

Image inpainting is a computer vision technique in which pixels missing from an image are filled in. It is often used to remove unwanted objects from an image or to recreate missing regions of occluded images. Inpainting is a common tool for predicting missing image data, but it's challenging to synthesize the missing pixels in a realistic and coherent way.

Researchers at the University of Tokyo have presented a frequency-based inpainting method that enables the use of both frequency and spatial information to generate missing image portions. Publishing in the Journal of Electronic Imaging (JEI), Hiya Roy et al. detail the technique in "Image inpainting using frequency domain priors." Current methods employ only spatial domain information during the learning process, which can allow details of interior reconstruction to be lost, resulting in the estimation of only a low-frequency part of the original patch. To solve that problem, the researchers looked to frequency-based image inpainting and demonstrated that converting inpainting to deconvolution in the frequency domain can predict the local structure of missing image regions.

"The frequency-domain information contains rich representations which allow the network to perform the image understanding tasks in a better way than the conventional way of using only spatial-domain information," Roy says. "Therefore, in this work, we try to achieve better image inpainting performance by training the networks using both frequency and spatial domain information."

Image inpainting algorithms historically fall into two broad categories. Diffusion-based image inpainting algorithms attempt to replicate the appearance of the image into the missing regions. This method can fill small holes well, but the quality of the results erodes as the size of holes increases. The second category is patch-based inpainting algorithms, which seek the best-fitting patch in the image to fill missing portions. This method can fill larger holes but is ineffective for complex or distinctive portions of an image.

"The originality of the research resides in the fact that the authors used the frequency domain representation, namely the spectrum of the images obtained by fast Fourier transform, at the first stage of inpainting with a deconvolution network," says Jenny Benois-Pineau of the University of Bordeaux, a senior editor for JEI. "This yields a rough inpainting result capturing the structural elements of the image. Then the refinement is fulfilled in the pixel domain by a GAN network. Their approach outperforms the state-of-the art in all quality metrics: PSNR, SSIM, and L1."

Roy and colleagues show that deconvolution in the frequency domain can predict the missing regions of the image structure using context from the image. In its first stage, their model learned the context using frequency domain information, then reconstructed the high-frequency parts. In the second stage, it used spatial domain information to guide the color scheme of the image and then enhanced the details and structures obtained in the first stage. The result is better inpainting outcomes.

"Experimental results showed that our method could achieve results better than state-of-the-art performances on challenging datasets by generating sharper details and perceptually realistic inpainting results," say Roy et al. in the research paper. "Based on our empirical results, we believe that methods using both frequency and spatial information should gain dominance because of their superior performance."

The group expects their research to become a springboard to extend the use of other types of frequency domain transformations to solve image restoration tasks such as image denoising.

A new study shows that several disagreements between Ethiopia, Sudan and Egypt around Africa's largest hydropower plant, the new Grand Ethiopian Renaissance Dam (GERD), could be alleviated by massively expanding solar and wind power across the region. Adapting GERD operation to support grid integration of solar and wind power would provide tangible energy and water benefits to all involved countries, creating regional win-win situations. "Our results call for integrated hydro-solar-wind planning to be taken up in the GERD negotiations," says Sebastian Sterl, energy planning expert at Vrije Universiteit Brussel (VUB) and KU Leuven in Belgium and lead author of the study, published in Nature Energy.

For several years, political tensions between Egypt, Sudan and Ethiopia have been escalating in a conflict surrounding Africa's largest hydropower plant: the nearly complete Grand Ethiopian Renaissance Dam (GERD) on the Blue Nile. Ethiopia, which started filling GERD's massive reservoir in 2020, says it needs GERD's electricity to lift millions of its citizens out of poverty. But Egypt is deeply concerned by the mega-dam's consequences for the Nile river, since its agriculture depends completely on Nile water -Egypt raised this issue to the UN Security Council earlier in 2020. Sudan, meanwhile, appears caught between both sides. Ongoing African Union-led mediation talks to agree on long-term operation of the dam have so far yielded little fruit. Certain tongues have even invoked the looming threat of a "water war" between Cairo and Addis Ababa.

Seasonal profiles

Sebastian Sterl, energy planning expert at VUB and KU Leuven and lead author of the study, explains: "The Blue Nile is a highly seasonal river. The GERD's reservoir is so large that it can store the river's full peak flow and deliver hydropower at a stable rate throughout the year, removing the flow seasonality. This makes a lot of sense from the Ethiopian perspective, but it overhauls the natural timing of the water reaching Sudan and Egypt. Behind many disagreements around GERD lies the question of who, if anyone, should be allowed to exert such control over the Nile river."

A group of researchers based in Belgium and Germany, led by Sterl, have now identified a surprising method that could solve multiple disagreements around the dam at once and benefit all three countries. The idea boils down to massively deploying modern, clean solar and wind power to serve as a complement to GERD's hydropower. More concretely: the researchers propose that Ethiopia and its neighbours deploy large-scale solar and wind farms, work towards a regionally integrated power grid, and then agree on Ethiopia operating GERD in synergy with solar and wind power. This would mean turbining less water on sunny and windy days, and more water during cloudy, windless spells and at nighttime, to "firm up" the always-fluctuating solar and wind power.

The researchers realised that sunshine and wind in many regions of Ethiopia, Sudan and their eastern African neighbours have opposite seasonal profiles to the Blue Nile flow. In these places, the sun shines brightest and the winds blow strongest during the dry season. This "seasonal synergy" between water, sun and wind lies at the heart of the researchers' findings.

The study found that, if GERD were operated to back up solar and wind power throughout the year - both hourly and seasonally - this would automatically mean producing less hydropower during the dry season, and more during the wet season, without negatively affecting GERD's yearly average power output. The water flowing out of the dam would then have a seasonality somewhat resembling the natural river flow, with a clear peak in the wet season.

According to Sterl, if GERD were operated in this way, "Essentially, Ethiopia would have all the expected benefits of a big dam - but for Sudan and Egypt, it would look as if the Ethiopians only built a modest, relatively small reservoir. There are many such reservoirs already on the Nile, so no country downstream of Ethiopia could really object to this."

