Welcome to the April 28, 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."
Rapid Covid-19 Test the Result of University-Industry Partnership
UC Davis Health
April 26, 2021


Researchers at the University of California, Davis (UC Davis) have developed a rapid Covid-19 test that combines mass spectrometry with robotics and an automated machine learning platform to detect SARS-CoV-2 in nasal swabs. The test, which takes about 20 minutes, was found to be 98.3% accurate for positive tests and 96% for negative tests, matching or outperforming many existing Covid-19 screening tests. The researchers used a mass spectrometer from Shimadzu Scientific Instruments to ionize 226 nasal swabs, and the hundreds of peaks and signals produced by the ionized particles were analyzed by the automated machine learning platform MILO (Machine Intelligence Learning Optimizer) to detect patterns that correspond to the presence or absence of the virus in the samples. SpectraPass, a startup launched by Allegiant Travel Company's Maurice J. Gallagher, Jr., a UC Davis alumnus, will produce the testing system.

Full Article

Security camera at co-working space in Houston How Data Is Changing the Way Offices Are Run
The New York Times
Patrick Sisson
April 27, 2021


Commercial real estate developers are tapping property technology (proptech) to use data to improve office buildings, in order to reduce costs and streamline operations. Property managers are applying proptech to enhance control systems like heating, lighting, air quality, and even the flow of workers via data collection and artificial intelligence (AI). Said Charlie Kuntz at real estate investment firm Hines, "There will be a dramatic increase in the information we have about how people use our buildings, and sensors will be more common." One example is Swedish developer Skanska's planned office tower in Houston, which will have environmental controls and smart building features, including a network of sensors tracking movement, occupancy, and efficiency for AI analysis. Data collection raises issues of privacy and cybersecurity, and Hines and other property developers say they strictly use anonymized data and do not track individuals.

Full Article
*May Require Paid Registration

Figure from Computational Design and Modeling of Nanobodies toward SARS CoV 2 Receptor Binding Domain Researchers Develop 11 SARS-CoV-2 RBD Nanobodies Using Computational Design
News-Medical.Net
Rayma Dwivedi
April 27, 2021


A team of Chinese researchers computationally designed nanobodies (Nbs) to block the interaction of the SARS-CoV-2 virus's receptor-binding domain (RBD) with the host cell's angiotensin-converting 2 receptor. The Nbs were generated using neutralizing/monoclonal antibodies specific to RBD of the spike protein. The researchers designed the Nbs by grafting the complementarity-determining regions of developed SARS-CoV, MERS-CoVs-specific neutralizing antibodies onto a known stable Nb framework. They identified hotspots between the Nbs and the RBD via a per-residue binding free energy decomposition analysis, while molecular dynamics simulation and binding free energy analysis singled out Nbs that stably bind to the virus's RBD. The team suggests 11 redesigned Nbs that could potentially neutralize the pathogen.

Full Article
How Close Is Ordinary Light to Doing Quantum Computing?
IEEE Spectrum
Neil Savage
April 27, 2021


Using mirrors to generate a light beam with multiple, classical entanglements is possible, according to researchers at China's Tsinghua University, the U.K.'s University of Southampton, and South Africa's University of Witswaterand (WITS). WITS' Andrew Forbes said this technique can entangle a potentially infinite number of photonic pathways, and his team demonstrated eight degrees of freedom within a single beam by changing the spacing between mirrors in the laser cavity. Said Forbes, "Not only could we make light that took many different paths at once, but we could encode information into those paths to make it look like we were holding a high-dimensional multi-photon quantum state." Forbes added that since quantum computing relies on particles existing in multiple states, some algorithms could be run using classically entangled light, bridging quantum and classical computers.

