Computers Math

The impact of deploying Artificial Intelligence (AI) for radiation cancer therapy in a real-world clinical setting has been tested by Princess Margaret researchers in a unique study involving physicians and their patients.
A team of researchers directly compared physician evaluations of radiation treatments generated by an AI machine learning (ML) algorithm to conventional radiation treatments generated by humans.
They found that in the majority of the 100 patients studied, treatments generated using ML were deemed to be clinically acceptable for patient treatments by physicians.
Overall, 89% of ML-generated treatments were considered clinically acceptable for treatments, and 72% were selected over human-generated treatments in head-to-head comparisons to conventional human-generated treatments.
Moreover, the ML radiation treatment process was faster than the conventional human-driven process by 60%, reducing the overall time from 118 hours to 47 hours. In the long term this could represent a substantial cost savings through improved efficiency, while at the same time improving quality of clinical care, a rare win-win.
The study also has broader implications for AI in medicine. More

Researchers have demonstrated “giant flexoelectricity” in soft elastomers that could improve robot movement range and make self-powered pacemakers a real possibility. In a paper published this month in the Proceedings of the National Academy of Sciences, scientists from the University of Houston and Air Force Research Laboratory explain how to engineer ostensibly ordinary substances like silicone rubber into an electric powerhouse.
What do the following have in common: a self-powered implanted medical device, a soft human-like robot and how we hear sound? The answer as to why these two disparate technologies and biological phenomena are similar lies in how the materials they are made of can significantly change in size and shape — or deform — like a rubber band, when an electrical signal is sent.
Some materials in nature can perform this function, acting as an energy converter that deforms when an electrical signal is sent through or supplies electricity when manipulated. This is called piezoelectricity and is useful in creating sensors and laser electronics, among several other end uses. However, these naturally occurring materials are rare and consist of stiff crystalline structures that are often toxic, three distinct drawbacks for human applications.
Human-made polymers offer steps toward alleviating these pain points by eliminating material scarcity and creating soft polymers capable of bending and stretching, known as soft elastomers, but previously those soft elastomers lacked significant piezoelectric attributes.
In a paper published this month in the Proceedings of the National Academy of Sciences, Kosar Mozaffari, graduate student at the Cullen College of Engineering at the University of Houston; Pradeep Sharma, M.D. Anderson Chair Professor & Department Chair of Mechanical Engineering at the University of Houston and Matthew Grasinger, LUCI Postdoctoral Fellow at the Air Force Research Laboratory, offer a solution.
“This theory engineers a connection between electricity and mechanical motion in soft rubber-like materials,” said Sharma. “While some polymers are weakly piezoelectric, there are no really soft rubber like materials that are piezoelectric.”
The term for these multifunctional soft elastomers with increased capability is “giant flexoelectricity.” In other words, these scientists demonstrate how to boost flexoelectric performance in soft materials. More

A new formula from Army scientists is leading to new insights on how to build an energy-efficient legged teammate for dismounted warfighters.
In a recent peer-reviewed PLOS One paper, the U.S. Army Combat Capabilities Development Command, known as DEVCOM, Army Research Laboratory’s Drs. Alexander Kott, Sean Gart and Jason Pusey offer new insights on building autonomous military robotic legged platforms to operate as efficiently as any other ground mobile systems.
Its use could lead to potentially important changes to Army vehicle development. Scientists said they may not know exactly why legged, wheeled and tracked systems fit the same curve yet, but they are convinced their findings drive further inquiry.
“If vehicle developers find a certain design would require more power than is currently possible given a variety of real-world constraints, the new formula could point to specific needs for improved power transmission and generation, or to rethink the mass and speed requirements of the vehicle,” Gart said.
Inspired by a 1980s formula that shows relationships between the mass, speed and power expenditure of animals, the team developed a new formula that applied to a very broad range of legged, wheeled and tracked systems — such as motor vehicles and ground robots.
Although much of the data has been available for 30 years, this team believes they are the first to actually assemble it and study the relationships that emerge from this data. Their findings show that legged systems are as efficient as wheeled and tracked platforms. More

Scientists have created a cybersecurity technology called Shadow Figment that is designed to lure hackers into an artificial world, then stop them from doing damage by feeding them illusory tidbits of success.
The aim is to sequester bad actors by captivating them with an attractive — but imaginary — world.
The technology is aimed at protecting physical targets — infrastructure such as buildings, the electric grid, water and sewage systems, and even pipelines. The technology was developed by scientists at the U.S. Department of Energy’s Pacific Northwest National Laboratory.
The starting point for Shadow Figment is an oft-deployed technology called a honeypot — something attractive to lure an attacker, perhaps a desirable target with the appearance of easy access.
But while most honeypots are used to lure attackers and study their methods, Shadow Figment goes much further. The technology uses artificial intelligence to deploy elaborate deception to keep attackers engaged in a pretend world — the figment — that mirrors the real world. The decoy interacts with users in real time, responding in realistic ways to commands.
“Our intention is to make interactions seem realistic, so that if someone is interacting with our decoy, we keep them involved, giving our defenders extra time to respond,” said Thomas Edgar, a PNNL cybersecurity researcher who led the development of Shadow Figment. More