Regional cooperation

By reconciling parties around common energy and water objectives, the researchers identified at least five concrete benefits of such integrated hydro-solar-wind planning. First, Ethiopia could become Africa's largest power exporter while reducing its dependence on hydropower and lowering its electricity generation costs on the long term. Second, consumption of polluting fossil fuels in Sudan and other eastern African countries could be displaced by solar and wind power, backed up by GERD. Third, thanks to the proposed operation scheme of GERD, Egypt could receive more water during dry years than before and would not need to change the operation of its own High Aswan Dam. Fourth, Ethiopia would make more efficient use of its mega-dam's more than a dozen turbines by frequently producing at peak power whenever solar and wind would be unavailable. And fifth, Nile river ecology across Sudan would be less affected by the new dam, as flow seasonality is an important component of rivers' ecological sustainability.

According to the authors, the entire eastern African region stands to contribute. "Ethiopia could theoretically go alone, using GERD to back up its own solar and wind power," says Sterl. "But it would work much better if, say, Sudan were to join in - it has better solar and wind resources than Ethiopia, allowing for better hydro-solar-wind synergies and reducing the overall costs of renewable power generation. Egypt has great solar and wind resources too, as do Djibouti, South Sudan and other eastern African countries. Regional cooperation in a common, Eastern African Power Pool could be key."

The results of the study suggest that integrated hydro-solar-wind planning could be a highly interesting option to discuss in the ongoing GERD negotations between Ethiopia, Sudan and Egypt. "You could call it a win-win situation," says prof. Wim Thiery, climate researcher at VUB and co-author of the study. "The entire region would benefit."

Model

The researchers obtained their results by using a dedicated, highly detailed computer model (REVUB) conceived to simulate the operation of hydropower dams alongside other renewables, like solar and wind power. The model was originally created by the same VUB-researchers in 2019 to study renewable electricity scenarios for West Africa. Later, as the GERD negotiations became more and more present in the media, the researchers realised they could directly apply the same tool to study solar and wind power as potential solutions to the GERD conflict.

Researchers have developed a new algorithm capable of identifying features of male zebra finch songs that may underlie the distinction between a short phrase sung during courtship, and the same phrase sung in a non-courtship context. Sarah Woolley of McGill University in Montreal, Canada, and colleagues present these findings in the open-access journal PLOS Computational Biology.

Like many animals, male zebra finches adjust their vocal signals for their audience. They may sing the same sequence of syllables during courtship interactions with females as when singing alone, but with subtle modifications. However, humans cannot detect these differences, and it was not clear that female zebra finches could, either.

For the new study, Woolley and colleagues first conducted behavioral experiments demonstrating that female zebra finches are indeed highly adept at discriminating between short segments of males' songs recorded in courtship versus non-courtship settings.

Next, they sought to expand on earlier studies that have focused on just a few specific song features that may underlie the distinction between courtship and non-courtship song. Taking a "bottom-up" approach, the researchers extracted over 5,000 song features from recordings and trained an algorithm to use those features to distinguish between courtship and non-courtship song phrases.

The trained algorithm uncovered features that may be key for song perception, some of which had not been identified previously. It also made predictions about the distinction capabilities of female zebra finches that aligned well with the results of the behavioral experiments.

These findings highlight the potential for bottom-up approaches to reveal acoustic features important for communication and social discrimination.

"As vocal communicators ourselves, we have a tendency to focus on aspects of communication signals that are salient to us," Woolley says. "Using our bottom-up approach, we identified features that might never have been on our radar."

Next, the researchers plan to test whether manipulating the acoustic features they discovered alters what female finches think about those songs. They also hope to evaluate how well their findings might generalize to courtship and non-courtship songs in other species.

Taiwan is an island of extremes: severe earthquakes and typhoons repeatedly strike the region and change the landscape, sometimes catastrophically. This makes Taiwan a fantastic laboratory for geosciences. Erosion processes, for example, occur up to a thousand times faster in the center of the island than in its far south. This difference in erosion rates influences the chemical weathering of rocks and yields insights into the carbon cycle of our planet on a scale of millions of years. A group of researchers led by Aaron Bufe and Niels Hovius of the German Research Center for Geosciences (GFZ) has now taken advantage of the different erosion rates and investigated how uplift and erosion of rocks determine the balance of carbon emissions and uptake. The surprising result: at high erosion rates, weathering processes release carbon dioxide; at low erosion rates, they sequester carbon from the atmosphere. The study will be published in Nature Geoscience.

Behind all this are tectonic and chemical processes. In rapidly growing mountains in particular, tectonic uplift and erosion constantly bring fresh rock material up from underground. There it is exposed to circulating acidic water which dissolves or alters the rock. Depending on the type of rock, this weathering has very different effects on Earth's climate. For example, if carbonic acid from the soil comes into contact with silicate minerals, limestone (calcium-carbonate or CaCO3) precipitates, in which the carbon is then bound for a very long time. In the case of a combination of sulfurous mineral, such as pyrite, and limestone, the opposite happens. The sulfuric acid that forms when pyrite comes into contact with water and oxygen dissolves carbonate minerals, thus producing CO2. This relationship between mountain building and chemical weathering is thought to affect our planet's climate on a scale of millions of years. But how exactly does the growth of the Alps or the Himalayas affect climate? Does silicate weathering accelerate, causing the climate to cool? Or does the dissolution of limestone by sulfuric acid dominate, driving the concentration of atmospheric CO2 up, with attendant global warming?

This question can be answered in southern Taiwan. Taiwan is located at a subduction zone, where an ocean plate slides under the Asian continent. This subduction causes rapid mountain growth. While the center of the island has been standing tall for several million years, the southern tip has just emerged from the sea. There, the mountains have low relief and they erode relatively slowly. Farther north, where the mountains are steep and tall, fresh rock is quickly brought to the Earth's surface to weather. Usefully, the rocks of southern Taiwan are typical of many young mountain ranges around the world, containing mostly silicate minerals with some carbonate and pyrite.

