Welcome to the May 18, 2022, edition of ACM TechNews, providing timely information for IT professionals three times a week.

ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).

To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version. The online version now has a button at the top labeled "Show Headlines."
New Approach for Faster Ransomware Detection
NC State University News
Matt Shipman
May 16, 2022


Engineering researchers at North Carolina State University (NC State) and Hewlett Packard Enterprise have come up with a new technique that can detect ransomware faster than previous systems. The Field-Programmable Gate Array-Accelerated XGBoost Inference for Data Centers using High-Level-Synthesis (FAXID) approach is a hardware-based implementation of the ransomware-detecting XGBoost algorithm. The researchers found FAXID was up to 65.8 times faster than software running XGBoost on a central processing unit, and 5.3 times faster than graphic processing unit-based deployment. NC State's Archit Gajjar said FAXID can allocate security hardware's computational muscle to separate problems. "For example, you could devote a certain percentage of the hardware to ransomware detection and another percentage of the hardware to another challenge—such as fraud detection," he explained.

Full Article

Contact lenses like this one can improve vision, but in the future they might also help treat glaucoma. Smart Contact Lens for Glaucoma Could Release Drugs When Needed
New Scientist
Carissa Wong
May 17, 2022


A prototype wireless contact lens developed by researchers at China's Sun Yat-Sen and Jinan universities could help treat glaucoma by automatically dispensing drugs when pressure inside the eye becomes excessive. The lens's outer layer has a ring of tiny copper plates surrounding the pupil, which detect eye deformation caused by increasing pressure. An antenna positioned near the eye sends this data to a computer, while the lens's inner layer contains the pressure-lowering drug brimonidine, which can be discharged via a signal from the computer. In tests on rabbits, the researchers found the animals' eye pressure declined by roughly a third after wearing the lenses for 30 minutes, and by more than 40% after wearing them for two hours, on average.

Full Article

Researchers added cameras to Quest 2 controllers to enable full body tracking in virtual reality. A Solution for One of VR’s Biggest Issues: Tracking Your Legs
CNN
Rachel Metz
May 12, 2022


Carnegie Mellon University (CMU) researchers have come up with a potential solution to the problem of tracking legs in virtual reality (VR). The ControllerPose system tracks a user's entire body by attaching cameras with fisheye lenses to two Meta Quest 2 headset controllers; as a VR headset wearer holds the VR system controllers, body images recorded by the cameras are streamed to a computer that controls a full-body avatar. Software screens out bad images and corrects lens distortion, then combines images in order to estimate the headset wearer's actual poses. CMU's Chris Harrison said ControllerPose cannot yet capture fine movements, but it performs "quite well" for tracking coarser actions.

Full Article

Robot-equipped beehives in an Israeli kibbutz aim to reduce bee mortality rates as bee species worldwide face an increasing threat of decline. Robot Hives in Israel Kibbutz Hope to Keep Bees Buzzing
Yahoo! News
Alexandra Vardi
May 17, 2022


Artificial intelligence-equipped apiaries at Kibbutz Beit Haemek in Israel's Galilee region are designed to maintain bee health. Beekeeper Shlomki Frankin said the Beewise startup's Beehome project sustains up to 24 hives using a multipurpose robot designed to keep the colonies viable. "The robot is equipped with sensors that allow it to know what is happening in the hive frames," said Beewise's Netaly Harari. "Thanks to artificial intelligence, our software knows what the bees need." The robots can automatically distribute sugar, water, and medication, while an application alerts beekeepers to problems. Harari said the solar-powered hives have adjustable temperatures, remove pests, and can extract honey via an integrated centrifuge.

Full Article

A machine-learning approach can learn to control a fleet of autonomous vehicles in a way that keeps traffic flowing smoothly. On the Road to Cleaner, Greener, Faster Driving
MIT News
Adam Zewe
May 17, 2022


Massachusetts Institute of Technology (MIT) researchers have developed a machine learning approach that can learn to control autonomous vehicle fleets so they arrive at intersections during green lights. The researchers found through simulations that the control system could increase travel speeds by 20% and reduce fuel consumption by 18% and carbon dioxide emissions by 25% if all vehicles on the road were autonomous. However, MIT's Cathy Wu said, "If we only control 25% of vehicles, that gives us 50% of the benefits in terms of fuel and emissions reduction. That means we don't have to wait until we get to 100% autonomous vehicles to get benefits from this approach."

Full Article
3D-Printed Acoustic Holograms Against Alzheimer's or Parkinson's
Universitat Politecnica de Valencia (Spain)
May 17, 2022


A team of researchers from Spain's Universitat Politècnica de València (UPV), the Spanish National Research Council, and Columbia University has created customizable, three-dimensionally-printed acoustic holograms that could be used to treat diseases like Alzheimer's and Parkinson's. The holograms focus ultrasound to open the blood-brain barrier in a controlled manner, so drugs that target central nervous system-affecting pathologies can pass through. They also correct for skull-induced aberrations, and can produce an ultrasonic multi-focal beam in critical brain structures. UPV's Noé Jiménez said the ultrasonic beam "focuses and adapts bilaterally and very precisely on parts of the brain that are of great therapeutic interest.”

Full Article
Body Cameras, Live Streaming Bring Search and Rescue into the Next Generation
Simon Fraser University (Canada)
Matt Kieltyka
May 16, 2022


The RescueCASTR system developed by researchers at Canada's Simon Fraser University (SFU) can improve the coordination of wilderness search and rescue (SAR) efforts. The system features 360-degree body cameras worn by rescuers that transmit live video and photos to a central command post. Coordinators at the central command post can access an interactive program that combines three-dimensional map data, live streams from each field team, and a timeline of milestone events and photos. Said SFU's Carman Neustaedter, "Search and rescue operations happen year-round and are often life-critical. It is highly important that SAR team members have ways to easily share information they come across in order to productively search for and find missing people in the wilderness.”