The myriad processes occurring in biological cells may seem unbelievably complex at first glance. And yet, in principle, they are merely a logical succession of events, and could even be used to form digital circuits. Researchers have now developed a molecular switching circuit made of DNA, which can be used to mechanically alter gels, depending on the pH. DNA-based switching circuits could have applications in soft robotics, say the researchers in their article in Angewandte Chemie.
DNA is a long molecule that can be folded and twisted in various ways. It has a backbone and bases that stick out from the backbone and pair up with counterparts in other DNA strands. When a series of these matching pairs comes together, they form a twisted, ladder-like double strand — the familiar DNA double helix. The flexibility of DNA, which makes it possible to produce bends, loops, and a wide variety of other shapes, has inspired researchers to build DNA switches. These switches change shape after receiving an input, and can then affect their surroundings.
Hao Pei from Shanghai Key Laboratory of Green Chemistry and Chemical Processes at the East China Normal University in Shanghai, China, and colleagues have now developed a configurable, multi-mode logic switching network that reacts differently with its surroundings depending on pH and DNA input. All the components of the switching circuit were produced from DNA.
The team developed a series of four DNA switches, each with slightly different lengths and combinations of bases. These variations meant they reacted differently with a single DNA strand depending on the pH of their surroundings. For example, at a slightly alkaline pH of 8, two of the switches formed triple-stranded DNA (triplexes), while the others remained loosely stretched out. These reactions and folds led to secondary reactions, which were utilized by the researchers as logic functions in the switching circuit. The result was, for example, a fluorescent signal that could be read as an output.
To demonstrate the use of the switching circuit in a real mechanical system, the team incorporated the DNA switches into polyacrylamide gels. The DNA acted as a crosslinker, joining the polymer molecules in the gel together. The shorter the crosslinker, or the more folded the DNA, the denser the gel became. Once a piece of DNA with matching bases was added as an input, a logic circuit was set in place, causing the DNA switches to unfold, form triplexes, or relax. The reaction circuit was also dependent on the pH. As a result, certain combinations of DNA input and pH range caused the DNA crosslinker to grow longer and the gel to swell up, in some cases nearly doubling in size.
As DNA switches have almost infinite possibilities for combinations of twists and folds, the researchers consider their switching circuits to be a vital step toward soft matter robotics, where controllable, miniaturized logic functional networks are important.
Story Source:
Materials provided by Wiley. Note: Content may be edited for style and length. More

What makes cheetah the fastest land mammal? Why aren’t other animals, such as horses, as fast? While we haven’t yet figured out why, we have some idea about how — cheetahs, as it turns out, make use of a “galloping” gait at their fastest speeds, involving two different types of “flight”: one with the forelimbs and hind limbs beneath their body following a forelimb liftoff, called “gathered flight,” while another with the forelimbs and hind limbs stretched out after a hind limb liftoff, called “extended flight”. Of these, the extended flight is what enables cheetahs to accelerate to high speeds, and it depends on ground reaction forces satisfying specific conditions; in the case of horses, the extended flight is absent.
Additionally, cheetahs show appreciable spine movement during flight, alternating between flexing and stretching in gathered and extended modes, respectively, which contributes to its high-speed locomotion. However, little is understood about the dynamics governing these abilities.
“All animal running constitutes a flight phase and a stance phase, with different dynamics governing each phase,” explains Dr. Tomoya Kamimura from Nagoya Institute of Technology, Japan, who specializes in intelligent mechanics and locomotion. During the flight phase, all feet are in the air and the center of mass (COM) of the whole body exhibits ballistic motion. Conversely, during the stance phase, the body receives ground reaction forces through the feet. “Due to such complex and hybrid dynamics, observations can only get us so far in unraveling the mechanisms underlying the running dynamics of animals,” Dr. Kamimura says.
Consequently, researchers have turned to computer modeling to gain a better dynamic perspective of the animal gait and spine movement during running and have had remarkable success using fairly simple models. However, few studies so far have explored the types of flight and spine motion during galloping (as seen in a cheetah). Against this backdrop, Dr. Kamimura and his colleagues from Japan have now addressed this issue in a recent study published in Scientific Reports, using a simple model emulating vertical and spine movement.
The team, in their study, employed a two-dimensional model comprising two rigid bodies and two massless bars (representing the cheetah’s legs), with the bodies connected by a joint to replicate the bending motion of the spine and a torsional spring. Additionally, they assumed an anterior-posterior symmetry, assigning identical dynamical roles to the fore and hind legs.
By solving the simplified equations of motion governing this model, the team obtained six possible periodic solutions, with two of them resembling two different flight types (like cheetah galloping) and four, only one flight type (unlike cheetah galloping), based on the criteria related to the ground reaction forces provided by the solutions themselves. Researchers then verified these criteria with measured cheetah data, revealing that cheetah galloping in the real world indeed satisfied the criterion for two flight types through spine bending .
Additionally, the periodic solutions also revealed that horse galloping only involves gathered flight due to restricted spine motion, suggesting that the additional extended flight in cheetahs combined with spine bending allowed them to achieve such great speeds!
“While the mechanism underlying this difference in flight types between animal species still remains unclear, our findings extend the understanding of the dynamic mechanisms underlying high-speed locomotion in cheetahs. Furthermore, they can be applied to the mechanical and control design of legged robots in the future,” speculates an optimistic Dr. Kamimura.
Cheetahs inspiring legged robots! Who would’ve thought?
Story Source:
Materials provided by Nagoya Institute of Technology. Note: Content may be edited for style and length. More