In their study, the researchers sampled rivers that collect water from these mountains at different erosion rates. From the material dissolved in the rivers, the researchers estimated the proportion of sulfide, carbonate, and silicate minerals in the weathering. These results allowed them to estimate the both the amount of CO2 that is sequestered and the amount of CO2 released by the weathering reactions. First author Aaron Bufe reports, "We found that in the southernmost part of Taiwan, atmospheric CO2 sequestration dominates. However, farther north, where mountains are eroding faster, carbonate and sulfide weathering rates dominate and CO2 is released."

So, does weathering of mountain ranges increase CO2 in the atmosphere? Aaron Bufe says, "we can make relatively good statements about Taiwan. It appears that chemical weathering in this most active of mountain belts is a net emitter of CO2 to the atmosphere due to chemical weathering. But, perhaps the story changes when sediments washed down from the mountains are trapped in vast alluvial plains; like at the foot of the Himalayas or the Alps. Those sediments are often rich in silicates, the weathering of which will sequester CO2. In addition, mountain building brings not only sedimentary rocks with pyrite and carbonate to the Earth's surface, but also rock types that have formed from solidified magma and contain many fresh silicates that weather quickly. Researchers have some mountains to climb before we fully know the net effect of weathering on the Earth's climate."

Every third-grader knows that plants absorb nutrients from the soil through their roots. The fact that they also release substances into the soil is probably less well known. And this seems to make the lives of plants a lot easier.

That is at least the conclusion of the current study. The participating researchers studied several maize varieties that differ significantly in their yield. In their search for the cause, they came across an enzyme, flavone synthase 2. "The high-yield inbred line 787 we studied contains large amounts of this enzyme in its roots", explains Dr. Peng Yu of the Institute of Crop Science and Resource Conservation (INRES) at the University of Bonn. "It uses this enzyme to make certain molecules from the flavonoid group and releases them into the soil."

Flavonoids give flowers and fruits their color. In the soil, however, they perform a different function: They ensure that very specific bacteria accumulate around the roots. And these microbes, in turn, cause the formation of more lateral branches on these roots, called lateral roots. "This allows the maize plant to absorb more nitrogen from the environment," explains Prof. Dr. Frank Hochholdinger of the Institute of Crop Science and Resource Conservation (INRES). "This means the plant grows faster, especially when nitrogen supplies are scarce."

Sterilized soil did not cause a growth spurt

The researchers were able to demonstrate in experiments how well this works. They did this using a maize variety with the abbreviation LH93, which normally produces rather puny plants. However, that changed when they planted this variety in soil where the high-performance line 787 had previously grown: LH93 then grew significantly better. The effect disappeared when the botanists sterilized the soil before repotting. This shows that the enriched bacteria are indeed responsible for the turbo growth, because they were killed during sterilization.

The researchers were able to demonstrate in another experiment that the microorganisms really do promote the growth of lateral roots. Here, they used a maize variety that cannot form lateral roots due to a mutation. However, when they supplemented the soil with the bacterium, the roots of the mutant started to branch out. It is not yet clear how this effect comes about. Additionally, with microbial support the maize coped far better with nitrogen deficiency.

Results may contribute to more sustainable agriculture

Nitrogen is extremely important for plant growth - so much so, that farmers artificially increase its amount in the soil by applying fertilizer. However, some of the fertilizer is washed off the fields into streams with the rain or enters the groundwater. It can also enter the atmosphere in the form of nitrogen oxides or as ammonium gas, where it contributes to the greenhouse effect. The production of nitrogenous fertilizers furthermore requires a great deal of energy. "If we breed crops that can improve their nitrogen usage with the help of bacteria, we might be able to significantly reduce environmental pollution," Yu hopes.

The study shows that plants help to shape the conditions of the soil in which they grow, in ways that ultimately benefit them. However, this aspect has been neglected in breeding until now. Dr. Peng Yu adds that, in general, many interactions of the root system with soil organisms are not yet well enough understood. He wants to help change that: He has just taken over the leadership of an Emmy Noether junior research group at the University of Bonn, which is dedicated to precisely this topic. With its Emmy Noether Program, the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) offers young researchers an opportunity to qualify for a university professorship within six years.

When it comes to choosing a partner, humans tend to be attracted by characteristics like personality and common interests. In contrast, insects tend to be a bit shallow, as they choose a mate based on appearance, and in some cases, smells. One example is the leaf beetle, which produces chemical pheromones that are on their cuticles, or the exterior surface of the beetle. They use these 'scents' to assess beetle sex and mating status (whether beetles are sexually mature or not).

Kari Segraves, professor of biology in the College of Arts and Sciences, is interested in researching the chemical and visual signals that contribute to mate selection by these beetles. This work is part of a larger project focused on understanding how new species are formed. By definition, species are related organisms that share common characteristics and are capable of interbreeding. Mating studies are essential in determining the mechanisms that might lead to reproductive barriers between newly formed species.

In the past, she and her colleagues have used a variety of models during these studies, including dead females and even round glass beads. To find a new and improved way of studying these insects, Segraves recently collaborated with Huai-Jun Xue and Si-Qin Ge from the Institute of Zoology, Chinese Academy of Sciences, to design, produce and test 3D printed beetle models to find out if they would be a feasible option when testing mate choice in these and possibly other insects. They believe this novel research to be the first time that 3D printed beetles have been used in mating tests.

According to Segraves, they designed this study after receiving some unusual results in a mating experiment that focused on the chemical signals used by leaf beetles. In that study, they used dead beetles as mating models and swapped the chemical signals between the sexes so that the male model smelled like a female and the female model smelled like a male. Contrary to what they expected, they found that males did not have a preference for female over male chemical signals. This suggested that the trial itself may have been flawed because they had shown that the beetles always preferred females in another experiment that used chemically unaltered dead beetle models. They were concerned that more chemicals were being released after they had washed the dead models and that this was altering the results.

"We thought it would be a good idea to try using 3D printed models instead of dead females because the plastic used in 3D printing doesn't have chemicals that would be confused as mating signals by the beetles," Segraves says. She and fellow researchers from the Chinese Academy of Sciences conducted the planning and design of the work, and the 3D beetles were produced and tested in Huai-Jun Xue's lab in China.

