Welcome to the October 4, 2021 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."
|
|
|
Robots Take Over Italy's Vineyards as Wineries Struggle with COVID-19 Worker Shortages
The Wall Street Journal
Ian Lovett
October 3, 2021
Vintners are turning to automation amid pandemic-driven labor shortages. Italian winemaker Mirko Cappelli purchased an automated grape harvester from French manufacturer Pellenc for his Tuscan vineyard, which shakes vine rows and sucks up the grapes that fall. Cappelli says he and his father completed harvesting their 13 hectares in about 10 days with the machine, compared to about 18 with handpickers. Thiébault Huber with a French vintners' trade group said farmers have doubts about the quality of the grapes the machines pick, and machine harvesting has been banned in France's Champagne region. Yet farmers in Italy's Valdesa region say robot harvesters do at least as good a job harvesting as people.
*May Require Paid Registration
|
UN Tool Maps Floods Since 1985 to Street Level, Will Aid Disaster Planning
United Nations University
October 1, 2021
A new online tool developed by researchers at United Nations University and various partners will help countries, especially those in the Global South, plan for disasters more effectively. The World Flood Mapping Tool produces instant, accurate 30-meter street-level resolution maps of floods worldwide since 1985. Users can tweak variables to help locate gaps in flood defenses and responses, and to plan future urban and agricultural development, disaster readiness, and supply-chain improvements. The tool uses the Google Earth Engine combined with decades of Landsat data; layers of data for a chosen region and specified timeframe identifies temporary and permanent water bodies while integrating site-specific elevation and land-use information, yielding a map of flooding in recent decades.
|
Ocean Drone Captures Video From Inside Hurricane
National Oceanic and Atmospheric Administration
September 30, 2021
An uncrewed surface vehicle (USV) built by California-based company Saildrone shot video from inside a hurricane as part of a U.S. National Oceanic and Atmospheric Administration (NOAA) project. The Saildrone Explorer SD 1045 was piloted into the category 4 Hurricane Sam in the Atlantic Ocean, and collected real-time observations for hurricane prediction models. Five saildrones operate in the Atlantic during hurricane season, gathering knowledge vital to improving storm forecasts. The USVs supply data directly to NOAA's Pacific Marine Environmental Laboratory and Atlantic Oceanographic and Meteorological Laboratory. NOAA's Greg Foltz said, "New data from saildrones and other uncrewed systems that NOAA is using will help us better predict the forces that drive hurricanes and be able to warn communities earlier."
|
Google, DeepMind Face Lawsuit Over Deal with Britain's National Health Service
CNBC
Sam Shead
October 1, 2021
Alphabet subsidiaries Google and DeepMind face litigation for obtaining and processing over a million patient health records in the U.K. without consent. British law firm Mishcon de Reya said it had filed suit with the High Court on behalf of about 1.6 million plaintiffs, whose medical records DeepMind acquired to develop a patient monitoring application called Streams. New Scientist previously disclosed that DeepMind's agreement with the U.K.'s National Health Service (NHS) exceeded what was publicly announced. The U.K. Information Commissioner's Office decreed the agreement did not comply with that nation’s data protection law, yet a subsequent audit by law firm Linklaters determined the Royal Free London NHS Foundation Trust's use of Streams was legal, and complied with the statutes. Mishcon's Ben Lasserson said the planned lawsuit "should help to answer fundamental questions about the handling of sensitive personal data."
|
Widely Used Bitcoin ATMs Have Major Security Flaws
Gizmodo
Tom McKay
September 30, 2021
Security researchers at crypto exchange Kraken warn that many bitcoin ATMs contain serious vulnerabilities that hackers could exploit. Kraken found software and hardware flaws within the General Bytes BATMtwo (GBBATM2) ATM model; Coin ATM Radar calculates that General Bytes has provided nearly 23% of all crypto ATMs globally, including 18.5% of U.S. units and 65.4% of European units. Owners have installed many such ATMs without changing the default admin quick response (QR) code that functions as a password, which is shared across units. Kraken also cited a lack of secure boot mechanisms, enabling hackers to fool GBBATM2s into running malware, as well as "critical vulnerabilities in the ATM management system." The exchange recommends bitcoin ATM users conduct cryptocurrency transactions in trustworthy locations overseen by surveillance cameras, and for operators to change the default QR code.
|
Virtual Care with Remote Monitoring Catches Drug Errors, Reduces Patient Pain
McMaster University (Canada)
September 30, 2021
Researchers at Canada's McMaster University found that virtual care and remote automated monitoring (RAM) can help reduce patients' pain and increase detection and correction of drug errors after non-elective surgery. As part of the study of 905 post-surgery patients, half used a cellular tablet and RAM equipment at home to measure their vital signs for 30 days after leaving the hospital, along with 24/7 virtual access to providers, while the other half received standard care involving an in-person visit within 30 days of discharge and encouraging them to call their surgeon with any concerns. The researchers found that only 22% of patients with the take-home technology returned to the hospital, versus 27% of patients receiving standard care. In addition, 30% of the virtual care group had a medication error detected and 28% had one corrected, compared with 6% and 4% of the standard care group, respectively.