Can you trust the map on your smartphone, or the satellite image on your computer screen?
So far, yes, but it may only be a matter of time until the growing problem of “deep fakes” converges with geographical information science (GIS). Researchers such as Associate Professor of Geography Chengbin Deng are doing what they can to get ahead of the problem.
Deng and four colleagues — Bo Zhao and Yifan Sun at the University of Washington, and Shaozeng Zhang and Chunxue Xu at Oregon State University — co-authored a recent article in Cartography and Geographic Information Science that explores the problem. In “Deep fake geography? When geospatial data encounter Artificial Intelligence,” they explore how false satellite images could potentially be constructed and detected. News of the research has been picked up by countries around the world, including China, Japan, Germany and France.
“Honestly, we probably are the first to recognize this potential issue,” Deng said.
Geographic information science (GIS) underlays a whole host of applications, from national defense to autonomous cars, a technology that’s currently under development. Artificial intelligence has made a positive impact on the discipline through the development of Geospatial Artificial Intelligence (GeoAI), which uses machine learning — or artificial intelligence (AI) — to extract and analyze geospatial data. But these same methods could potentially be used to fabricate GPS signals, fake locational information on social media posts, fabricate photographs of geographic environments and more.
In short, the same technology that can change the face of an individual in a photo or video can also be used to make fake images of all types, including maps and satellite images. More

Researchers in Penn State’s College of Agricultural Sciences have developed a robotic mechanism for mushroom picking and trimming and demonstrated its effectiveness for the automated harvesting of button mushrooms.
In a new study, the prototype, which is designed to be integrated with a machine vision system, showed that it is capable of both picking and trimming mushrooms growing in a shelf system.
The research is consequential, according to lead author Long He, assistant professor of agricultural and biological engineering, because the mushroom industry has been facing labor shortages and rising labor costs. Mechanical or robotic picking can help alleviate those problems.
“The mushroom industry in Pennsylvania is producing about two-thirds of the mushrooms grown nationwide, and the growers here are having a difficult time finding laborers to handle the harvesting, which is a very labor intensive and difficult job,” said He. “The industry is facing some challenges, so an automated system for harvesting like the one we are working on would be a big help.”
The button mushroom — Agaricus bisporus — is an important agricultural commodity. A total of 891 million pounds of button mushrooms valued at $1.13 billion were consumed in the U.S. from 2017 to 2018. Of this production, 91% were for the fresh market, according to the U.S. Department of Agriculture, and were picked by hand, one by one, to ensure product quality, shelf life and appearance. Labor costs for mushroom harvesting account for 15% to 30% of the production value, He pointed out.
Developing a device to effectively harvest mushrooms was a complex endeavor, explained He. In hand-picking, a picker first locates a mature mushroom and detaches it with one hand, typically using three fingers. A knife, in the picker’s other hand, is then used to remove the stipe end. Sometimes the picker waits until there are two or three mushrooms in hand and cuts them one by one. Finally, the mushroom is placed in a collection box. A robotic mechanism had to achieve an equivalent picking process.
The researchers designed a robotic mushroom-picking mechanism that included a picking “end-effector” based on a bending motion, a “4-degree-of-freedom positioning” end-effector for moving the picking end-effector, a mushroom stipe-trimming end-effector, and an electro-pneumatic control system. They fabricated a laboratory-scale prototype to validate the performance of the mechanism.
The research team used a suction cup mechanism to latch onto mushrooms and conducted bruise tests on the mushroom caps to analyze the influence of air pressure and acting time of the suction cup.
The test results, recently published in Transactions of the American Society of Agricultural and Biological Engineers, showed that the picking end-effector was successfully positioned to the target locations and its success rate was 90% at first pick, increasing to 94.2% after second pick.
The trimming end-effector achieved a success rate of 97% overall. The bruise tests indicated that the air pressure was the main factor affecting the bruise level, compared to the suction-cup acting time, and an optimized suction cup may help to alleviate the bruise damage, the researchers noted. The laboratory test results indicated that the developed picking mechanism has potential to be implemented in automatic mushroom harvesting.
Button mushrooms for the study were grown in tubs at Penn State’s Mushroom Research Center on the University Park campus. Fabrication and experiments were conducted at the Fruit Research and Extension Center in Biglerville. A total of 70 picking tests were conducted to evaluate the robotic picking mechanism. The working pressures of the pneumatic system and the suction cup were set at 80 and 25 pounds per square inch, respectively.
Story Source:
Materials provided by Penn State. Original written by Jeff Mulhollem. Note: Content may be edited for style and length. More