"In the study, we learned that the 3D models worked and were more effective than models that weren't shaped like the beetles such as glass beads," Segraves notes. Another interesting result revealed that males use color in mate selection as males tested on black versus white 3D models all chose the black models.

In the end, the researchers determined that when given an option, the beetles opt for the real thing over the 3D counterpart. When males were offered dead females with the reapplied chemical signals, males preferred dead females over black 3D models coated with the female scent. But Segraves says that's to be expected given that there are probably other types of signals that can't be replicated by 3D printing such as tactile signals.

Their results indicate that in the absence of a dead female beetle, 3D-printed models can provide a feasible and cost-effective method for mating studies of insects. Segraves believes that going forward, researchers studying beetles and other similar insects should use a combination of models. While dead females are most realistic, the 3D printed models allows scientists to more tightly control the chemicals that are present in a given experiment.

The network of nerves connecting our eyes to our brains is sophisticated and researchers have now shown that it evolved much earlier than previously thought, thanks to an unexpected source: the gar fish.

Michigan State University's Ingo Braasch has helped an international research team show that this connection scheme was already present in ancient fish at least 450 million years ago. That makes it about 100 million years older than previously believed.

"It's the first time for me that one of our publications literally changes the textbook that I am teaching with," said Braasch, as assistant professor in the Department of Integrative Biology in the College of Natural Science.

This work, published in the journal Science on April 8, also means that this type of eye-brain connection predates animals living on land. The existing theory had been that this connection first evolved in terrestrial creatures and, from there, carried on into humans where scientists believe it helps with our depth perception and 3D vision.

And this work, which was led by researchers at France's Inserm public research organization, does more than reshape our understanding of the past. It also has implications for future health research.

Studying animal models is an invaluable way for researchers to learn about health and disease, but drawing connections to human conditions from these models can be challenging.

Zebrafish are a popular model animal, for example, but their eye-brain wiring is very distinct from a human's. In fact, that helps explain why scientists thought the human connection first evolved in four-limbed terrestrial creatures, or tetrapods.

"Modern fish, they don't have this type of eye-brain connection," Braasch said. "That's one of the reasons that people thought it was a new thing in tetrapods."

Braasch is one of the world's leading experts in a different type of fish known as gar. Gar have evolved more slowly than zebrafish, meaning gar are more similar to the last common ancestor shared by fish and humans. These similarities could make gar a powerful animal model for health studies, which is why Braasch and his team are working to better understand gar biology and genetics.

That, in turn, is why Inserm's researchers sought out Braasch for this study.

"Without his help, this project wouldn't have been possible," said Alain Chédotal, director of research at Inserm and a group leader of the Vision Institute in Paris. "We did not have access to spotted gar, a fish that does not exist in Europe and occupies a key position in the tree of life."

To do the study, Chédotal and his colleague, Filippo Del Bene, used a groundbreaking technique to see the nerves connecting eyes to brains in several different fish species. This included the well-studied zebrafish, but also rarer specimens such as Braasch's gar and Australian lungfish provided by a collaborator at the University of Queensland.

In a zebrafish, each eye has one nerve connecting it to the opposite side of the fish's brain. That is, one nerve connects the left eye to the brain's right hemisphere and another nerve connects its right eye to the left side of its brain.

The other, more "ancient" fish do things differently. They have what's called ipsilateral or bilateral visual projections. Here, each eye has two nerve connections, one going to either side of the brain, which is also what humans have.

Armed with an understanding of genetics and evolution, the team could look back in time to estimate when these bilateral projections first appeared. Looking forward, the team is excited to build on this work to better understand and explore the biology of visual systems.

"What we found in this study was just the tip of the iceberg," Chédotal said. "It was highly motivating to see Ingo's enthusiastic reaction and warm support when we presented him the first results. We can't wait to continue the project with him."

Both Braasch and Chédotal noted how powerful this study was thanks to a robust collaboration that allowed the team to examine so many different animals, which Braasch said is a growing trend in the field.

The study also reminded Braasch of another trend.

"We're finding more and more that many things that we thought evolved relatively late are actually very old," Braasch said, which actually makes him feel a little more connected to nature. "I learn something about myself when looking at these weird fish and understanding how old parts of our own bodies are. I'm excited to tell the story of eye evolution with a new twist this semester in our Comparative Anatomy class."

URBANA, Ill. - When scientists need to understand the effects of new infant formula ingredients on brain development, it's rarely possible for them to carry out initial safety studies with human subjects. After all, few parents are willing to hand over their newborns to test unproven ingredients.

Enter the domestic pig. Its brain and gut development are strikingly similar to human infants - much more so than traditional lab animals, rats and mice. And, like infants, young pigs can be scanned using clinically available equipment, including non-invasive magnetic resonance imaging, or MRI. That means researchers can test nutritional interventions in pigs, look at their effects on the developing brain via MRI, and make educated predictions about how those same nutrients will affect human infants.

For nearly a decade, scientists have relied on an MRI-based map, or atlas, of the pig brain - developed at the University of Illinois using 4-week-old pigs - to understand where and how nutrients and other interventions affect the developing brain. Now, Illinois scientists have updated that atlas, increasing its resolution by a factor of four, and they have also added a new atlas for adolescent 12-week-old pigs.

The new atlases are freely available for download at pigmri.illinois.edu.

"That improvement in spatial resolution makes a huge difference when you're looking at development in a small pig brain and trying to see how your intervention is changing structure, size, or even function in the brain," says Brad Sutton, professor in the Department of Bioengineering, technical director of the Biomedical Imaging Center at Illinois' Beckman Institute, and a co-author on the brain atlas study, published in the Journal of Neuroscience Methods.

Ryan Dilger, associate professor in the Department of Animal Sciences and senior author on the atlas study, adds, "It's about our ability to discern one part of the brain from another. The higher the resolution, the more reliably we can say this piece is the hippocampus, for example. Part of the need for an atlas is for every research group working in this area to be referencing the same parts or regions of the brain consistently. We have to have common terms and infrastructure to speak the same language."