|
U.S. Soybean, Corn Yields Could Increase Through ML
Penn State News
Amy Duke
September 29, 2021
Research by American and Greek scientists suggests a machine learning (ML) algorithm could help to increase U.S. soybean and corn yields. The researchers sifted through crop yield and management data from trials conducted in 28 states between 2016 and 2018 for corn and between 2014 and 2018 for soybeans. They also analyzed correlations among weather-related factors, determining seven variables for corn and eight for soybean for specific regions. The researchers developed algorithms or datasets for each crop, which they applied to various models and tested over two growing seasons in a randomly selected field in Wisconsin. Spyridon Mourtzinis at Greece's Agstat Consulting said, "Our approach can accelerate agricultural research, identify sustainable practices, and help overcome future food demands."
|
Exoskeleton Research Demonstrates Importance of Training
Stanford News
Taylor Kubota
September 29, 2021
Stanford University scientists found training is critical for reaping the benefits of exoskeleton use. Stanford's Katherine Poggensee and Steve Collins tracked the progress of 15 people equipped with an ankle exoskeleton emulator that accommodated customization and refinement through laboratory-based controllers. One group of participants received generic assistance from the emulator; a second group received continuous optimization of the device to their specific needs, and a third group underwent daily optimization resets. Subjects receiving continuous optimization and exposure to moderate variety spent nearly 40% less energy than those walking with the exoskeleton deactivated. Training also contributed about half of the overall benefit offered by the exoskeleton across all participants, while about 25% was due to customization.
|
Tying Quantum Computing to AI Prompts Smarter Power Grid
Cornell Chronicle
Blaine Friedlander
September 29, 2021
Combining quantum computing with artificial intelligence could lead to the rapid diagnosis and identification of remedies for electrical grid problems, according to Cornell University's Fengqi You and Akshay Ajagekar. The researchers proposed a hybrid quantum-computing-based deep learning "intelligent system" design for a fault-diagnosis framework that can accurately pinpoint problems in electrical power systems. They demonstrated the solution in a large-scale Institute of Electrical and Electronics Engineers test electric grid; the hybrid approach could be scaled efficiently for quick diagnosis in larger power systems without sacrificing performance. Ajagekar said, "Integrating quantum computing with intelligence—even though it is not yet a mature technology—will solve real problems now."
|
Deep Learning Framework to Enable Material Design in Unseen Domain
KAIST (South Korea)
September 29, 2021
Researchers at South Korea's KAIST and the University of California, Berkeley have developed a framework that uses a deep neural network to facilitate more efficient material or structure design beyond the domain of the initial training set. The method compensates for the weak predictive power of neural networks. This involves three steps: • using genetic algorithms to search for candidates with improved properties close to the training set, and mixing superior designs in the training set; • determining whether the candidates actually have improved properties, and using data augmentation to duplicate validated designs and expand the training set; and • using transfer learning to update the neural network with newly generated superior designs to broaden the reliable prediction domain. Researchers are using the optimization framework to design metamaterial structures, segmented thermoelectric generators, and optimal sensor distributions.
|
X-Ray Technology Reveals Marie Antoinette's Censored Secret Correspondence
Smithsonian
Brian Handwerk
October 1, 2021
Scientists have used modern technology to retrieve secret correspondence from French queen Marie-Antoinette to Swedish count Axel von Fersen, that a censor blotted out with dark ink. Anne Michelin and colleagues at the French National Museum of Natural History used X-ray fluorescence spectroscopy to generate fluorescent fingerprints of the inks, and mapped the distribution and ratios of different components in the inks of the original letter writers and of the censor; this made the redacted words more readable. The researchers also used data processing to decipher more difficult sections, and trained algorithms to produce images that maximized the inks' chemical differences, to enhance the text's legibility. The University of Wisconsin-Madison's Uwe Bergmann said this research "will have an impact on people who use all kinds of imaging techniques and can learn from the application of these approaches to get exceptional results."
|
Dissolved Salt Can Reassemble at Nanoscale, Simulations Show
University of Nebraska-Lincoln
Scott Schrage
October 1, 2021
The University of Nebraska-Lincoln's Xiao Cheng Zeng and colleagues ran computer models to simulate the behavior of sodium chloride and lithium chloride when submerged in a nanoscopic stream of water enclosed by two water-repellent walls. Zeng said following their initial dissolution, the substances would spontaneously reassemble into two-dimensional layers, with sodium chloride reconstituting into its solid state and lithium chloride forming hexagonal rings. The researchers theorize this unexpected response is partly due to nanoscale confinement reducing interaction strength between a charged atom and the water molecules that typically cohere into a surrounding hydration shell. The team hopes these predictions will spur other researchers to run experiments that confirm or challenge its simulation, which Zeng said may lead to nanofluidic devices that transport charged atoms to reproduce neuronal activity.
|
|