Full Article
Cancer Algorithm Flags Genetic Weaknesses in Tumors
University of Cambridge (U.K.)
April 27, 2021


The MMRDetect clinical algorithm developed by researchers at the U.K.'s University of Cambridge can flag tumors with mismatch repair (MMR) deficiencies, and then enhance personalized therapies to exploit those genetic weaknesses. The team used CRISPR-Cas9 gene-editing technology to render repair genes inoperative in healthy human stem cells, and noticed the failure of strong mutational signatures. The implication is that these signatures of repair pathway defects are continuous, and could function as critical biomarkers in personalized medicine. MMRDetect employs these signatures, and was trained on whole genome sequencing data from U.K. National Health Service cancer patients in the 100,000 Genomes Project, to detect tumors with MMR deficiency. The algorithm could maximize effectiveness if it is applied as soon as a patient has received a cancer diagnosis and their tumor has been characterized by genome sequencing.

Full Article
AI Tool Calculates Materials' Stress and Strain Based on Photos
MIT News
Daniel Ackerman
April 22, 2021


A technique for rapidly assessing material properties like stress and strain, based on an image of the internal structure, has been developed by Massachusetts Institute of Technology (MIT) researchers. The team trained a Generative Adversarial Neural Network with thousands of paired images—respectively depicting a material's internal microstructure subject to mechanical forces, and its color-coded stress and strain values; the network iteratively determined relationships between a material's geometry and its ensuing stresses using principles of game theory. MIT's Markus Buehler said the computer can essentially predict the various forces that act on the material, as opposed to the conventional way, in which "you would need to code the equations and ask the computer to solve partial differential equations. We just go picture to picture." Buehler said the network is well-suited for describing material properties, as it can process data through a series of convolutions, which analyze the images at progressively larger scales.

Full Article

Apple CEO Tim Cook Apple Will Spend $1 Billion to Open 3,000-Employee Campus in North Carolina
CNBC
Kif Leswing
April 26, 2021


Apple plans to open a new campus in the Raleigh, NC, area, with 3,000 employees working on software engineering, machine learning, and other technologies. The company will spend more than $1 billion on the campus, adding to previous plans to open a $1 billion campus in Austin, TX, next year. The NC campus will be located in the Research Triangle area, referring to its proximity to North Carolina State University, Duke University, and the University of North Carolina at Chapel Hill. The move comes as other technology companies, like Oracle, Google, and Amazon, expand outside the San Francisco Bay area to tap a larger pool of talent in areas with lower costs of living. Meanwhile, Apple announced that it would add 20,000 jobs nationwide over the next five years in cities including San Diego, Culver City, Boulder, and Seattle.

Full Article

3D image of examined lattice structure as measured with the aid of computed tomography Energy-Saving Gas Turbines From the 3D Printer
Technical University of Munich
April 23, 2021


A team including researchers at the German Federal Institute for Materials Research and Testing (BAM) and the Technical University of Munich (TUM) developed a method for improving the three-dimensional printing process for gas turbine buckets to minimize internal stress in the components that can result in cracks. Laser Powder Bed Fusion is used to create the complex lattice structures inside the hollow turbine buckets that provide stability. The researchers used neutrons from TUM's Research Neutron Source Heinz Maier-Leibnitz (FRM II) to detect internal stresses in the components. BAM's Tobias Fritsch said, "We know that we have to modify the production process parameters and thus the way in which the component is built up during printing. The more localized the heat application is during the melting process, the more internal stress results. So we have to distribute the heat as evenly as possible during the printing process."

Full Article

SolarWinds sign on building SolarWinds, Microsoft Hacks Prompt Focus on Zero-Trust Security
The Wall Street Journal
James Rundle
April 26, 2021


At an April 22 virtual event hosted by Cyber Education Institute LLC's Billington Cybersecurity unit, U.S. Department of Defense's John Sherman said the public and private sectors should adopt zero-trust models that constantly verify whether a device, user, or program should be able to do what it is asking to do. Ericom Software Ltd.'s Chase Cunningham said, "No one who actually understands zero trust says abandon the perimeter. But the reality of it is that you need to understand your perimeter's probably already compromised, especially when you're in a remote space." Carnegie Mellon University's Gregory Touhill stressed that zero trust is not a technology but a strategy, and "we've got too many folks in industry that are trying to peddle themselves as zero-trust vendors selling the same stuff that wasn't good enough the first time."