To build the updated atlas, the researchers anesthetized and scanned 4- and 12-week old pigs at Beckman's Biomedical Imaging Center using a state-of-the-art Siemens Prisma 3 T MRI scanner. Scans from multiple pigs in each age class were averaged into a single atlas for each age, to account for variation among individuals. Subsequently, the researchers identified and digitally isolated 26 regions of interest, such as the cerebellum, medulla, right and left cortex, and others, and provided volumetric standards for each in the pig.

"We provide the absolute and relative volumes for not only the whole brain, but tissues such as gray matter, white matter, cerebrospinal fluid, as well as all the different regions of interest. That normative data can act as a reference for other individuals who might be interested in seeing how a particular intervention influences brain growth or development in the pig," says Joanne Fil, doctoral student in the Neuroscience Program at Illinois and lead author on the atlas study.

The previous pig brain atlas has been used by researchers to advance neuroscience around the world, with some 450 downloads to date. The collective discoveries made possible by the atlas go well beyond pediatric nutrition to include deeper understanding of the microbiota-gut-brain axis, which appears to relate to common clinical situations.

Dilger says the new atlas will give researchers an even more precise view of the brain, enabling more advanced discoveries. And with the addition of the atlas for older pigs, they'll be able to extend their findings even farther.

"At 24 weeks of age, or six months, the pig is sexually mature. We would expect that by this age, the pig would have most, if not all, of its brain development completed," Fil says. "So now we're able to see how our interventions impact development not only at an early age, but also into adulthood in the pig."

Fil adds the study also provides a detailed account of the process they used to create the atlas, giving researchers the blueprints to create additional atlases for other animals.

But there's a lot to be said for pigs as an important biomedical research animal.

"You can study brain development in a mouse, but for some studies, the mouse brain is not similar to a human brain in some important aspects. Also, you can't really study intervention effects on the brain directly in humans, because although we can get people in the scanner, we can't always modify their diet and test out different components," Sutton says.

"So the pig is right in that sweet spot: its brain is the right size to use human MRI scanners and pig brain development closely matches that of humans. And we have tools to be able to study it in great detail, especially on this campus, and do great things with it. The pig is perfect for studying the brain."

Dilger adds, "We are using the actual human clinical equipment in the pig. We're effectively, non-invasively, taking a microscope to the pig brain while it's still alive. That's the benefit. We can take a virtual peek inside the pig brain multiple times throughout the life of the pig to see how the brain is structurally developing."

Detecting and analyzing breast cancer goes beyond the initial discovery of the cancer itself. If a patient has a tumor removed and it needs to be analyzed to determine further treatment, it might be OK for the results to take 24 hours. But if the patient is still on the operating table and clinicians are waiting to make sure no cancer cells are present along the edges of the removed tumor, results need to be nearly immediate.

A paper titled, "Breast cancer histopathology using infrared spectroscopic imaging: The impact of instrumental configurations," was published in Clinical Spectroscopy, an extension of previous work by Beckman Institute Postdoctoral Fellow Shachi Mittal and Rohit Bhargava, bioengineering professor and director of the Cancer Center at Illinois. Two alumni of the Beckman Postdoctoral Fellows program, Michael Walsh and Tomasz Wrobel, are co-authors. It seeks to advance breast cancer detection methodologies when using spectroscopic imaging.

They conducted the research at the Beckman Institute for Advanced Science and Technology at the University of Illinois Urbana-Champaign.

Because time is of the essence in breast cancer detection, the research identifies how clinicians can choose the right detection methods and tools during the correct scenario to still drive accurate results.

"The goal of the study was to give people a roadmap of how to plan and design an infrared imaging-based study for clinical work like digital histopathology," Mittal said. "In one scenario, you determine what instrument configuration will be better, and then you determine what types of methodologies you can use to develop accurate models with that instrument."

The team set out to examine the tradeoffs between using a standard Fourier transform infrared image or a high-definition Fourier transform infrared image.

"We wanted to see how the different types of instruments, especially different resolutions and how it impacts the capability of that data set to be used for different diagnostic purposes," Mittal said.

Instead of determining that one method is always superior, the researchers uncovered that the answer is much more complex. While high-definition imaging may seem like the best option regardless of the circumstances, sometimes standard definition infrared spectroscopy suffices in accuracy in instances when clinicians need a speedy result, for example.

"As technology expands and provides more capabilities with new features, it becomes more difficult to choose the optimal technology from the many options available," Bhargava said. "This study provides a nice comparison and guidelines to design a more useful and practical technology."

For Mittal, the research is an important step towards better dissecting, understanding, and curing cancer.

"Cancer is something we understand very little about," she said. "It's not only about the journey of battling the disease, but also the mental burden. Patients oftentimes don't understand, and they recognize the doctor may not fully understand either."

Mittal began her focus on breast cancer during her first graduate project, and continues to specialize her research in hopes to achieve something impactful that she can then apply to other types of cancers later in her academic career. She notes breast cancer has one of the highest mortality rates when looking at cancers that affect women.

"For females, it's not about just battling the disease, it also comes with a lot of cosmetic and life changing challenges," she said.

The interdisciplinary work that resulted in the paper was made possible by the Beckman Institute, Bhargava said.

"Developing new technology for human use requires a breadth of expertise that places like Beckman can bring together," he said. "I note that this study involves one current (Shachi) and one former (Tomasz) Beckman fellow as well as a Carle-Beckman Fellow (Michael). Together, these academicians' work is helping incubate a new community that provides chemical imaging technology to address human disease."

The almost 15-million-year-old Nördlinger Ries is an asteroid impact crater filled with lake sediments. Its structure is comparable to the craters currently being explored on Mars. In addition to various other deposits on the rim of the basin, the crater fill is mainly formed by stratified clay deposits. Unexpectedly, a research team led by the University of Göttingen has now discovered a volcanic ash layer in the asteroid crater. In addition, the team was able to show that the ground under the crater is sinking in the long term, which provides important insights for the exploration of craters on Mars, such as the ancient Gale and Jezero crater basin lakes, currently being explored by the NASA Curiosity and Perseverance Rovers. The results of the study have been published in the Journal of Geophysical Research Planets.