Full Article

A building destroyed by an earthquake in Izmir, Turkey, in October 2020. Quantum Communication System Could Detect Earthquakes
New Scientist
Alex Wilkins
May 17, 2022


University of Science and Technology of China (USTC) scientists demonstrated twin-field quantum key distribution (QKD), a process that can measure tiny vibrations in the ground in order to potentially detect landslides and earthquakes. Twin-field QKD can encrypt data by leveraging how single photons interfere with each other; the researchers used the method to transmit encrypted data over a 658-km. (408-mile) cable, with minimal data loss. The rate of data transfer needs to be improved before the technology can be built into a large-scale quantum communication network, said Timothy Spiller at the U.K.’s University of York, adding that if such improvement can be achieved, vibration sensing could be a useful by-product.

Full Article
Searching for Features to ID Modern, Fossil Leaves
Penn State News
Matthew Carroll
May 12, 2022


Researchers at Pennsylvania State (Penn State) and Brown universities combined machine learning (ML) and botanical terminology to find and describe new features for identifying fossil leaves. The researchers analyzed ML-generated heat maps across different plant families using a manual scoring system. Penn State's Edward Spagnuolo examined more than 3,000 heat maps highlighting leaves of 930 genera in 14 angiosperm families, scoring the top-five and top-one hot spot regions, and describing their locations on the leaves in traditional botanical language. The scientists said ML extrapolated hot spots on families like Rosaceae correlated with traditional observations, while the computer highlighted details on leaves in other families that lack distinctive features and are largely unidentified in the fossil record. Said Spagnuolo, "These new features can lead to additional studies to hopefully delineate new fossil-identifying characters."

Full Article

The common skin cancer squamous cell carcinoma. Melding Data to 3D-Map Cells' Activities
Princeton University School of Engineering and Applied Science
Molly Sharlach
May 16, 2022


Princeton University researchers have developed a method of integrating information from multiple slices of the same tissue sample to generate a three-dimensional view of gene expression within a tumor or developing organ, which could help identify rare cell types or more precisely determine cancer treatment options. The PASTE (Probabilistic Alignment of ST Experiments) method also can combine information from multiple tissue slices into a two-dimensional consensus slice with richer gene expression information, and categorize cells based on their gene expression profiles while preserving the physical location of the cells within the tissue. Said lead study author Ron Zeira, now at the precision health company Verily, "After we integrate several of these slices and effectively increase the coverage of the data, we get more spatially coherent groupings of cells, which is more reasonable than every cell type being randomly positioned in the tissue."

Full Article
Algorithm Trained to Detect Unhappiness on Social Networks
Universitat Oberta de Catalunya News (Spain)
Beatriz de Vera
May 11, 2022


Scientists at Spain's Universitat Oberta de Catalunya have created a deep learning algorithm that analyzes social network content to help diagnose users' potential for mental health problems. The researchers trained the model to identify image content and to categorize text by assigning labels proposed by psychologists, who compared the results to a database containing roughly 30,000 images, captions, and comments. The researchers analyzed 86 Instagram profiles in Spanish and Persian, and the results of their analysis supported the idea that human choices do not always stem from one basic need. The researchers suggested, "Studying data from social networks that belong to non-English speaking users could help build inclusive and diverse tools and models for addressing mental health problems in people with diverse cultural or linguistic backgrounds."

Full Article
Eavesdroppers Can Hack 6G Frequency with DIY Metasurface
Rice University News
Jade Boyd
May 16, 2022


Hackers can use common tools to construct a metasurface that allows them to listen in on 6G wireless transmissions. Researchers at Rice and Brown universities demonstrated that attackers could employ a sheet of office paper covered with two-dimensional foil symbols to reroute part of a 150-gigahertz "pencil beam" signal between two users, calling it a Metasurface-in-the-Middle exploit. In such a situation, the eavesdropper designs a metasurface to diffract part of a signal to their location; Rice's Zhambyl Shaikhanov said they then laser-print the metasurface by feeding metal foil through a laminator. Brown's Daniel Mittleman said the hot-stamping technique was developed to simplify metasurface manufacturing for quick, affordable testing. Warns Rice's Edward Knightly, "Next-generation wireless will use high frequencies and pencil beams to support wide-band applications like virtual reality and autonomous vehicles."

Full Article
Government Websites, Apps Use Same Tracking Software as Commercial Sites
Concordia University (Canada)
Patrick Lejtenyi
May 17, 2022


Researchers at Canada's Concordia University discovered that both governments and businesses worldwide are using the same data-harvesting software to track users of their websites. The researchers analyzed more than 150,000 government websites from 206 countries and more than 1,150 Android applications from 71 countries. They estimated 17% of government sites and 37% of government Android apps host Google trackers; 27% of Android apps also leak sensitive information to third parties or potential hackers. Concordia's Mohammad Mannan said some government tracker usage may be unintentional, as developers likely use existing software suites to build sites and apps containing tracking scripts or including links to tracker-filled social media sites. Said Mannan, "Governments are becoming more aware of online threats to privacy, but at the same time, they are enabling these potential violations through their own services."

Full Article
Concurrency:  The Works of Leslie Lamport
 
ACM Special Interest Groups
 

Association for Computing Machinery

1601 Broadway, 10th Floor
New York, NY 10019-7434
1-800-342-6626
(U.S./Canada)



ACM Media Sales

If you are interested in advertising in ACM TechNews or other ACM publications, please contact ACM Media Sales or (212) 626-0686, or visit ACM Media for more information.

To submit feedback about ACM TechNews, contact: [email protected]