Full Article
*May Require Paid Registration
Researchers Rev Up Innovative ML Strategies to Reclaim Energy, Time, and Money Lost in Traffic Jams
National Renewable Energy Laboratory
April 22, 2021


A researchers team led by the National Renewable Energy Laboratory (NREL) found that next-generation sensors and controls in combination with high-performance computing, analytics, and machine learning could minimize road congestion. The researchers used real-time data gathered by a wide range of sensors to develop a series of simulations of traffic conditions in Chattanooga, TN, and identify which controls can achieve the greatest energy efficiency while optimizing travel time, highway speed, and safety. The researchers also analyzed the underlying causes of congestion using machine learning, data from GPS devices and vehicle sensors, and visual analytics. The data could help urban planners, technology developers, automakers, and fleet operators design systems and equipment to make commutes and deliveries more efficient. NREL's Juliette Ugirumurera said, "The intersection of high-performance computing, high-fidelity data, machine learning, and transportation research can deliver powerful results, far beyond what has been possible in the past with legacy technology."

Full Article
A Growing Problem of 'Deepfake Geography': How AI Falsifies Satellite Images
University of Washington
Kim Eckart
April 21, 2021


Researchers at the University of Washington (UW), Oregon State University, and Binghamton University used satellite photos of three cities and manipulation of video and audio files to identify new methods of detecting deepfake satellite images. The team used an artificial intelligence framework that can infer the characteristics of satellite images from an urban area, then produce deepfakes by feeding the characteristics of the learned satellite image properties onto a different base map. The researchers combined maps and satellite imagery from Tacoma, WA, Seattle, and Beijing to compare features and generate deepfakes of Tacoma, based on the characteristics of the other cities. UW's Bo Zhao said, "This study aims to encourage more holistic understanding of geographic data and information, so that we can demystify the question of absolute reliability of satellite images or other geospatial data. We also want to develop more future-oriented thinking in order to take countermeasures such as fact-checking when necessary."

Full Article
Researchers Develop First-of-Its-Kind Extended Reality Testbed to Speed Virtual, Augmented Reality Innovation
University of Illinois Urbana-Champaign
Laura Schmitt
April 26, 2021


University of Illinois at Urbana-Champaign (UIUC) researchers have launched an open source extended reality (XR) testbed to democratize XR systems research, development, and benchmarking. XR is an umbrella term for virtual, augmented, and mixed reality, and UIUC's Sarita Adve cited an orders of magnitude gap between the performance, power, and usability of current and desirable XR systems. The first-of-its-kind Illinois Extended Reality (ILLIXR) testbed is an end-to-end XR system that all types of XR scientists can use to research, develop, and benchmark concepts in the context of a complete XR system, and observe the effect on end-user experience. Facebook's Rod Hooker said, "ILLIXR's open source modular architecture enables the XR research community to address challenging problems in the areas of optimizing algorithms, system performance/power optimizations, scheduler development, and quality-of-service degradation."

Full Article
Algorithm Reveals Birdsong Features That May be Key for Courtship
McGill University (Canada)
April 22, 2021


Scientists at Canada's McGill University and the University of California, San Francisco have developed a new algorithm that can identify characteristics of male zebra finch songs that may underpin differences between a phrase sung during and outside of courtship. The researchers used a bottom-up approach to extract more than 5,000 song features from recordings, and trained the algorithm to apply those features to distinguish courtship from non-courtship song phrases. The algorithm flagged features that had not been identified previously, and predicted the distinction capabilities of female finches that were in line with the results of behavioral experiments. This highlights the potential for bottom-up approaches to uncover acoustic features important for communication and social discrimination. The researchers hope to test whether manipulating acoustic features changes what female finches think about those songs, and assess how their findings might generally apply to courtship and non-courtship songs in other species.

Full Article
Semantic Web For The Working Ontologist, Third Edition: Effective Modeling In RDFs And Owl
 
ACM Insurance for Members
 

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]