Until now, it was assumed that these lake deposits had settled on a stable crater floor. The same is assumed for crater deposits on Mars, although some of them show significantly inclined sediment strata. The layers of these crater fills appear on the surface as ring-shaped structures. However, a precise understanding of the underlying conditions and the temporal interrelationships of the deposits is important for reconstructing the chemical development of a crater lake and habitability for possible lifeforms that might have developed there in the past.

For the first time, the researchers have now been able to detect a volcanic ash layer in the lake sediments of the 330-metre-thick crater filling in the Ries. "This is surprising, as volcanic rocks were not expected here since the circular basin was identified as an asteroid crater," says first author Professor Gernot Arp from the Geosciences Centre at the University of Göttingen. "The ash was blown in from a volcano 760 kilometres further east in Hungary. The age of the ash can be dated to 14.2 million years ago," adds his colleague and co-author István Dunkl.

The ash, which in the meantime has transformed into nitrogen-rich silicate minerals, reveals a surprisingly strong bowl-shaped geometry: at the edge of the basin the ash is found at the current ground surface, while in the centre of the basin it comes to rest at a depth of about 220 metres. A subsequent systematic evaluation of drillings and geological mapping has now also revealed an arrangement of concentric rings - the "outcropping strata" - for the Ries crater filling, with the oldest deposits at the rim and the most recent in the centre.

Calculations show that this bedding geometry cannot be explained solely by the fact that the underlying lake sediments are settling. In fact, an additional subsidence of about 135 metres had to be accounted for. This can only be explained by subsidence phenomena of the crater bedrock, which is fractured kilometres deep. While further research is needed to explain the exact mechanisms of this subsidence of the crater floor, a simple model calculation can already show that subsidence of this magnitude is basically possible due to settlement phenomena of the fractured underground rocks. This means that inclined strata in the fillings of craters on Mars can now be better explained, at least for craters that show a close timely association of crater formation, flooding by water, and sedimentation.

COLUMBUS, Ohio - Researchers have found a way to use chaos to help develop digital fingerprints for electronic devices that may be unique enough to foil even the most sophisticated hackers.

Just how unique are these fingerprints? The researchers believe it would take longer than the lifetime of the universe to test for every possible combination available.

"In our system, chaos is very, very good," said Daniel Gauthier, senior author of the study and professor of physics at The Ohio State University.

The study was recently published online in the journal IEEE Access.

The researchers created a new version of an emerging technology called physically unclonable functions, or PUFs, that are built into computer chips.

Gauthier said these new PUFs could potentially be used to create secure ID cards, to track goods in supply chains and as part of authentication applications, where it is vital to know that you're not communicating with an impostor.

"The SolarWinds hack that targeted the U.S. government really got people thinking about how we're going to be doing authentication and cryptography," Gauthier said.

"We're hopeful that this could be part of the solution."

The new solution makes use of PUFs, which take advantage of tiny manufacturing variations found in each computer chip - variations so small that they aren't noticeable to the end user, said Noeloikeau Charlot, lead author of the study and a doctoral student in physics at Ohio State.

"There's a wealth of information in even the smallest differences found on computers chips that we can exploit to create PUFs," Charlot said.

These slight variations - sometimes seen only at the atomic level - are used to create unique sequences of 0s and 1s that researchers in the field call, appropriately enough, "secrets."

Other groups have developed what they thought were strong PUFs, but research showed that hackers could successfully attack them. The problem is that current PUFs contain only a limited number of secrets, Gauthier said.

"If you have a PUF where this number is 1,000 or 10,000 or even a million, a hacker with the right technology and enough time can learn all the secrets on the chip," Gauthier said.

"We believe we have found a way to produce an uncountably large number of secrets to use that will make it next to impossible for hackers to figure them out, even if they had direct access to the computer chip."

The key to creating the improved PUF is chaos, a topic that Gauthier has studied for decades. No other PUFs have used chaos in the way demonstrated in this study, he said.

The researchers created a complex network in their PUFs using a web of randomly interconnected logic gates. Logic gates take two electric signals and use them to create a new signal.

"We are using the gates in a non-standard way that creates unreliable behavior. But that's what we want. We are exploiting that unreliable behavior to create a type of deterministic chaos," Gauthier said.

The chaos amplifies the small manufacturing variations found on the chip. Even the smallest differences, when amplified by chaos, can change the entire class of possible outcomes - in this case, the secrets that are being produced, according to Charlot.

"Chaos really expands the number of secrets that are available on a chip. This will likely confuse any attempts at predicting the secrets," Charlot said.

One key to the process is letting the chaos run just long enough on the chip, according to Gauthier. If you let it run too long, it becomes - well, too chaotic.

"We want the process to run long enough to create patterns that are too complex for hackers to attack and guess. But the pattern must be reproducible so we can use it for authentication tasks," Gauthier said.

The researchers calculated that their PUF could create 1077 secrets. How big is that number? Imagine if a hacker could guess one secret every microsecond - 1 million secrets per second. It would take the hacker longer than the life of the universe, about 20 billion years, to guess every secret available in that microchip, Gauthier said.

As part of the study, the researchers attacked their PUF to see if it could be successfully hacked. They attempted machine learning attacks, including deep learning-based methods and model-based attacks - all of which failed. They are now offering their data to other research groups to see if they can find a way to hack it.

Gauthier said the hope is that PUFs like this could help beef up security against even state-sponsored hacker attacks, which are generally very sophisticated and backed up with a lot of computer resources.

For example, Russia is suspected of backing the SolarWinds hack that was uncovered in December. That hack reportedly gained access to email accounts of officials in the Department of Homeland Security and the department's cybersecurity staff.

"It is a constant battle to come up with technology that can stay ahead of hackers. We are trying to come up with technology that no hacker - no matter your resources, no matter what supercomputer you use - will be able to crack."

The researchers have applied for an international patent for their PUF device.

The goal of the team is to move beyond research and to move quickly to commercialize the technology. Gauthier and two partners recently founded Verilock, with a goal of bringing a product to market within a year.

"We see this technology as a real game changer in cybersecurity. This novel approach to a strong PUF could prove to be virtually un-hackable," said Jim Northup, CEO of Verilock.

A novel way to pinpoint and illuminate bone damage promises to make X-rays more efficient at diagnosing bone and other injuries, Flinders University researchers say.

The new technique, looking at potential biomedical applications of an ancient inorganic salt-based aggregation induced emission (AIE) radio-luminescence material, could open new frontiers in medicine including X-ray dosimetry, bioimaging and advanced applications such as optogenetics, says Professor Youhong Tang, from Flinders University's College of Science and Engineering.

The review article, published by Professor Tang, postdoctoral student Dr Javad Tavokoli, colleagues in Hong Kong and Australian technology company Micro-X and, examined the potential of the AIEgen luminogens (AIEgens) in deep tissue imaging. The study used X-ray testing provided by Adelaide-based Micro-X.

"We were able to use Micro-X advanced X-ray machines at the Tonsley Innovation District to show the benefits of this AIEgen system which can be excited by X-ray (as the radioluminescence emitter) and UV light (as the photoluminescence emitter) compared to current AIEgens which mostly only act as the photoluminescence emitter," he says.

"The study highlighted the disadvantages of autofluorescence, poor signal-to-noise radio, and poor tissue penetration depth of traditional photoluminescence emitters which could be elegantly solved by these radioluminescence luminogens," says Professor of Mechanical and Materials Engineering, Dr Tang.

"Not only do they pinpoint bone and soft tissue damage for better diagnosis and treatment but we suggest further studies could see these AIE-based materials with multifunctionalities used for improved drug delivery, biosensors, bioimaging, and tissue engineering."

Lead author on the journal article in Aggregate, Dr Tavokoli, how based at the Centre for Health Technologies at University of Technology Sydney, says the next generation of fluorescent gels could also capitalise on additional light-emitting properties making them attractive for different applications.

The latest work not only explores a series of inorganic AIE systems but also "fundamentally helps to understand both the unconventional organic and inorganic clusteroluminescence phenomena, Professor Tang concludes.

Membranes that allow certain molecules to quickly pass through while blocking others are key enablers for energy technologies from batteries and fuel cells to resource refinement and water purification. For example, membranes in a battery separating the two terminals help to prevent short circuits, while also allowing the transport of charged particles, or ions, needed to maintain the flow of electricity.

The most selective membranes - those with very specific criteria for what may pass through - suffer from low permeability for the working ion in the battery, which limits the battery's power and energy efficiency. To overcome trade-offs between membrane selectivity and permeability, researchers are developing ways to increase the solubility and mobility of ions within the membrane, therefore allowing a higher number of them to transit through the membrane more rapidly. Doing so could improve the performance of batteries and other energy technologies.

Now, as reported today in the journal Nature, researchers have designed a polymer membrane with molecular cages built into its pores that hold positively charged ions from a lithium salt. These cages, called "solvation cages," comprise molecules that together act as a solvent surrounding each lithium ion - much like how water molecules surround each positively charged sodium ion in the familiar process of table salt dissolving in liquid water. The team, led by researchers at the U.S. Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab), found that solvation cages increased the flow of lithium ions through the membrane by an order of magnitude compared to standard membranes. The membrane could allow high-voltage battery cells to operate at higher power and more efficiently, important factors for both electric vehicles and aircraft.

"While it's been possible to configure a membrane's pores at very small length scales, it's not been possible until now to design sites to bind specific ions or molecules from complex mixtures and enable their diffusion in the membrane both selectively and at a high rate," said Brett Helms, a principal investigator in the Joint Center for Energy Storage Research (JCESR) and staff scientist in Berkeley Lab's Molecular Foundry, who led the work.

The research is supported by JCESR, a DOE Energy Innovation Hub whose mission is to deliver transformational new concepts and materials for electrodes, electrolytes, and interfaces that will enable a diversity of high-performance next-generation batteries for transportation and the grid. In particular, JCESR provided the motivation to understand how ions are solvated in porous polymer membranes used in energy storage devices, Helms said.

To pinpoint a design for a cage in a membrane that would solvate lithium ions, Helms and his team looked to a widely practiced drug discovery process. In drug discovery, it's common to build and screen large libraries of small molecules with diverse structures to pinpoint one that binds to a biological molecule of interest. Reversing that approach, the team hypothesized that by building and screening large libraries of membranes with diverse pore structures, it would be possible to identify a cage to temporarily hold lithium ions. Conceptually, the solvation cages in the membranes are analogous to the biological binding site targeted by small molecule drugs.

Helms' team devised a simple but effective strategy for introducing functional and structural diversity across multiple length scales in the polymer membranes. These strategies included designs for cages with different solvation strengths for lithium ions, as well as arrangements of cages in an interconnected network of pores. "Before our work, a diversity-oriented approach to the design of porous membranes had not been undertaken," said Helms.

Using these strategies, Miranda Baran, a graduate student researcher in Helms' research group and a Ph.D. student in the Department of Chemistry at UC Berkeley and lead author on the paper, systematically prepared a large library of possible membranes at the Molecular Foundry. She and co-authors experimentally screened each one to determine a leading candidate whose specific shape and architecture made its pores best suited for selectively capturing and transporting lithium ions. Then, working with Kee Sung Han and Karl Mueller at the Environmental Molecular Sciences Laboratory, a DOE user facility at Pacific Northwest National Laboratory, Baran and Helms revealed, using advanced nuclear magnetic resonance techniques, how lithium ions flow within the polymer membrane compared to other ions in the battery.

"What we found was surprising. Not only do the solvation cages increase the concentration of lithium ions in the membrane, but the lithium ions in the membrane diffuse faster than their counter anions," said Baran, referring to the negatively charged particles that are associated with the lithium salt when it enters the membrane. The solvation of lithium ions in the cages helped to form a layer that blocked the flow of those anions.

To further understand the molecular reasons for the new membrane's behavior, the researchers collaborated with Artem Baskin, a postdoctoral researcher working with David Prendergast, another investigator in JCESR. They performed calculations, using computing resources at Berkeley Lab's National Energy Research Scientific Computing Center (NERSC), to determine the precise nature of the solvation effect that occurs as lithium ions associate with the cages in the membrane's pores. This solvation effect causes lithium ions to concentrate more in the new membrane than they do in standard membranes without solvation cages.

Finally, the researchers investigated how the membrane performed in an actual battery, and determined the ease with which lithium ions are accommodated or released at a lithium metal electrode during the battery's charge and discharge. Using X-ray tools at Berkeley Lab's Advanced Light Source, they observed lithium flow through a modified battery cell whose electrodes were separated by the new membrane. The X-ray images showed that, in contrast to batteries that used standard membranes, lithium was deposited smoothly and uniformly at the electrode, indicating that the battery charged and discharged quickly and efficiently thanks to the solvation cages in the membrane.

With their diversity-oriented approach to screening possible membranes, the researchers achieved the goal of creating a material that helps to transport ions rapidly without sacrificing selectivity. Parts of the work - including component analysis, gas sorption, and X-ray scattering measurements - were also supported by the Center for Gas Separations Relevant to Clean Energy Technologies, a DOE Energy Frontier Research Center led by UC Berkeley.

Future work by the Berkeley Lab team will expand the library of membranes and screen it for enhanced transport properties for other ions and molecules of interest in clean energy technologies. "We also see exciting opportunities to combine diversity-oriented synthesis with digital workflows for accelerated discovery of advanced membranes through autonomous experimentation," said Helms.

The global production of sheep's milk is one the rise, in the vast majority of cases used to produce cheese. However, a relatively large amount of milk is needed to produce it, so science is looking for ways to increase its yield; that is, to obtain more cheese using less milk.

Immersed in this task, a team from the Department of Animal Production at the University of Cordoba, led by Professor Ana Garzón, has collaborated with the University of Leon in the search for genetic parameters affecting the cheese production of milk from Churra sheep, one of the oldest and most rustic breeds on the Iberian Peninsula.

After analysing traits related to rennet and milk properties (pH, milk yield, fat and protein content) in a sample of more than 1,000 sheep, the research team found a low to moderate heritability of these traits, suggesting that their improvement can be achieved through genetic selection. In addition, the need to consider milk pH at the beginning of the coagulation process as a characteristic to be taken into accountas a selection index for the improvement of Churra quality was confirmed, as it will augment the 'cheesemaking capacity' of the milk from this breed.

The team,formed by Ana Garzón andthe researchers Antonio Figueroa and Javier Caballero-Villalobos, comprise the Dairy Laboratory, where the milk samples of this work were analysed, measuring their pH; the physical-chemical parameters of the milk, such as its proteins, fats, and lactose; and technological parameters, such as coagulation time and curd hardening speed; with the aim of providing information for the selection of values to be included in the genetic selection scheme of the Churra breed in order to obtain ewes that give milk with a higher yield in terms of cheese production.

The UCO Dairy Laboratory | The science that cares for milk

This service, part of the UCO's Department of Animal Production, has been working since 2003 on the study of the composition, quality and technological parameters of ruminant milk with the aim of transferring knowledge to the livestock sector to improve milk quality, productivity and yields.

The Dairy Laboratory specialises in the study of the Manchega breed, which is the most important class of sheep in terms of product quality and economic weight in the sector. In this regard, the search for a faster, cheaper and more efficient method to measure the quality of milk according to its composition is one of their main lines of research, as they are trying to determine whether chromaticity can provide information sufficient for the livestock sector to evaluate milk quickly and cheaply.

They are also striving to solve the problem of water retention in curd, which reduces the milk'syield, thus requiring a lot of milk to obtain cheese. In their latest work, they develop mathematical models to achieve more efficient milk for cheese production. Finally, they analyse the correlation between the health of the sheep's udder and these coagulation parameters of the milk slated for cheese production.

In short, this work directly transfers the science carried out in the Laboratory to the livestock sector, which benefits from improvements in the quality and efficiency of its dairy farms.

It is more complicated than copy and paste, but digital twins could be way of future manufacturing according to researchers from the University of Kentucky. They developed a virtual environment based on human-robot interactions that can mirror the physical set up of a welder and their project. Called a digital twin, the prototype has implications for evolving manufacturing systems and training novice welders. They published their work in the IEEE/CAA Journal of Automatica Sinica (Volume 8, Issue 2, February 2021).

"This human-robot interaction working style helps to enhance the human users' operational productivity and comfort; while data-driven welder behavior analysis benefits further novice welder training," said paper author YuMing Zhang, James R. Boyd Professor in electrical engineering at the University of Kentucky.

The researchers had a human demonstrate welder operations using a manual welding torch and a motion tracker. Their movements are transmitted to a machine that is actually welding. Sensors in the physical welding environment feedback data to the human. The physical environment, and incoming data, is accessible via an augmented virtual reality in which the human can make adjustments accordingly.

"In current developed digital twins, humans are the observers of the physical systems --information flow is one way," Zhang said. "For processes where intelligence from humans is needed, like precise welding, human-robot interaction needs to be integrated with the digital twins such that the humans' operative ability can be enhanced and the roles they play transmit from observers to dominators."

The researchers also tracked the behavior of six welders with different experience levels in the digital twin system. All welders were able to complete the same welding task, to varying levels of satisfaction. Analysis revealed the distinct patterns in the skilled and unskilled welders' operating behaviors and, ultimately, their work.

"The successful pattern recognition in skilled welder operations should help accelerate novice welder training," Zhang said.

The digital twin environment could, for example, provide a safe space for novice welders to practice techniques without the risk of dangerous or costly damages, as the system could be trained to recognize potentially harmful patterns and shut down.

"In future work, we plan to investigate efficient novice welder training based on this developed human-robot interactive welding with the recognized patterns from skilled welders and also upgrade the system to support multi-robot collaboration such that some more complex welding operations can be completed by this system," Zhang said. "As such, the system applicability can be increased greatly."